Plug and abandonment of subsea wells

Abstract

Methods related to the field of subsea well intervention, e.g. plug and abandonment, and in particular to methods for plugging of subsea wells and substantiating the quality of the well plug. Method for a barrier setting in a wellbore, the method comprising planning the barrier setting, simulating the planned barrier setting, creating an operation plan for the barrier setting; performing a risk assessment for the operation plan; executing the operation plan setting the barrier in the wellbore; and verifying the barrier in the wellbore.

Description

Plug and abandonment of subsea wells

[0001] The present invention relates to the field of subsea well intervention such as well abandonment. In particular, it relates to plugging of subsea wells and substantiating the quality of the well plug. The present invention also relates to validation of a well annulus washing operation. There is disclosed a method and a tool for validating completeness of an annulus washing operation.Background

[0002] The oil and gas industry has produced hydrocarbons from subsea wells for many decades. When the wells are no longer commercially viable, they must be abandoned in a safe manner, such that hydrocarbons are prevented from leaking into the environment. Such abandonment involves permanently plugging the well with a cement plug. The plug must be tight and lasting.

[0003] The regulations governing how a subsea well shall be plugged vary between different states. For instance, the NORSOK standard that governs permanent plug and abandonment (PP&A) on the Norwegian continental shelf has relatively stringent rules.

[0004] While the rules for PP&A may be sufficiently clear, it is not always easy to confirm that a plugging operation has in fact been conducted according to rules. It goes without saying that an operation taking place several hundreds or even thousands of meters below the seabed may not be easy to overlook. Verifying and substantiating the success of a plugging operation is therefore a challenge.

[0005] As an example, during a PWC operation (perforate wash and cement) one must ensure that the annulus is without debris, cuttings, mud, or other components that may hamper the formation of the cement plug. The annulus is therefore washed before introducing the cement. However, even if the washing step has been performed, the operator will wish to know if the washing has been successful. Moreover, the operator will wish to substantiate the success of the washing step.

[0006] Another example involves removal of a control cable or gauge cable being present in the annulus, as the presence of a control cable may lead to leakage through the plug. It is known to cut the control cable with a gun that blasts through the casing and cuts off the control cable outside of it. Again, while the cutting operation has been performed according to plan, the operator will wish to confirm that the control cable has been cut and removed from the annulus that shall be plugged.

[0007] In some situations, during the lifetime of a hydrocarbon well, it may be necessary to wash the well annulus. A typical scenario is when a plug shall be installed in the well by setting a cement plug in the annulus and main bore.

[0008] Before cement is introduced in the annulus, the operator washes the annulus such that the quality of the installed plug is not jeopardized. If the cement is introduced in an unwashed annulus, impurities (e.g. barite from settled drilling mud) may result in a poor plug quality.

[0009] Moreover, although having performed a washing operation, the operator may wish to know if the washing operation has been successful or not. An object of the present invention may be to provide a method and a tool for verifying a successful annulus washing operation.Summary of the InventionAn aspect is directed towards a method for a barrier setting in a wellbore, the method comprising: planning the barrier setting; simulating the planned barrier setting; creating an operation plan for the barrier setting; performing a risk assessment for the operation plan; executing the operation plan setting the barrier in the wellbore; and verifying the barrier in the wellbore.The method may further comprise calculating a quality of the barrier in the wellbore. The method may further comprise calculating the barrier quality over the barrier length. The barrier length may correspond to an ablated length of the wellbore. Executing the operation plan may further comprise ablating a control line in an annulus of the wellbore and confirming ablation of the control line. Abarrier quality may be calculated for a length of the barrier. A barrier quality may be displayed for a length of the barrier. Diagnostics of the well may be performed before simulation.An aspect is also directed towards a method for a barrier setting in a wellbore, the method comprising: executing an operation plan setting the barrier in the wellbore; and verifying the barrier in the wellbore.The method may further comprise calculating a quality of the barrier in the wellbore. The method may further comprise calculating the barrier quality over the barrier length The barrier length may correspond to an ablated length of the wellbore. Executing the operation plan may further comprise ablating a control line in an annulus of the wellbore and confirming ablation of the control line. A barrier quality may be calculated for a length of the barrier. A barrier quality may be displayed for a length of the barrier. Diagnostics of the well may be performed before simulation.Another aspect is directed towards a method in a barrier setting in a wellbore, the method comprising: detecting a control line location and orientation in the wellbore; ablating the control line over at least a length of the wellbore; and verifying ablation of the control line over the ablated length of the wellbore.The method may further comprise verifying ablation of the control line by performing downhole diagnostics. Verifying ablation of the control line may be performed by ultrasonic logging. Verifying ablation of the control line may be performed using an imaging tool.Another aspect is directed towards a method in a barrier setting in a wellbore, the method comprising: washing an annulus; and detecting by use of acoustics, a particle flow in the annulus during washing.The method may further comprise calculating a total mobilized mass of particles in the annulus based on the detected particle flow.Another aspect is directed towards a method for establishing a quality of a barrier setting in a wellbore, the method comprising calculating at least one quality parameter for the barrier based on at least one parameter obtained during the barrier setting and at least one verification parameter obtained in a verification of the barrier.The at least on parameter obtained during the barrier setting may comprise cementing flow rate and pressure. The at least one parameter obtained during the barrier setting may be related to ablation of a control line along the barrier. The at least one parameter obtained during the barrier setting may be washing pressure.The at least one parameter obtained during the barrier setting may be related to monitoring of washing particles during the washing of the wellbore.The at least one quality parameter of the barrier may be calculated for a number of intervals over the barrier length. The at least one quality parameter may be displayed in intervals along the depth of the wellbore.Another aspect is directed towards a product for qualification of a barrier setting in a well comprising:- a number of tracks for parameters obtained before, during and after the barrier setting in intervals along a depth of the wellbore; and- a calculated quality parameter track representing a quality of the barrier in intervals along the depth of the wellbore.The parameters obtained before the barrier setting may comprise parameters related to at least one of gamma ray log, casing collar locator log, cement bond log, variable density log. The parameters obtained before the barrier setting may comprise parameters related to control line and clamp location and orientation. The parameters obtained during the barrier setting may comprise parameters related to cementing flow rate and pressure, preferably cementing flow rate and pressure versus set parameters. The parameters obtained during the barrier setting comprising parameters related to detected particles during a washing operation, preferably estimated particle mass lifted from an annulus versusparticles in place in the annulus before washing. The parameters obtained during the barrier setting may comprise parameters related to ablation of control line over the length of the barrier. The parameters obtained after the barrier setting comprising parameters related to acoustic monitoring of the barrier along the barrier length to verify the barrier. The quality parameter track is calculated based on values calculated for the execution of the barrier setting and verification values for the set barrier along the barrier length. The number of tracks for the parameters obtained before, during and after the barrier setting and the calculated quality parameter track represent a visualization of the parameters.Another aspect is directed towards a method for a barrier setting in a wellbore, the method comprising: logging the wellbore; ablating a control line in the wellbore; confirming ablation of the control line; washing the wellbore; placing a fluidized plugging material in the wellbore creating a barrier in the wellbore; and verifying the barrier.The present invention provides an improved method for controlling the end result in a barrier setting in a well. A barrier setting may be a high-risk operation. Verification of the end result is chosen based on risk profile and cost.The invention provides an answer product that is able to calculate and verify the actual delivered barrier in the well.The invention provides a unique workflow including ablation and confirmation of ablation.The invention provides a new approach to define a successful barrier and calculate actual barrier length delivered based on actual ablated length.The invention also provides detection of particle flow and the ability to calculate the total mobilized mass during the washing process.

[0001] There is disclosed a method of validating a well annulus washing operation. The method comprises: a) with an annulus washing tool arranged at aperforated section of a well pipe, providing a washing fluid flow into the annulus of the well pipe. The method further comprises: b) with an acoustic sensor arranged inside the inner bore of the wellpipe, detecting acoustic signals generated by the washing fluid flow.

[0002] The acoustic sensor can be part of the annulus washing tool.

[0003] Furthermore, the acoustic sensor can be, during step a), located at the location of the perforations.

[0004] In some embodiments of the method, the acoustic sensor can comprise an upper acoustic sensor which is part of the annulus washing tool above the wash fluid ports (e.g. nozzles) of the annulus washing tool. Alternatively, the acoustic sensor can comprise an upper acoustic sensor being supported on a string that supports the annulus washing tool, at a position above the annulus washing tool. The upper acoustic sensor can, during step a), be arranged inside the inner bore, above the location of the perforations.

[0005] In some embodiments, the method can comprise the following step: c) based on detected acoustic signals, calculating an amount of material that has been washed out of the annulus with the washing fluid flow.

[0006] In such embodiments, the method may further comprise the following steps: d) calculating an amount of material present in the portion of the annulus before the washing operation in step a); and e) calculating an annulus wash index by dividing the amount of material, as calculated in step c), by the amount of material as calculated in step d).

[0007] There is also disclosed an annulus washing assembly, comprising an annulus washing tool with wash fluid ports (e.g. nozzles) configured to provide a washing fluid flow. The annulus washing assembly further comprises an acoustic sensor.

[0008] In some embodiments, the annulus washing tool comprises the said acoustic sensor.

[0009] The annulus washing assembly may comprise an upper acoustic sensor arranged above the wash fluid ports. The upper acoustic sensor can be inaddition to or instead of the abovementioned acoustic sensor that is part of the annulus washing tool.

[0010] Advantageously, the upper acoustic sensor can be arranged several meters, for instance at least twenty meters, forty meters or even eighty meters above the wash fluid ports.

[0011] In some embodiments of the annulus washing assembly, the annulus washing tool further comprises an upper casing bore sealing arrangement and a lower casing bore sealing arrangement. Moreover, the wash fluid ports (e.g. nozzles) can be arranged between the upper and lower casing bore sealing arrangements, and the acoustic sensor can be arranged above the wash fluid ports. The casing bore sealing arrangements may be in the form of wash cups. Alternatively, the casing bore sealing arrangements can be in the form of flexible expandable elements, such as expandable rubber elements.

[0012] Advantageously, the annulus washing assembly can comprise or can be connected to a computer-readable memory unit or a computing unit programmed to calculate an annulus wash index.

[0013] In some embodiments, the detected acoustic signals can be used to calculate the removed or washed-out mass in real time. In such embodiments one may have a data link to surface, for instance with a through-wired drill pipe or through-wired coiled tubing. This enables the operator to monitor the progress of the washing process during operation.

[0014] Alternatively, the washing assembly may comprise a data storage unit for storing the detected acoustic signals. The operator may then perform calculations when the assembly has been retrieved to surface.

[0010] An aspect is directed towards a method of validating a subsea well annulus washing operation, comprising the following steps: a) with an annulus washing tool arranged at a perforated section of a well pipe, providing a washing fluid flow from wash fluid ports (e.g. nozzles) through perforations in the well pipe, into the annulus, and back through perforations into the inner bore of the well pipe;b) with an acoustic sensor being part of the annulus washing tool, detecting acoustic signals generated by the washing fluid flow; and c) detecting a change of acoustic signals detected in step b).

[0011] When the washing fluid flow is circulated through the annulus, material will be washed away from the annulus. The washing fluid flow will generate acoustic signals that is detected with the acoustic sensor. Moreover, when the washing fluid flow contains material to be washed out, typically particles and debris, it will generate acoustic signals that differ from the sound produced when the washing fluid flow does not contain such material. The operator can thus confirm that the annulus has been appropriately washed. This will typically be performed before a cement plug is installed in the annulus.

[0012] In some embodiments, step b) can further include detecting, with an auxiliary acoustic sensor that is part of the annulus washing tool and arranged above the said acoustic sensor, acoustic signals of the washing fluid flow flowing through the inner bore of the well pipe.

[0013] The method may also further comprise the following step: d) based on at least the acoustic signals detected with the auxiliary sensor, calculating an amount of material that has been washed out of the annulus with the washing fluid flow.

[0014] In this manner, the operator can compare the volume of removed or washed material with the volume of the annulus. This gives a further indication of a complete or successful washing operation.

[0015] The method can further comprise storing detected acoustic signals in a computer-readable memory unit and, with a computing system, performing step c) with acoustic signals stored in the computer-readable memory unit.

[0016] Another aspect relates to an annulus washing tool for cleaning of a subsea well annulus, the annulus washing tool comprising wash fluid ports (e.g. nozzles) configured to provide a washing fluid flow. The annulus washing tool further comprises an acoustic sensor.

[0017] With the acoustic sensor, the acoustic signals generated from the washing fluid flow can be detected.

[0018] The annulus washing tool may further comprise an auxiliary acoustic sensor arranged a distance above the said acoustic sensor.

[0019] The annulus washing tool can comprise or be connected to a computer- readable memory unit and a computing unit programmed to perform any of the methods disclosed herein.

[0020] A further aspect relates to a method of setting a permanent plug in a subsea well having a well pipe and a control line located in an annulus outside the well pipe. The method comprises a) with a perforation tool, providing perforations in the well pipe for removal of the control line along a portion of the well pipe by severing the control line when providing said perforations; b) washing the annulus and installing cement in the annulus and the bore of the well pipe;

[0021] The method further comprises:

[0022] with an ultrasonic tool arranged in the bore of the well pipe, recording ultrasonic images of the perforations;

[0023] inspecting the ultrasonic images and verifying removal of the control line.

[0024] A further aspect relates to a method for controlling a Plug and Abandonment (PA) procedure for a well, wherein the PA procedure comprises washing and setting a permanent plug in the well to form a barrier. The method comprises obtaining a set of diagnostic parameters from a network of sensors or an application database. The set of diagnostic parameters comprises initial well parameters, cement bond logs or variable density logs over a target depth range of the well before the PA procedure. Using the set of diagnostic parameters to plan the PA procedure, one or more operational steps is determined. A simulation of the one or more operational steps is performed, wherein performing the simulation comprises performing hydraulics calculations and estimating downhole forces based on the set of diagnostic parameters. The simulation is analyzed to generatean operation plan comprising the one or more operational steps. Once the PA procedure is initiated according to the operation plan, the PA procedure is monitored using data collected from the network of sensors to determine an operation status of the PA procedure, wherein the network of sensors comprise an ultrasonic imaging tool and an acoustic sensor. The PA procedure is validated based on determining a barrier quality value, wherein the barrier quality is associated with whether there is a leak in the barrier. Data associated with execution of the PA procedure is then stored in an application database.

[0025] A further aspect relates to a method for simulating operation of a Plug and Abandonment (PA) procedure for a well, wherein the PA procedure comprises washing and installing cement in the well. The method comprising: obtaining, from an application database, a set of diagnostic parameters, wherein the set of diagnostic parameters comprises hydraulics parameters and conveyance parameters associated with a planned PA procedure; generating a hydraulics module configured to perform hydraulics calculations comprising ball drop calculations, pressure drop calculations, volume calculations, or calculations associated with cementing the well; generating a conveyance module configured to estimate downhole forces based on torque and drag calculations; and using the hydraulics module and the conveyance module to simulate one or more operational steps of the planned PA procedure based on the set of diagnostic parameters.

[0026] Optionally, the method further comprises generating a computational fluid dynamics (CFD) module configured to perform further hydraulics calculations comprising any of: cement slumping calculations, washing efficiency calculations, listing of particles calculations, multiple annuli calculations, or calculations associated with tubular, tubing, or casing designs; and using the CFD module to simulate one or more operational steps of the planned PA procedure. Further optionally, the method comprises generating a perforation ablation module configured to verify correct perforation and ablation or determine mechanical perforation and ablation parameters based on geometry of perforations, perforation depth, or perforation gun setup; and using the perforation ablation module to simulate one or more operational steps of the planned PA procedure.

[0027] A further aspect relates to a method for monitoring operation of a Plug and Abandonment (PA) procedure for a well, the method comprising obtaining, from a network of sensors comprising at least one acoustic sensor, a first set of parameters associated with the well. The method further comprises generating one or more processed parameters by performing one or more processing steps on each parameter of the first set of parameters, the processing steps comprising: merging two or more parameters of the first set of parameters, resulting in data fusion; filtering one or more parameters of the first set of parameters; or extracting a target feature from one or more parameters of the first set of parameters. An estimated state of the PA procedure is predicted based on the one or more processed parameters. The method further comprises obtaining, from the network of sensors, a second set of parameters associated with monitoring the well after the PA procedure is initiated, wherein the second set of parameters comprises one or more acoustic signals. A current state of the PA procedure is generated based on the second set of parameters and an operation status determined based on comparing the estimated state of the PA procedure to the current state of the PA procedure.

[0028] A further aspect relates to a system for diagnosing a Plug & Abandonment (PA) operation comprising an artificial neural network (ANN), wherein the system is configured to perform the following steps: obtain sensor data from a network of sensors, the network of sensors comprising one or more surface sensors and one or more downhole sensors; obtain historical data from an application database, the historical data comprising historical fault data or trend data associated with previous PA operations; extract one or more relevant features from the data from the sensor data, wherein relevant features are determined based on the sensor data; use the ANN, trained using the historical data, to identify one or more conditions indicators based on the one or more relevant features, wherein the one or more conditions indicators are associated with performing the PA operation;determine a predicted state of the PA operation based on the one or more conditions indicators; and predict whether a failure mode is probable based on comparing the predicted state of the PA operation to a current state of the PA operation, the current state of the PA operation based on either: a protocol for the PA operation, the protocol comprising a plurality of planned operational steps or instructions to perform the PA operation; or updated sensor data from the network of sensors.

[0029] A further aspect relates to a method for monitoring a Plug and Abandonment (PA) procedure for a well, wherein the PA procedure comprises washing an annulus of the well and setting a permanent plug in the well. The method comprises determining that a control line is present in the well based on a cement bond log or an ultrasonic log; obtaining one or more ultrasonic images from an ultrasonic imaging tool within the well after ablation of the control line is initiated; and verifying ablation of the control line over an ablated length based on the ultrasonic images associated with perforations of the control line. After washing of the annulus is initiated, one or more acoustic signals is obtained from a first acoustic sensor within the well at a first time point and a second time point. It is then confirmed whether of the well is complete based a change in the one or more acoustic signals detected at the first time point and the one or more acoustic signals detected at the second time point. The method further comprises obtaining a set of verification parameters from a second acoustic sensor after installation of cement in the annulus is initiated, wherein the set of verification parameters are indicative of the presence of leaks in the cement plug.

[0030] Alternatively, a method for monitoring a Plug and Abandonment (PA) procedure for a well, wherein the PA procedure comprises washing an annulus of the well and setting a permanent plug in the well, comprises: determining that a control line is present in the well based on data obtained from an application database; obtaining a first set of diagnostic parameters collected by one or more sensors before ablation of the control line, wherein the one or more sensors comprises at least one ultrasonic imaging tool, and wherein the first set of diagnostic parameters comprises initial well parameters, cement bond logs orvariable density logs over a target depth range of the well, or ultrasonic sensor data; determining a position of the control line based on the first set of diagnostic parameters, wherein the position comprises a location or an orientation of the control line relative to the annulus; and, after perforation of the control line is initiated, obtaining a second set of diagnostic parameters from the one or more sensors comprising one or more ultrasonic signals indicative of whether the control line is perforated. Ablation of the control line over an ablated length associated with the target depth range is verified based on the second set of diagnostic parameters and, if ablation of the control line is complete and verified, installation of cement within the annulus is initiated for setting the permanent plug.

[0031] A further aspect relates to a method for validating a Plug and Abandonment (PA) procedure for a well, wherein the PA procedure comprises forming a barrier over a target depth range. The method comprises obtaining a set of diagnostic parameters comprising gamma ray signals, cement bond logs or variable density logs over the target depth range of the well collected before the PA procedure, wherein the set of diagnostic parameters are associated with planning the PA procedure. A set of procedure parameters collected during the PA procedure is obtained from a network of sensors comprising an acoustic sensor, wherein the set of procedure parameters are associated with monitoring execution of the PA procedure, and wherein the set of procedure parameters comprises: a washing pressure flowrate track over the target depth range, and a cementing track over the target depth range associated with installation of cement within the well. The method further comprises obtaining, from the network of sensors after the PA procedure is executed, a set of verification parameters comprising an acoustic track over the target depth range, and analyzing the set of verification parameters to determine whether there is a leak in the barrier. Based on whether a leak is detected over the target depth range, the target depth range is assigned a barrier quality value indicative of whether the PA procedure was executed successfully. The barrier quality value is then output for validation of the PA procedure.

[0032] In an aspect relates to a method of validating a well annulus (101 ) washing operation, comprising the following steps:a) with an annulus washing tool (100) arranged at a perforated section of a well pipe (103), providing a washing fluid flow (110) into the annulus(101 ) of the well pipe; b) with an acoustic sensor (115, 117) arranged inside an inner bore(102) of the well pipe, detecting acoustic signals generated by the washing fluid flow (110).

[0033] The acoustic sensor (115) may be part of the annulus washing tool (100) and is, during step a), located at the location of the perforations (103a). The acoustic sensor may comprise an upper acoustic sensor (117) being part of the annulus washing tool (100) above wash fluid ports (e.g. nozzles) (111 ) of the annulus washing tool; or an upper acoustic sensor (117) being supported on a string that supports the annulus washing tool (100), at a position above the annulus washing tool;

[0034] wherein the upper acoustic sensor (117) during step a) is arranged inside the inner bore (102), above the location of the perforations (103a). The method may further comprise c) based on detected acoustic signals, calculating an amount of material (113) that has been washed out of the annulus (101 ) with the washing fluid flow (110).The method may further comprise d) calculating an amount of material present in the portion of the annulus (101 ) before the washing operation in step a); e) calculating an annulus wash index by dividing the amount of material (113), as calculated in step c), by the amount of material as calculated in step d).

[0035] A further aspect relates to an annulus washing assembly, comprising an annulus washing tool (100) comprising wash fluid ports (111 ) configured to provide a washing fluid flow (110), wherein the annulus washing assembly further comprises an acoustic sensor (115, 117).

[0036] The annulus washing tool (100) may comprise the acoustic sensor (115). The annulus washing assembly may further comprise an upper acoustic sensor (117) arranged above the wash fluid ports (111 ). The annulus washing tool (100) may further comprise an upper wash cup (107) and a lower wash cup (109), wherein the wash fluid ports (111 ) arearranged between the upper and lower wash cups (107, 109), and wherein the acoustic sensor (115, 117) is arranged above the wash fluid ports (111). The annulus washing assembly may comprise or is connected to a computer-readable memory unit or a computing unit programmed to calculate an annulus wash index.

[0037] A further aspect relates to a method for controlling a Plug and Abandonment (PA) procedure for a well, wherein the PA procedure comprises washing and setting a permanent plug in the well to form a barrier, the method comprising: obtaining a set of diagnostic parameters from a network of sensors or an application database; using the set of diagnostic parameters to plan the PA procedure, determining one or more operational steps; performing a simulation of the one or more operational steps; analysing the simulation to generate an operation plan comprising the one or more operational steps; monitoring the PA procedure using data collected from the network of sensors to determine an operation status of the PA procedure; validating the PA procedure based on determining a barrier quality value, wherein the barrier quality is associated with whether there is a leak in the barrier; and storing, in the application database, data associated with execution of the PA procedure.

[0038] The set of diagnostic parameters may comprise initial well parameters, cement bond logs or variable density logs over a target depth range of the well before the PA procedure. Performing the simulation may comprise performing hydraulics calculations and estimating downhole forces based on the set of diagnostic parameters. The method may further comprise monitoring the PA procedure once the PA procedure is initiated according to the operation plan. The network of sensors may comprise an ultrasonic imaging tool and an acoustic sensor.

[0039] A further aspect relates to a system comprising: a processor configured to communicate with an application database; a network of sensors, wherein the network of sensors preferably comprising an ultrasonic imaging tool and an acoustic sensor; and a memory storinginstructions that, when executed, cause the system to perform the method according to above.

[0040] A further aspect relates to a method for simulating operation of a Plug and Abandonment (PA) procedure for a well, wherein the PA procedure comprises washing and installing cement in the well, the method comprising: obtaining, from an application database, a set of diagnostic parameters; generating a hydraulics module; generating a conveyance module; and using the hydraulics module and the conveyance module to simulate one or more operational steps of the planned PA procedure based on the set of diagnostic parameters.

[0041] The set of diagnostic parameters may comprise hydraulics parameters and conveyance parameters associated with a planned PA procedure. The hydraulics module may be configured to perform hydraulics calculations comprising ball drop calculations, pressure drop calculations, volume calculations, or calculations associated with cementing the well. The conveyance module is configured to estimate downhole forces based on torque and drag calculations.

[0042] A further aspect relates to a method for monitoring operation of a Plug and Abandonment (PA) procedure for a well, the method comprising: obtaining, from a network of sensors a first set of parameters associated with the well; generating one or more processed parameters by performing one or more processing steps on each parameter of the first set of parameters, the processing steps comprising: merging two or more parameters of the first set of parameters, resulting in data fusion; filtering one or more parameters of the first set of parameters; or extracting a target feature from one or more parameters of the first set of parameters; predicting an estimated state of the PA procedure based on the one or more processed parameters; obtaining, from the network of sensors, a second set of parameters associated with monitoring the well after the PA procedure is initiated; generating a current state of the PA procedure based on the second set of parameters; and determining an operationstatus based on comparing the estimated state of the PA procedure to the current state of the PA procedure.

[0043] The method may include use of a least one acoustic sensor. The second set of parameters comprises one or more acoustic signals.

[0044] A further aspect relates to a system for diagnosing a Plug & Abandonment (PA) operation comprising an artificial neural network (ANN), wherein the system is configured to perform the following steps: obtain sensor data from a network of sensors, the network of sensors comprising one or more surface sensors and one or more downhole sensors; obtain historical data from an application database, the historical data comprising historical fault data or trend data associated with previous PA operations; extract one or more relevant features from the data from the sensor data, wherein relevant features are determined based on the sensor data; use the ANN, trained using the historical data, to identify one or more conditions indicators based on the one or more relevant features, wherein the one or more conditions indicators are associated with performing the PA operation; determine a predicted state of the PA operation based on the one or more conditions indicators; and predict whether a failure mode is probable based on comparing the predicted state of the PA operation to a current state of the PA operation, the current state of the PA operation based on either: a protocol for the PA operation, the protocol comprising a plurality of planned operational steps or instructions to perform the PA operation; or updated sensor data from the network of sensors.

[0045] The network of sensors may comprise one or more surface sensors and one or more downhole sensors. The historical data may comprise historical fault data or trend data associated with previous PA operations. Relevant features may be determined based on the sensor data. The one or more conditions indicators are associated with performing the PA operation.

[0046] A further aspect relates to a method for monitoring a Plug and Abandonment (PA) procedure for a well, wherein the PA procedure comprises washing an annulus of the well and setting a permanent plug in the well, the method comprising: determining that a control line is present in the well; verifying ablation of the control line over an ablated length; after washing of the annulus is initiated, obtaining one or more acoustic signals from a first acoustic sensor within the well; confirming whether washing of the well is complete based on a change in the one or more detected acoustic signals; and obtaining a set of verification parameters after installation of cement in the annulus is initiated, wherein the set of verification parameters are indicative of the presence of leaks in the cement plug.

[0047] The method may further comprise determining that a control line is present in the well based on a cement bond log or an ultrasonic log. After ablation of the control line is initiated, one or more ultrasonic images may be obtained from an ultrasonic imaging tool within the well. Verifying ablation of the control line over an ablated length may be based on the ultrasonic images associated with perforations of the control line. After washing of the annulus is initiated, one or more acoustic signals may be obtained from a first acoustic sensor within the well at a first time point and a second time point. The method may further comprise confirming whether washing of the well is complete based on a change in the one or more acoustic signals detected at the first time point and the one or more acoustic signals detected at the second time point. The method may further comprise obtaining a set of verification parameters from a second acoustic sensor after installation of cement in the annulus is initiated.

[0048] A further aspect relates to a method for monitoring a Plug and Abandonment (PA) procedure for a well, wherein the PA procedure comprises washing an annulus of the well and setting a permanent plug in the well, the method comprising: determining that a control line is present in the well based on data obtained from an application database; obtaining a first set of diagnostic parameters collected by one or more sensors;determining a position of the control line based on the first set of diagnostic parameters, wherein the position comprises a location and / or an orientation of the control line; verifying ablation of the control line over an ablated length associated with the target depth range based on a second set of diagnostic parameters; and if ablation of the control line is complete and verified, initiating installation of cement within the annulus for setting the permanent plug.

[0049] The method may further comprise obtaining a first set of diagnostic parameters collected by one or more sensors before ablation of the control line, wherein the one or more sensors comprises at least one ultrasonic imaging tool. The first set of diagnostic parameters may comprise initial well parameters, cement bond logs or variable density logs over a target depth range of the well, or ultrasonic sensor data. After perforation of the control line is initiated, obtaining a second set of diagnostic parameters from the one or more sensors comprising one or more ultrasonic signals indicative of whether the control line is perforated.

[0050] A further aspect relates to a method for validating a Plug and Abandonment (PA) procedure for a well, wherein the PA procedure comprises forming a barrier over a target depth range, the method comprising: obtaining a set of diagnostic parameters, wherein the set of diagnostic parameters are associated with planning the PA procedure; obtaining, from a network of sensors, a set of procedure parameters collected during the PA procedure, wherein the set of procedure parameters are associated with monitoring execution of the PA procedure, obtaining, from the network of sensors after the PA procedure is executed, a set of verification parameters; and analysing the set of verification parameters to determine whether there is a leak in the barrier; based on whether a leak is detected over the target depth range, assigning the target depth range a barrier quality value indicative of whether the PA procedure was executed successfully; and outputting the barrier quality value for validation of the PA procedure.

[0051] Obtaining a set of diagnostic parameters comprising gamma ray signals, cement bond logs or variable density logs over the target depth range of the well collected before the PA procedure. The network of sensors may comprise an acoustic sensor. The set of procedure parameters comprises: a washing pressure flowrate track over the target depth range, and a cementing track over the target depth range associated with installation of cement within the well. The set of verification parameters comprising an acoustic track over the target depth range.

[0052] A further aspect relates to a method for validating a Plug and Abandonment (PA) procedure for a well, wherein the PA procedure comprises forming a barrier over a target depth range, the method comprising: obtaining a set of diagnostic parameters; using the set of diagnostic parameters for planning the PA procedure, wherein planning the PA procedure comprises determining one or more operational steps followed during execution of the PA procedure; obtaining, from a network of sensors, a set of procedure parameters collected during the PA procedure; using the set of procedure parameters to monitor execution of the PA procedure.

[0053] The set of verification parameters may comprise an acoustic track over the target depth range. One or more signal peaks in the acoustic track may be associated with one or more leaks.

[0054] A further aspect relates to a computer-readable medium storing instructions for performing any of the methods according to above.

[0055] A further aspect relates to a system comprising: a network of sensors, wherein the network of sensors are configured to detect data associated with monitoring planning or operation of a Plug and Abandonment (PA) procedure for a well; a processor configured to obtain data from the network of sensors and communicate with an applicationdatabase; and a memory storing instructions that, when executed, cause the system to perform any of the method according to above.The aspects above are explained with reference to a plug and abandonment procedure, but the aspects are also applicable for well intervention procedures.A further aspect related to a method for validating a well intervention procedure for a well, wherein the well intervention procedure comprises forming a barrier over a target depth range. The method comprising obtaining a set of diagnostic parameters, wherein the set of diagnostic parameters are associated with planning the well intervention procedure. Obtaining, from a network of sensors, a set of procedure parameters collected during the well intervention procedure, wherein the set of procedure parameters are associated with monitoring execution of the well intervention procedure. Obtaining, from the network of sensors after the well intervention procedure is executed, a set of verification parameters. Analysing the set of verification parameters to determine whether there is a leak in the barrier. Based on whether a leak is detected over the target depth range, assigning the target depth range a barrier quality value indicative of whether the well intervention procedure was executed successfully. Outputting the barrier quality value for validation of the well intervention procedure.Obtaining a set of diagnostic parameters may comprise obtaining initial well parameters. The initial well parameters may be over the target depth range of the well. Obtaining a set of diagnostic parameters may comprise obtaining cement bond logs over the target depth range of the well. Obtaining a set of diagnostic parameters may comprise obtaining variable density logs over the target depth range of the well. Obtaining a set of diagnostic parameters may comprise obtaining hydraulics parameters and conveyance parameters. The set of diagnostic parameters may be collected before the well intervention procedure. The set of procedure parameters may comprise a washing pressure flowrate track over the target depth range. The set of procedure parameters may comprise a cementing track over the target depth range. The set of verification parameters may comprise an acoustic track over the target depth range. The network of sensors may comprise an acoustic sensor. The network of sensorsmay comprise an ultrasonic logging tool. The set of verification parameters may comprise an acoustic track over the target depth range. The acoustic track may be associated with detecting acoustic signals generated by a washing fluid flow. If a control line is present, the set of procedure parameters may comprise a perforation ablation track over the target depth range. The procedure parameters for the perforation ablation track over the target depth range may comprise direct control line detection through holes or windows in the tubing I casing using ultrasonic images or ultrasonic measurements, and I or control line detection through the tubing I casing using an acoustic sensor. The set of diagnostic parameters may comprise any of a multifinger caliper track over the target depth range, or ultrasonic measurement track over the target depth range, or a control line track over the target depth range. The set of procedure parameters may comprise a washing particles track over the target depth range.The method may output one or more of: the set of diagnostic parameters for planning the well intervention procedure; the set of procedure parameters for generating an operation plan or monitoring the well intervention procedure; or the set of verification parameters for verifying execution of the well intervention procedure. The set of verification parameters may comprise a drill out and cement bond line track over the target depth range. The set of verification parameters may comprise measurements from a pressure or temperature sensor below the target depth range or below the barrier. The set of verification parameters may comprise validation of barrier integrity derived from one or more acoustic signals across the barrier; or measurements from a gas permeability test of the barrier.The barrier quality value may comprise a plurality of quality values, each of the quality values associated with a predetermined length within the target depth range. The plurality of quality values may be determined based on the set of verification parameters for each predetermined length. The barrier quality value quality parameter track may be calculated based on values calculated for the execution of the barrier setting and verification values for the set barrier along the barrier length. The barrier quality value may be a number of tracks for theparameters obtained before, during and after the barrier setting. The barrier quality value may represent a visualization of the parameters.The well intervention procedure may be a Plug and Abandonment (PA) procedure.In a further aspect a method for simulating operation of a well intervention procedure for a well is discloses. The well intervention procedure may comprise washing and installing cement in the well. The method comprising obtaining, from an application database, a set of diagnostic parameters; obtaining a hydraulics module; obtaining a conveyance module; and using the hydraulics module and the conveyance module to simulate one or more operational steps of the planned well intervention procedure based on the set of diagnostic parameters. The set of diagnostic parameters may comprise hydraulics parameters associated with a planned well intervention procedure. The set of diagnostic parameters may comprise conveyance parameters associated with a planned well intervention procedure. The hydraulics module may be configured to perform hydraulics calculations comprising ball drop calculations, pressure drop calculations, volume calculations, or calculations associated with cementing the well. The conveyance module may be configured to estimate downhole forces based drag calculations. The conveyance module may be configured to estimate downhole forces based on torque calculations.The method may further comprise obtaining a computational fluid dynamics (CFD) module configured to perform further hydraulics calculations comprising any of: cement slumping calculations, washing efficiency calculations, listing of particles calculations, multiple annuli calculations, or calculations associated with tubular, tubing, or casing designs. The CFD module may be used to simulate one or more operational steps of the planned well intervention procedure.The method may further comprise generating a perforation ablation module configured to verify correct perforation and ablation or determine mechanical perforation and ablation parameters based on e.g. geometry of perforations, perforation depth, or perforation gun setup. The perforation ablation module may be used to simulate one or more operational steps of the planned well intervention procedure. Simulation of the one or more operational steps may beused to generate or verify an operation plan comprising operational steps used for planning the well intervention procedure. Simulation of the one or more operational steps may be used to generate or verify an operational envelope comprising a range of operation parameters for performing the well intervention procedure optimally. The method may comprise comparing the operation plan to a monitored well intervention procedure to determine an operational status of the monitored well intervention procedure.A machine learning module may comprise the hydraulics module and the conveyance module such that the machine learning module simulates operation of the well intervention procedure. The machine learning module may comprise a generative artificial intelligence (Al) algorithm configured to generate the operation plan. A data analytics interface may initiate simulation of the one or more operational steps, and may be configured to display the simulation of the one or more operational steps. Simulation of the one or more operational steps associated with the set of diagnostic parameters may be stored in the application database. The method may comprise identifying faults or risks in the well or well intervention procedure based on simulation of the one or more operational steps. The hydraulics parameters and conveyance parameters may be obtained from one or more sensors associated with the well.The invention and the technique for validation of perforate, wash and cement quality are relevant for any well intervention in general.The well intervention procedure may be a Plug and Abandonment (PA) procedure.In a further aspect a method for a barrier setting in a wellbore is disclosed. The method comprising: planning the barrier setting, simulating the planned barrier setting, creating an operation plan for the barrier setting; performing a risk assessment for the operation plan; executing the operation plan setting the barrier in the wellbore; and verifying the barrier in the wellbore.A method for a barrier setting in a wellbore is disclosed where the method comprising executing an operation plan setting the barrier in the wellbore; and verifying the barrier in the wellbore.The methods above may further comprising calculating a quality of the barrier in the wellbore. The barrier quality may be calculated over the barrier length. The barrier length may correspond to an ablated length of the wellbore, preferably the ablated length of the wellbore versus total ablated length. Executing the operation plan may further comprise ablating / severing a control line in an annulus of the wellbore and confirming ablation / severing of the control line. A barrier quality may be calculated for a length of the barrier. A barrier quality may be displayed for a length of the barrier. Diagnostics of the well may be performed before simulation.It is disclosed a method in a barrier setting in a wellbore, the method comprising: detecting a control line location and orientation in the wellbore; ablating the control line over at least a length of the wellbore; verifying ablation of the control line over the ablated length of the wellbore.Verifying the ablated length of the wellbore may comprises verifying the ablated length versus total ablated length. Determining location and orientation of the control line may be based on ultrasonic imaging or ultrasonic measurements or an acoustic log. Verifying ablation of the control line may be performed by performing downhole diagnostics. Verifying ablation of the control line may be performed by ultrasonic logging. Verifying ablation of the control line may be performed using an imaging tool.It is disclosed a method in a barrier setting in a wellbore, the method comprising: washing an annulus; and detecting by use of acoustics, a particle flow in the annulus during washing. Calculating a total mobilized mass of particles in the annulus based on the detected particle flow.It is disclosed a method for establishing a quality of a barrier setting in a wellbore, the method comprising calculating at least one quality parameter for the barrier based on at least one parameter obtained during the barrier setting and at least one verification parameter obtained in a verification of the barrier.At least on parameter obtained during the barrier setting may comprise cementing flow rate and pressure. The at least one parameter obtained duringthe barrier setting may be related to ablation of a control line along the barrier. The at least one parameter obtained during the barrier setting may be washing pressure. The at least one parameter obtained during the barrier setting may be related to monitoring of washing particles during the washing of the wellbore. The at least one quality parameter of the barrier may be calculated for a number of intervals over the barrier length. The at least one quality parameter maybe displayed in intervals along the depth of the wellbore.It is disclosed a product for qualification of a barrier setting in a well comprising: a number of tracks for parameters obtained before, during and after the barrier setting in intervals along a depth of the wellbore, and a calculated quality parameter track representing a quality of the barrier in intervals along the depth of the wellbore. The parameters obtained before the barrier setting may comprise parameters related to at least one of gamma ray log, casing collar locator log, cement bond log, variable density log. The parameters obtained before the barrier setting may comprise parameters related control line and clamp location and orientation. The parameters obtained during the barrier setting may comprise parameters related to cementing flow rate and pressure, preferably cementing flow rate and pressure versus set parameters. The parameters obtained during the barrier setting comprising parameters may be related to detected particles during a washing operation, preferably estimated particle mass lifted from an annulus versus particles in place in the annulus before washing. The parameters obtained during the barrier setting may comprise parameters related to ablation of control line over the length of the barrier. The parameters obtained after the barrier setting may comprise parameters related to acoustic monitoring of the barrier along the barrier length to verify the barrier. The quality parameter track may be calculated based on values calculated for the execution of the barrier setting and verification values for the set barrier along the barrier length. The number of tracks for the parameters obtained before, during and after the barrier setting and the calculated quality parameter track may represent a visualization of the parameters.It is disclosed a method for a barrier setting in a wellbore, the method comprising: logging the wellbore; ablating a control line in the wellbore; confirming ablation ofthe control line; washing the wellbore; placing a fluidized plugging material in the wellbore creating a barrier in the wellbore; and verifying the barrier. A method for a barrier setting in a wellbore, the method comprising logging the wellbore; washing the wellbore; placing a fluidized plugging material in the wellbore creating a barrier in the wellbore; and verifying the barrier. Determining that a control line is present in the wellbore may be based on an ultrasonic log. The ultrasonic log may comprise obtaining ultrasonic images or ultrasonic measurements from an ultrasonic tool within the well after ablation of the control line is initiated; or verifying ablation of the control line over an ablated length based on the ultrasonic images or ultrasonic measurements associated with perforations of the control line; or a phased-array oriented beam. Determining that a control line is present in the wellbore may be based on use of an acoustic sensor.A presence of a control line in a wellbore is determined by use of an ultrasonic log. Ablation of a control line in the wellbore is confirmed by using an ultrasonic log. The ultrasonic log may comprise obtaining ultrasonic images or ultrasonic measurements from an ultrasonic tool within the well after ablation of the control line is initiated; or verifying ablation of the control line over an ablated length based on the ultrasonic images or ultrasonic measurements associated with perforations of the control line; or a phased-array oriented beam.It is disclosed a method for controlling a well intervention procedure for a well, wherein the well intervention procedure comprises washing and setting a permanent plug in the well to form a barrier, the method comprising obtaining a set of diagnostic parameters from a network of sensors or an application database; using the set of diagnostic parameters to plan the well intervention procedure, determining one or more operational steps; performing a simulation of the one or more operational steps, analysing the simulation to generate an operation plan comprising the one or more operational steps; monitoring the well intervention procedure using data collected from the network of sensors to determine an operation status of the well intervention procedure; validating the well intervention procedure based on determining a barrier quality value, wherein the barrier quality is associated with whether there is a leak in the barrier; andstoring, in the application database, data associated with execution of the well intervention procedure. The set of diagnostic parameters may comprise at least one of initial well parameters, cement bond logs or variable density logs over a target depth range of the well before the well intervention procedure. Performing the simulation may comprise performing hydraulics calculations and estimating downhole forces based on the set of diagnostic parameters. The well intervention procedure may be monitored once the well intervention procedure is initiated according to the operation plan. The network of sensors may comprise an ultrasonic imaging tool and / or an acoustic sensor. The well intervention procedure may be a Plug and Abandonment (PA) procedure.It is disclosed a method for monitoring operation of a well intervention procedure for a well, the method comprising: obtaining, from a network of sensors a first set of parameters associated with the well; generating one or more processed parameters by performing one or more processing steps on each parameter of the first set of parameters, the processing steps comprising: merging two or more parameters of the first set of parameters, resulting in data fusion; filtering one or more parameters of the first set of parameters; or extracting a target feature from one or more parameters of the first set of parameters; predicting an estimated state of the well intervention procedure based on the one or more processed parameters; obtaining, from the network of sensors, a second set of parameters associated with monitoring the well after the well intervention procedure is initiated; generating a current state of the well intervention procedure based on the second set of parameters; and determining an operation status based on comparing the estimated state of the well intervention procedure to the current state of the well intervention procedure. The method may further comprise at least one acoustic sensor. The second set of parameters may comprise one or more acoustic signals.It is disclosed a system for diagnosing a well intervention operation comprising an artificial neural network (ANN), wherein the system is configured to perform the following steps: obtain sensor data from a network of sensors, the network of sensors comprising one or more surface sensors and one or more downhole sensors; obtain historical data from an application database, the historical datacomprising historical fault data or trend data associated with previous well intervention operations; extract one or more relevant features from the data from the sensor data, wherein relevant features are determined based on the sensor data; use the ANN, trained using the historical data, to identify one or more conditions indicators based on the one or more relevant features, wherein the one or more conditions indicators are associated with performing the well intervention operation; determine a predicted state of the well intervention operation based on the one or more conditions indicators; and predict whether a failure mode is probable based on comparing the predicted state of the well intervention operation to a current state of the well intervention operation, the current state of the well intervention operation based on either: a protocol for the well intervention operation, the protocol comprising a plurality of planned operational steps or instructions to perform the well intervention operation; or updated sensor data from the network of sensors.The network of sensors may comprise one or more surface sensors and one or more downhole sensors. The historical data may comprise historical fault data or trend data associated with previous well intervention operations. The relevant features may be determined based on the sensor data. The one or more conditions indicators may be associated with performing the well intervention operation.It is disclosed a method for monitoring a well intervention procedure for a well, wherein the well intervention procedure comprises washing an annulus of the well and setting a permanent plug in the well, the method comprising: determining that a control line is present in the well; verifying ablation of the control line over an ablated length; after washing of the annulus is initiated, obtaining one or more acoustic signals from a first acoustic sensor within the well; confirming whether washing of the well is complete based on a change in the one or more detected acoustic signals; and obtaining a set of verification parameters after installation of cement in the annulus is initiated, wherein the set of verification parameters are indicative of the presence of leaks in the cement plug. The method may further comprise determining that a control line is present in the well based on a cement bond log or an ultrasonic log or use of an acoustic sensor. The method may further comprise after ablation of the control line isinitiated, obtaining one or more ultrasonic images or measurements from an ultrasonic tool within the well. Verifying ablation of the control line over an ablated length may be based on the ultrasonic images or measurements associated with perforations of the control line. The method may further comprise after washing of the annulus is initiated, obtaining one or more acoustic signals from a first acoustic sensor within the well at a first time point and a second time point. Confirming whether washing of the well is complete may be based on a change in the one or more acoustic signals detected at the first time point and the one or more acoustic signals detected at the second time point. A set of verification parameters from a second acoustic sensor may be obtained after installation of cement in the annulus is initiated.It is disclosed a method for monitoring a well intervention procedure for a well, wherein the well intervention procedure comprises washing an annulus of the well and setting a permanent plug in the well, the method comprising: determining that a control line is present in the well based on data obtained from an application database; obtaining a first set of diagnostic parameters collected by one or more sensors; determining a position of the control line based on the first set of diagnostic parameters, wherein the position comprises a location and / or an orientation of the control line; verifying ablation of the control line over an ablated length associated with the target depth range based on a second set of diagnostic parameters; and if ablation of the control line is complete and verified, initiating installation of cement within the annulus for setting the permanent plug. The method may comprise obtaining a first set of diagnostic parameters collected by one or more sensors before ablation of the control line, wherein the one or more sensors comprises at least one ultrasonic tool. The first set of diagnostic parameters may comprise initial well parameters, cement bond logs or variable density logs over a target depth range of the well, or ultrasonic sensor data. After perforation of the control line is initiated, obtaining a second set of diagnostic parameters from the one or more sensors may comprise one or more ultrasonic signals indicative of whether the control line is perforated.It is disclosed a method for validating a well intervention procedure for a well, wherein the well intervention procedure comprises forming a barrier over a target depth range, the method comprising: obtaining a set of diagnostic parameters; using the set of diagnostic parameters for planning the well intervention procedure, wherein planning the well intervention procedure comprises determining one or more operational steps followed during execution of the well intervention procedure; obtaining, from a network of sensors, a set of procedure parameters collected during the well intervention procedure; using the set of procedure parameters to monitor execution of the well intervention procedure. The set of verification parameters may comprise an acoustic track over the target depth range. One or more signal peaks in the acoustic track may be associated with one or more leaks.The well intervention procedure disclosed above may be a Plug and Abandonment (PA) procedure.It is disclosed a computer-readable medium storing instructions for performing any of the methods disclosed above.It is disclosed a system comprising: a network of sensors, wherein the network of sensors are configured to detect data associated with monitoring planning or operation of a well intervention procedure for a well; a processor configured to obtain data from the network of sensors and communicate with an application database; and a memory storing instructions that, when executed, cause the system to perform any of the methods disclosed above. The the well intervention procedure may be a Plug and Abandonment (PA) procedure.Diagnostic parameters from a network of sensors or from application database may include one or more of the following: initial well parameters, such as tubing schematic, casing schematic, well history, pressures, temperature, well condition etc.; cement bond logs or variable density logs over a target depth range of the well; ultrasonic imaging tool, an acoustic sensor or gamma ray log.Diagnostic parameters from application database may include hydraulics parameters (such as such as flow rate and pressure) and conveyance parameters (such as torque and drag) associated with a planned well intervention procedure.Parameters obtained before the barrier setting may include one or more of the following: gamma ray log, casing collar locator log, cement bond log, variable density log, parameters for control line and clamp location and orientation.Parameters obtained during the barrier setting may include detected particles during a washing operation, preferably estimated particle mass lifted from an annulus versus particles in place in the annulus before washing; or ablation of control line over the length of the barrier.Simulation may include at least one of the following:- hydraulics calculations such as ball drop calculations, pressure drop calculations, volume calculations, or calculations associated with cementing the well;- conveyance module configured to estimate downhole forces based on torque and drag calculations;- estimating downhole forces based on the set of diagnostic parameters.A computational fluid dynamics (CFD) module may be configured to perform further hydraulics calculations comprising any of: cement slumping calculations; washing efficiency calculations; listing of particles calculations; multiple annuli calculations; or calculations associated with tubular, tubing, or casing designs.Perforation ablation module may be configured to verify correct perforation and ablation or determine mechanical perforation and ablation parameters including one or more of geometry and depth of perforations; perforation depth; or perforation gun setup.Determining that a control line is present in the well may be based on at least one of:- a cement bond log; or- an ultrasonic log including obtaining ultrasonic images or measurements from an ultrasonic tool within the well after ablation of the control line is initiated; verifying ablation of the control line over an ablated length based on the ultrasonic images or measurements associated with perforations of the control line; or phased-array oriented beam;- acoustic sensor;- based on data obtained from an application database.Determining position (location or orientation) of control line may be based on ultrasonic imaging or measurement, or acoustic log.Procedure parameters collected during the well intervention procedure may be one or more of:- a washing pressure flowrate track over the target depth range;- a washing pressure flowrate track over the target depth range associated with detecting acoustic signals generated by a washing fluid flow, or- a cementing track over the target depth range associated with installation of cement within the well.The washing pressure flowrate track over the target depth range may be associated with detecting acoustic signals generated by a washing fluid flow. The washing pressure may be the pressure from topside.Verification parameters may be at least one of:- an acoustic track over the target depth range;- a pressure and / or temperature sensor arranged below the barrier;- validation of barrier integrity derived from acoustic signal across the barrier;- barrier gas permeability (e.g. attach a canister of gas below the barrier that is emitted in a controlled way and with gas detection on surface);- Drill out CBL / VDL.Validation based on determining a barrier quality value may include:- a quality parameter track calculated based on values calculated for the execution of the barrier setting and verification values for the set barrier along the barrier length;- a number of tracks for the parameters obtained before, during and after the barrier setting and the calculated quality parameter track representing a visualization of the parameters.The parameters and tracks involved may vary depending e.g. on the well and procedure performed.Brief Description of the Drawings

[0056] While various features of the invention have been presented in general terms above, some examples of embodiment are given in the following with reference to the drawings, in which:

[0057] Figure 1 is a schematic view of an annulus washing tool arranged inside a well pipe;

[0058] Fig. 1 B is a particle count time based showing a reduced number of particles over time indicating progressing washing / clean-up effect.;

[0059] Fig. 1C is an energy histogram time based with dark line showing flow noise / noise base line.

[0060] Fig. 1 D is a signature log time based showing particle impacts versus particle size.

[0061] Figure 2 depicts a subsea well with a casing before perforation;

[0062] Figure 3 depicts a casing after the perforation tool has perforated the casing and severed the control line at several places;

[0063] Figure 4 shows an ultrasonic imaging tool arranged inside the casing;

[0064] Figure 5 depicts a schematic illustration of a perforation with a control line still being present;

[0065] Figure 6 depicts a schematic illustration of a perforation without the control line being present;

[0066] Figure 7 shows a flowchart illustrating a method of validating a subsea well annulus washing operation;

[0067] Figure 8 shows a flowchart illustrating a method of setting a permanent plug in a subsea well having a well pipe and a control line located in an annulus outside the well pipe;

[0068] Figure 9 shows a system for optimizing a PA procedure;

[0069] Figure 10 illustrates an example implementation of system 900 used for monitoring a PA procedure;

[0070] Figure 11 shows a flowchart illustrating an analytics system for use in monitoring or planning a PA procedure;

[0071] Figures 12A-12C show a flowchart illustrating a circular method for optimizing barrier setting and verification;

[0072] Figure 13A illustrates an example confirmation and verification output from a PA procedure performed using any the methods, apparatus, or systems described herein;

[0073] Fig. 13B illustrating tracks of an exemplary answer product. Figure 13C illustrating exemplary parameters and tracks of a further exemplary answer product.

[0074] Figure 14 depicts a flowchart illustrating a method for controlling a PA procedure;

[0075] Figure 15 depicts a flowchart illustrating a method for simulating a PA procedure;

[0076] Figure 16 depicts a flowchart illustrating a method for monitoring a PA procedure;

[0077] Figure 17 depicts a flowchart illustrating a method for diagnosing a PA procedure;

[0078] Figure 18 depicts a flowchart illustrating another method for monitoring a PA procedure;

[0079] Figure 19 depicts a flowchart illustrating a further method for monitoring a PA procedure;

[0080] Figure 20 depicts a flowchart illustrating another method for diagnosing a PA procedure;

[0081] Figure 21 illustrates a system for diagnosing, monitoring, or optimizing a PA procedure according to the methods disclosed herein; and

[0082] Figure 22 shows an example computing system for performing any of the method steps or system processes described herein.

[0083] The included drawings are for illustrative purposes as to provide examples of possible structures and arrangements for the disclosed inventive apparatuses, systems and methods associated with well intervention methods e.g. the plugging and abandonment of wells, such as subsea wells. The drawings listed above do not restrict or limit amendments to the described embodiments of the invention that may be performed by a person skilled in the art without departing from the spirit and scope of the invention. The embodiments of the detailed description below, either alone or in combination with the accompanying drawings listed above, do not limit the inventive concepts described herein to the exact embodiments and drawings disclosed.Description of the example Embodiments

[0084] Example embodiments are described with reference to the drawings. The same reference numerals are used for the same or similar features in all the drawings and throughout the description. The example embodiments are examples only and not limiting for the invention. The examples are described for P&A procedures, however the examples may also be applicable for other well intervention procedures that include well barriers.A permanent barrier may be a single barrier (about 30 meter) or a dual barrier (about 60 meter). Whether a single barrier or dual barrier is selected depends on e.g. the requirements from regulators and the risk scenario for the particular well.Before an operation can be performed in a field, planning and simulation of the operation must be performed and the plan and the result of the simulations including the risk of the operation, must be presented to the authorities / regulators. The authorities / regulators must approve the plan before execution of the operation may take place. The plan must show in detail how the procedure of plugging the well is to be performed. There is a need to show to the regulators the best approach for the P&A procedure.P&A is a high-risk procedure and there is a need for a in improved way of controlling the end result. There is also a need for choosing the verification method based on risk profile and also with respect to costs for the specific operation.After the well has been plugged in a P&A operation the authorities / regulators want proof that the well has been safely and permanently plugged for eternity according to the requirements for the specific field.The steps of the barrier setting and verification workflow comprise a number of steps: planning, simulating, risk assessment, execution and verification.The barrier may be set in a single or multiple tubular / annulus scenario, including e.g. a single tubular with annulus, a dual tubular with annuli, a triple tubular / annulus. Dual annulus means that there are either two tubulars and open hole on the outside or three tubular strings with two annuli between tubulars or open hole on the outside. Single annulus means one or two tubulars or open hole on the outside of first tubular.A control line may be present or not in an annulus. Control line could be a single hydraulic (hollow) tubular, a number of tubulars inside a flatpack or an electric gaugeline on the outside of typically a tubing string. A control line represents a risk for a potential leak path through the permanent barrier formed in the P&Aoperation. It is thus important to ensure that the control line is completely ablated (severed). The control line is fixed to the inside of the annulus by use of clamps. Identification of the position and direction of these clamps are also important to ensure complete ablation of the control line. The control line cannot be ablated in a position of a clamp. Ablation of control line takes place between the clamps.A method is disclosed for a barrier setting in a wellbore. The method comprising: planning the barrier setting; simulating the planned barrier setting; creating an operation plan for the barrier setting; performing a risk assessment for the operation plan; executing the operation plan setting the barrier in the wellbore; and verifying the barrier in the wellbore.Executing the operation plan may include ablating a control line in an annulus of the wellbore and confirming ablation of the control line. A quality of the barrier in the wellbore may be calculated. The barrier quality may be calculated over the barrier length. The barrier length may correspond to an ablated length of the wellbore. A barrier quality may be calculated for a length of the barrier. A barrier quality may be displayed for a length of the barrier. Diagnostics of the well may be performed before simulation.A method for a barrier setting in a wellbore includes executing an operation plan setting the barrier in the wellbore; and verifying the barrier in the wellbore. Executing the operation plan may include ablating a control line in an annulus of the wellbore and confirming ablation of the control line. A quality of the barrier in the wellbore may be calculated. The barrier quality may be calculated over the barrier length. The barrier length may correspond to an ablated length of the wellbore. A barrier quality may be calculated for a length of the barrier. A barrier quality may be displayed for a length of the barrier. Diagnostics of the well may be performed before simulation.A method in a barrier setting in a wellbore includes detecting a control line location and orientation in the wellbore. The control line is ablated over at least a length of the wellbore. Verifying ablation of the control line is performed over the ablated length of the wellbore. Verifying ablation of the control line by performing downhole diagnostics. Verifying ablation of the control line may be performed byultrasonic logging. Verifying ablation of the control line may be performed using an imaging tool.A method in a barrier setting in a wellbore includes washing an annulus, and detecting, by use of acoustics, a particle flow in the annulus during washing. A total mobilized mass of particles in the annulus may be calculated based on the detected particle flow.A method for establishing a quality of a barrier setting in a wellbore, the method comprising calculating at least one quality parameter for the barrier based on at least one parameter obtained during the barrier setting and at least one verification parameter obtained in a verification of the barrier.The at least on parameter obtained during the barrier setting may comprise cementing flow rate and pressure. The at least one parameter obtained during the barrier setting may be related to ablation of a control line along the barrier. The at least one parameter obtained during the barrier setting may be washing pressure.The at least one parameter obtained during the barrier setting may be related to monitoring of washing particles during the washing of the wellbore.The at least one quality parameter of the barrier may be calculated for a number of intervals over the barrier length. The at least one quality parameter may be displayed in intervals along the depth of the wellbore.A product for qualification of a barrier setting in a well. The product includes a number of tracks for parameters obtained before, during and after the barrier setting in intervals along a depth of the wellbore. The product also includes a calculated quality parameter track representing a quality of the barrier in intervals along the depth of the wellbore.The parameters obtained before the barrier setting may include parameters related to at least one of gamma ray log, casing collar locator log, cement bond log, variable density log. The parameters obtained before the barrier setting may include parameters related to control line and clamp location and orientation. Theparameters obtained during the barrier setting may include parameters related to cementing flow rate and pressure, preferably cementing flow rate and pressure versus set parameters. The parameters obtained during the barrier setting may include parameters related to detected particles during a washing operation, preferably estimated particle mass lifted from an annulus versus particles in place in the annulus before washing. The parameters obtained during the barrier setting may include parameters related to ablation of control line over the length of the barrier. The parameters obtained after the barrier setting may include parameters related to acoustic monitoring of the barrier along the barrier length to verify the barrier. The quality parameter track is calculated based on values calculated for the execution of the barrier setting and verification values for the set barrier along the barrier length. The number of tracks for the parameters obtained before, during and after the barrier setting and the calculated quality parameter track represent a visualization of the parameters.A method for a barrier setting in a wellbore including: logging the wellbore; ablating a control line in the wellbore; confirming ablation of the control line; washing the wellbore; placing a fluidized plugging material in the wellbore creating a barrier in the wellbore; and verifying the barrier.Method for barrier setting and verificationA method for barrier setting and verification is explained below for a single or multiple tubulars / annuli with reference to Fig.12A-C.First it is decided 1204 whether a control line is present or not.If a control line is not present in the annulus, the following steps are performed: o Perform diagnostics I is an optional step 1206. Diagnostics is performed using a cement bond log or optionally ultrasonic logging. o Perform TCP (Tubing Conveyed Perforation) 1208. o Install a cement base 1210.If a control line is present in the annulus, the following steps are performed:o Perform diagnostics I is an optional step 1212. Diagnostics is performed using a cement bond log or optionally ultrasonic logging. o Perform diagnostics II 1214. Diagnostics II identify control line orientation and / or location. WL ultrasonic log using phased-array oriented beam may be used for identifying location and direction of the clamps and control lines. Identifying of location and orientation of control line and clamps is important to be able to efficiently ablate the control line. o Ablate control line 1216. Ablation (cutting / severing) of Control lines or flatpack(s) may be performed using explosives, abrasive, mechanical, plasma, chemical explosives, laser etc. Ablation must not penetrate a second tubular. o Perform diagnostics III 1218 to verify ablation of control line. Verification of ablation of control line may e.g. be performed by wireline ultrasonic log using phased-array oriented beam. Ultrasonic log may “see through” slots / cuts in the tubular to verify control line is removed (“not present”). The verification of ablation of control line is performed over the total ablated length (i.e. the total length of the annulus where ablation / severing of control line has been performed). o Install a cement base 1220.After installing the cement base, an evaluation is performed 1222 to establish whether the pipe is free or whether there is material in annulus.If there are completion fluids 1224 in the annuli, the annulus is washed 1226. The cement plug is placed 1238 in the annuli and in the main bore.If there is mud (settled mud and solids) 1228 in the annuli, the annulus is washed 1230. The cement plug is placed 1238 in the annuli and in the main bore.If the annuli is cemented 1232, the cement is broken up 1234 and annulus washed 1236. The cement plug is placed 1238 in the annuli and in the main bore.During washing 1226, 1230, 1236, acoustic detection of particle flow may be used to confirm washing efficiency in single or multiple annuli. Washing ismonitored and also the total produced mass is monitored to determine if annulus is cleaned.After cement plug is place in 1238, drill out and log are optionally performed.If drill out and logging is performed, drill out of cement plug is performed in 1242. Cement bond logging or ultrasonic logging is performed in 1244. Logging may be performed downwards and sideways. The cement plug is then placed in the well in 1246.The cement plug is verified 1248 with pressure test and / or by use of acoustic detection techniques. The barrier is now tested successfully.New barrier depth is optionally selected and replanned.A barrier answer product qualification report is made available for the barrier setting.Further details of the method described above are disclosed later in the description.Barrier setting workflowAn exemplary barrier setting and verification workflow includes, but is not limited to:1 . Initial Well Parameters (The «well» as is)2. Planning I Objectives - Intervention Target (What needs to be done in the well)3. Application Data Experience - Previous experience in similar well scenarios4. Diagnostics (Pre-intervention log data)5. Initial Risk Assessment (Evaluation of risk factors during planning stage)6. SIM - Simulations to confirm well intervention operational steps7. Operation Plan (DOP - Detailed Operating Procedure)8. Final Risk Assessment (Evaluate risk before starting operation)9. Execution as per Operation Road Map (Human Intelligence) or Operation Protocol (Al)10. Answer Product (Combining Diagnostics, Execution and Verification in Auditable Report)11 . Learnings (collect observations and learnings for next well planning)12. Applications Data Base (Collect all relevant data in data base)Fig. 9 is a block diagram for an exemplary method and system for barrier setting and verification workflow including job planning, simulation, risk assessment, execution and verification.In Fig. 9, the method and system 900 comprises the following units: initial well parameters 902; planning objectives 904; application data experience 906; diagnostics 908; initial risk assessment 910; simulations 912; operation plan (DOP) 914; risk assessment 916; execution 918; answer product 920; learnings 922; and application database 924. Simulations 912 comprises any of computation fluid dynamics (CFD) module 912-A, conveyance module 912-B, hydraulics module 912-C, or perforation / ablation module 912-D. Execution 918 comprises any of analysis module 918-A, sensor / data stitching module 918-B, road map module 918-C, or operation protocol module 918-D.In the kick off P&A planning objectives for the intervention target; i.e. what needs to be done in the specific well, is defined. The planning objective is based on well conditions, platform and vessel conditions and equipment selection. The well conditions include initial well parameters (the well as is). The P&A rules are determined, well conditions assessed, and the number and position of barriers in the well are optimized.The P&A barrier setting may be designed for setting a barrier without a rig (rigless) or by use of a rig.A simulation of the well operation for the barrier setting is performed by CFD (Computational Fluid Dynamics) and / or OPS based on all the data, information, rules etc in the planning step. The simulation is performed for all the steps of the planned barrier settings for the different scenarios of single or multiple tubulars / annuli for the operation. The simulation of the well operation for the barrier setting is explained in detail later in the description.Answer ProductAfter the well has been plugged in a P&A operation the authorities / regulators want proof that the well has been safely and permanently plugged for eternity according to the requirements for the specific field. As explained above, a barrier answer product qualification report is made available for the barrier setting.The answer product can be viewed by a person on a suitable medium including, but not limited to, a display, a screen, on paper or a format easy to understand by the owners of the well and the authorities / regulators. The answer product may e.g. be a digital product, a software. The answer product may provide information about the barrier setting process as the operation proceeds in e.g. real time or near real time. The answer product may be viewed or optionally interacted with by e.g. a human, a machine, a humanoid, a robot or an android.An exemplary answer product is illustrated in Figure 13, e.g. Figure 13B.The answer product report has a number of sections including:A. New diagnostics or old logs: Information for execution planning. Diagnostics information used for planning well intervention operation.B. Execution road map / protocol: Delivery & QC of step-by-step protocol of operation. Execution Road Map I Protocol displays actual operation data and progress versus planning, with pass (operational parameters fulfilled) or no pass (executed parameters fall out of accepted range).C. Verification: Validates barrier placed during execution with pass or no pass. Verification has binary outcome pass (no leak detected) or no pass (leak detected).D. Results: Display barrier image as actual length of placed barrier and verification. Results display barrier placed as length required and as tested(pass or no pass). The barrier length image and the barrier quality image is displayed based on the results for the execution in B and the verification in C.In Fig. 13B each of the sections A, B, C and D of the answer product are each divided into tracks. The sections all show the length of the barrier with respect to well measured depth. The information in all the tracks are plotted to show the parameter(s) in the track along the barrier length. The parameters and information in each section and in each track of each section are therefore easy to visually relate to the barrier length image and the barrier quality image.The sections and their tracks are explained in more detail below.A. Diagnostics or “old” logsEither «new» or «old» data acquisition (logging). Information for depth control inside the tubing, tubular condition and bonding and material in annulus. Data is used for planning of well intervention operations in the well execution phase. The values in each track are plotted with respect to depth.Diagnostics Tracks:A 1: Depth, gamma ray (GR), casing collar locator (CCL) Track. The depth, GR and CCL are presented side by side: Track = Depth control.A 2: Cement Bond Log (CBL) = CBL value and VDL image (variable density log). CBL and VDL are presented side by side. CBL value used to define CBL wash indicator (Wl) to indicate success chance of washing annulus section. Wl is proportional to the CBL value and the CBL scale is typically displayed between 0 and 100 percent bonding which is further separated into good, medium and bad sections. A CBL value > 15 indicates typically a good washability and is then generally good to proceed. A CBL value between 10 and 15 indicates a medium washability and risks must be assessed. A CBL value < 10 indicates typically a bad washability and requires further considerations; risks assessed. The CBL value can be converted into Bl (Bond Index to normalize). Could be based onCBL value (washability index) / CBL index, Ultrasonic log or mass flow rate / permeability or any other methods suitable.As : Ultrasonic and / or Multifinger track. Log = Cement Bond evaluation and Inner casing corrosion & deformation. Optional based on availability.A4 : Control line track. Control line and clamp location and orientation image. Optional based on presence of control lines.B. Execution road map / protocol.Displays actual operation data and progress versus planning.Execution tracks:Bi: Perforation on ablation track. Shows the ablations on the tubular along the barrier. Shows the ablations with respect to the control line. Optional based on presence of control lines.B2a: Washing pressure flowrate track. Optional based on presence of control lines.B2b: Washing particles track. Optional.Bs: Cementing track.Execution of the operation road map or protocol is performed. For presence of control line in annuli, control line ablation is important. The barrier road map or protocol divides the planned barrier section into incremental steps to verify and log planned parameters versus actual parameters. The output is an Execution Quality Flag (EOF) calculated from the incremental operational steps that build up the barrier. If the parameters are within an expected range for the specific incremental interval, then the execution quality flag for this step is equal to 1 . EQF is calculated by (E1 +E2+E3+E4+...+En) / n.Bi Ablation = Confirmation of ablation depths and calculation of actual ablated length LAA. LAA = Confirmed ablated length [m] over total perforated length [m] (either defined by length or by number of ablation points).B2aWashing Pressure = The washing pressure and flowrates are to stay between the set maximum flow pressure and the minimum flow pressure toensure good washing efficiency. Pressure and flowrate are set in the road map / protocol.B2b Washing Particles = Using acoustic particle flow detection to monitor successful lifting of particles from annuli (estimated particle mass lifted versus originally in place).Bs Cement Placement = Monitor cementing flow rate and pressure versus set parameters.Example:Calculation of EQF for incremental depth step n (n=2500m-2501m) f EQF) = Ablation Length + Washing Pressure + Washing Particles+ Cement PlacementAblation + Wash Pressure +Wash Particles + Cementing 1 = Execution Rate Ex= (1+1 +1 +1 ) / 4 = 1EQF rating: 0.8 < 1 :Good; 0.6 < 0.8 : Medium; < 0.6 BadC. VerificationVerification of placed barrier during execution.Verification tracks:Ci : Acoustic track. The barrier is tested by acoustic monitoring along the entire barrier. Defects in the barrier may be displayed with visualization of barrier defect and corresponding signal.C2 : Drill out and CBL / US track: CBL / Ultrasonic log interpretation. Optional.C3 : Pressure temperature (P / T) track: sensor arranged below barrier. Optional.C4 : Barrier gas permeability test track: Measures gas permeability of the barrier. Optional.Verification outcomeVerification has binary outcome pass. Result of each verification method is pass (no leak detected) or no pass (leak detected).Ci : Verification: No leakC2 : Verification: CBL / Ultrasonic log interpretation.C3 : Verification: No leakC4 : Verification: No leakD. ResultsPresent barrier as actual length of placed barrier and verification. The barrier may be displayed as barrier image as actual length of placed barrier and verification. The barrier may be displayed on a screen, as an image, presented on paper, visualized for a human, a robot, a humanoid, an android, provided in a digital format etc.Tracks:Di :Actual Barrier Length track, (from depth 1 to depth 2) D2 : Actual Barrier Quality track.The barrier length image and the barrier quality image are based on the results for the execution in B and the verification in C as follows.Execution Bi-4 * Verification C1-4 = Barrier Length & Quality ImageThe barrier quality track may represent the barrier quality by use of flag.Flag - Good = pass. Flag - No Good = no pass. Color may be used as flag. E.g. green = good and yellow = no good. Should be easy to visually understand the quality and at which depth of the barrier there are potential issues. (%) = (A + B + C) * DI * D2 * D3 * D4 f(x) = (Ablation Length + Washing Efficiency + Cement Placement) * Verification 1-4A+B+C = Actual Barrier LengthD1 ...4 = Absolute Barrier Verification (go / nogo)Example establish rock-to-rock barrier and verify by answer product to calculate and verify actual delivered barrier in the well.Example layout of product o Before log data o Access to annulei & ablation o Execution wash & cement o Verification o Final Barrier DeliveredExample sequence: o Log before - WL Diagnostics o Perforation & Ablation o Confirm Ablation o Washing also including e.g. plug and perforation o Cement Placement o VerificationInital scenario showing:Tracks for depth, GR, CL (before), control line and clamps locations. Inital scenario including visualizations of identified position of clamps and control line performed by e.g. use of imaging tool.Execution scenario showing:- Perforation and ablation track: e.g. with oriented ablation and 360 helix.- Ablation confirmation track: total ablation = ablated / non-ablated length.- PWC washing efficiency track including subtracks: good - medium - bad washing logged each meter (e.g. by use of color coding green, yellow, red) and acoustic detection of particle flow to confirm washing in annulus 1 & 2.- Cementing track: Good - Medium - Bad Cement Placement logged each incremental interval (e.g. vizualized by use of color coding).Execution scenario including:-Confirm ablation depths of control line and total length of barrier without control line.- Washing efficiency log by depth- Cement placement execution log by depthOptions for control line / clamp logging before and after ablation may be found in both initial scenario and in execution scenario. Control line and clamp logging may be performed by imaging tool.Verification scenario:- Verification acoustics I track.- Verification II drill out & CBL track.- Verification III - Pressure / temperature (P / T) sensor below barrier with or without data transmission.- Verification IV - Barrier permeability test with gas released below barrier.If no drill out, then stationary acoustic log from above downwards Continuous acoustic log if drilled out.Final barrier result success scenario:L1 + L2 > Required barrier length.Displayed in green color if success.Answer product provides:- calculations and verified actual delivered barrier in the well.- unique workflow including ablation and confirmation of ablation.- a new approach to define successful barrier and calculate actual barrier length delivered based on actual ablated length.Ablation confirmation is provided.Detection of particle flow with acoustics enabling calculation of total mobilized mass of particles in the washing process.ExampleAn exemplary answer product for a well intervention procedure is illustrated in Figure 13C. An output 1300 comprises relevant data for assessing and verifying the well intervention procedure. Output 1300 is therefore suitable for determining the success of the well intervention procedure. In the example in Figure 13C the most relevant data is shown in a machine-readable and / or displayable format. The number of parameters involved varies depending on the procedure. Measurements are performed during the procedure to check whether the procedure are in line with the expectations. Output 1300 comprises depth (meter) and velocity (meter / min) 1302, inflow (liters / minute); pressure (bar) 1304; hookload (metric ton) 1308; depth range 1310; a first barrier quality flag 1320; and a second barrier quality flag 1322. In Figure 13C the column to the right illustrates the barrier. The barrier is divided into segments along the length of the barrier illustrated with the quality flags in Figure 13C. The upper curve 1302 is velocity (meter / minute). The second curve from above is inflow (LPM: liter / minute). The third curve from above is pressure (bars). The lowest curve is hookload (metric ton). The measured parameters involved in Figure13 C are hydraulic parameters and mechanical parameters. The hydraulic parameters are flow rate vs pressure and hookload. Hookload indicate downhole weight on the tool. Each parameter in isolation indicates if the execution of the well intervention operation is in line with the expectations for that parameter. The combination of the monitored hydraulic parameters is used to verify if the execution of the well intervention procedure is in line with expectations and thus indicate whether the overall outcome will be successful or not.The barrier illustrated in 1304 is provided with indication as to the quality of each segment of the barrier. The indication may e.g. be use of colors which has been explained above.

[0085] Figure 1 depicts an annulus washing tool 100 configured for washing an annulus 101 outside a well pipe, here in the form of a casing 103 installed in a subsea well 105. The well pipe (i.e. casing) 103 has an inner bore 102.

[0086] The annulus washing tool 100 is part of an annulus washing assembly 10, which is installed in the subsea well 105. The annulus washing assembly 10 can typically comprise a drill string or coiled tubing (not shown) or another string for suspending the annulus washing tool 100.

[0087] It shall be appreciated that, although the shown tool is termed an annulus washing tool, it may be part of a plug setting tool string or other combination tool, configured to perform several additional steps needed to set a wellbore plug (typically perforation, washing and cementing - PWC). The annulus washing tool 100 is run on a drill string that extends down from a surface structure (not shown) at the sea surface. In some embodiments, the washing tool 100 could instead be run on a coiled tubing or other means.

[0088] It should also be appreciated that the enclosed apparatus, e.g. the annulus washing tool 100, systems, e.g. system 900, or methods, e.g. method 700, are not limited to operations associated with subsea wells, and may be applied to nonsubsea wells or alternative structures.

[0089] In the situation shown in Figure 1 , the casing 103 has already been perforated with a perforation tool (not shown, e.g. perforation tool 123 of Figure 3). The resulting perforations 103a constitute fluid communication between the inner bore of the casing 103 and the annulus 101 outside of it.The annulus washing tool (100) is arranged at a perforated section of a well pipe (103), providing a washing fluid flow (110) through perforations (103a) in the well pipe, into the annulus (101 ), and back through perforations into the inner bore (102) of the well pipe.

[0090] The annulus washing tool 100 comprises an upper wash cup 107 and a lower wash cup 109 (sometimes also called swab cups), which are arranged at a mutual distance along the tool. In the shown embodiment, there are also arranged an upper back-up wash cup 107a and a lower back-up wash cup 109a. The washcups are configured to seal against the inner bore of the casing 103. The skilled reader will appreciate that other types of washing tools can be without wash cups.

[0091] Between the upper and lower wash cups 107, 109, there is a plurality of wash fluid ports (e.g. nozzles) 111 . As indicated with one of the arrows shown in Figure 1 , a washing fluid flow 110 exits the wash fluid nozzles 111 and flows through perforations 103a at the location between the upper and lower wash cups 107, 109. The washing fluid flow 110 flows further upwards and re-enters the casing bore through perforations 103a above the upper back-up wash cup 107a. The washing fluid flow 110 further flows upwards (i.e. towards the surface) through the inner bore 102 of the casing 103.

[0092] This washing process cleans the annulus 101 for particles 113, such as settled barite. During the cleaning process, the annulus washing tool 100 is moved downwards through the casing 103, thus cleaning the annulus 101 along a predetermined length.

[0093] Also schematically shown in Figure 1 , the annulus washing tool 100 comprises an acoustic sensor 115. The acoustic sensor 115 detects acoustic signals, e.g. sounds, originating from the flow of washing fluid. The presence of particles 113 during the washing procedure will affect this sound. Thus, the sound produced by washing fluid containing particles will be different from the sound produced by washing fluid without particles. Consequently, the operator is enabled to detect when the particles in the annulus 101 has been successfully washed away with the washing fluid.

[0094] Still referring to Figure 1 , the annulus washing tool 100 can further comprise an auxiliary acoustic sensor 117. The auxiliary acoustic sensor 117 is arranged above the acoustic sensor 115 and can detect sound originating from the upwardly directed flow in the main bore of the casing 103. Preferably, the auxiliary acoustic sensor 117 can be arranged above the perforated section of the casing 103 during the washing procedure. All the washed-out particles 113 will then flow through the inner bore 102 of the casing and past the auxiliary acoustic sensor 117.

[0095] Based on detected acoustic signal, e.g. recorded sound, from the auxiliary acoustic sensor 117, or recorded sound from both the auxiliary acoustic sensor 117 and the acoustic sensor 115, one may calculate the amount or mass ofparticles 113 that has been washed out from the annulus with the washing fluid flow 110.

[0096] The annulus washing tool 100 can further or alternatively comprise an upper acoustic sensor 117. The upper acoustic sensor 117 is arranged above the acoustic sensor 115 and can detect acoustic signals originating from the upwardly directed washing fluid flow 110 in the casing bore 102. Advantageously, the upper acoustic sensor 117 is arranged above the perforated section of the casing 103 during the washing procedure. All the washed-out particles 113 will then flow through the casing bore 102 and past the upper acoustic sensor 117.

[0097] While the upper acoustic sensor 117 in the shown embodiment is a part of the annulus washing tool 100, in other embodiments the upper acoustic sensor 117 could be located further above the annulus washing tool 100. For instance, the upper acoustic sensor 117 can be supported above the annulus washing tool 100, by the string (not shown) supporting the annulus washing tool 100.

[0098] Based on detected acoustic signals or sound from the upper acoustic sensor 117, and / or recorded acoustic signals from the acoustic sensor 115, one may calculate the amount or mass of particles 113 that has been washed out from the annulus with the washing fluid flow 110.

[0099] Knowing the amount of material or particles 113 that has been removed from the annulus 101 may assist in validating that the annulus 101 has been sufficiently washed. By knowing the volume of the washed section of the annulus and comparing it to the amount of washed-out material, one can indicate how much of the material has been removed (washed out).

[0100] One can calculate an annulus wash index by dividing the amount of washed-out material by the initial amount of material present in the annulus. For instance, the amount of washed-out material divided by the initial amount of material can equal 0.95, giving an annulus wash index of 0.95 or, alternatively 95 %.

[0101] Estimation of the initial amount of material in the portion of the annulus that shall be washed can be based on one or more of well logs, fluid type, and well age.

[0102] With the acoustic sensor 115 and / or the upper acoustic sensor 117, one can provide a particle energy histogram, of which an example is shown in Fig. 1 B. The particle energy histogram shows the distribution of the particle sizes of the material that has been removed from the annulus 101 and that is lifted with the washing fluid flow 110. This enables the operator (i.e. with a computer) to calculate the amount of material that has been removed and lifted up from the annulus 101 .

[0103] Fig. 1 C and Fig. 1 D are showing energy histogram and signature log respectively which can be used to compute distribution of the particle sizes of the material that has been removed from the annulus. To distinguish the detected acoustic signals that originates from particles from the noise of the fluid flow, one may use multi-frequency filtering / thresholding. Furthermore, one can advantageously use simultaneous (parallel) processing in both time domain and frequency domain.

[0104] The thick horizontal line in Fig.1 C is the flow noise (noise baseline). The acoustic sensor 115 and / or the upper acoustic sensor 117 can preferably be optimized for particle detection. The sensors that are optimized to be used in this application are typically fast response, high frequency sensors.

[0105] As the skilled person will appreciate, when using terms like above and below, or upper and lower, it is referred to the positions along the extension of the well, which indeed may be non-vertical and even horizontal. Hence, above and upper are closer to the wellhead than the terms below and lower.

[0106] While Figure 1 depicts a situation where one well pipe (casing) is perforated, there could also be two perforated well pipes such that the washing fluid flows into a first and a second annulus.

[0107] Figure 2 depicts the subsea well 105 with the casing 103 before perforation. Furthermore, a control line 119 is arranged in the annulus 101 , typically attached to the casing 103. To prevent the control line 119 from constituting a possible leak path when installing a plug in the subsea well 105, the control line 119 is removed along a portion of the casing 103.

[0108] This is done by severing the control line 119 with a perforation tool 123, as shown in Figure 3. When the perforation tool provides the perforations 103a in the casing 103, the control line 119 is simultaneously severed.

[0109] Figure 3 depicts the situation after the perforation tool 123 has perforated the casing 103 and severed the control line 119 at several places. Remaining portions of the control line 119 are visible above and below the perforated section. Moreover, a plurality of control line pieces 119a are shown.

[0110] Figure 4 shows an ultrasonic imaging tool 121 arranged inside the casing 103. With the ultrasonic imaging tool 121 , the operator obtains images of the inner bore 102 of the casing 103, including the perforations 103a. The operator can then see if there is a remaining control line 119 behind the perforations 103a. This is schematically illustrated with the example images shown in Figure 4 and Figure 5.

[0111] Figure 5 depicts a schematic illustration of a perforation 103a with a control line 119 still being present. Figure 6 shows a similar illustration, however without the control line 119 being present.

[0112] As the skilled person may know, before perforating the casing 103 for severing the control line 119, an imaging tool can be used to detect the angular location of the control line 119. In this manner, the operator can perforate the casing 103 at the angular position of the control line 119. If the angular location of the control line 119 is unknown, a perforation pattern can be used, which will ensure that the control line 119 is severed. A helical perforation pattern can for instance be used.

[0113] Figure 7 shows a flowchart illustrating a method of validating a subsea well annulus washing operation. Method 700 comprises the following steps: step 702, providing a washing fluid flow from wash fluid nozzles through perforations; step704, detecting acoustic signals generated by the washing fluid flow; and step 706, detecting a change of acoustic signals.

[0114] Step 702 comprises, with an annulus washing tool arranged at a perforated section of a well pipe, providing a washing fluid flow from wash fluid nozzles through perforations in the well pipe, into the annulus, and back through perforations into the inner bore of the well pipe. For example, annulus washing tool 100 provides a washing fluid flow 110 from wash fluid nozzles 111 through perforations 103a.

[0115] Step 704 comprises, with an acoustic sensor being part of the annulus washing tool, detecting acoustic signals generated by the washing fluid flow. For example, acoustic sensor 115 detects acoustic signals generated by the washing fluid flow 111.

[0116] Step 706 comprises detecting a change of acoustic signals detected in step 704, e.g. using acoustic sensor 115.

[0117] Figure 8 shows a flowchart illustrating a method of setting a permanent plug in a subsea well having a well pipe and a control line located in an annulus outside the well pipe. The method comprises the following steps: step 802, providing perforations in the well pipe for removal of the control line; step 804, recording ultrasonic images of the perforations; step 806, inspecting the ultrasonic images and verifying removal of the control line; and step 808, washing the annulus and installing cement in the annulus.

[0118] Step 802 comprises, with a perforation tool, providing perforations in the well pipe for removal of the control line along a portion of the well pipe by severing the control line when providing said perforations. For example, perforation tool 123 provides perforations 103a for removal of control line 119.

[0119] Step 804 comprises, with an ultrasonic imaging tool arranged in the bore of the well pipe, recording ultrasonic images of the perforations. For example, ultrasonic imaging tool 121 records ultrasonic images of the perforations 103a.

[0120] Step 806 comprises inspecting the ultrasonic images and verifying removal of the control line, e.g. removal of control line 119.

[0121] Step 808 comprises washing the annulus and installing cement in the annulus and the bore of the well pipe, e.g. using annulus washing tool 100.

[0122] Figure 9 shows a system for optimizing a PA procedure. System 900 represents a collection of units, e.g. processes and / or hardware components, that can be utilised in isolation or in combinations not depicted in Figure 9. Specifically, Figure 9 shows a block diagram of an embodiment of a system 900 according to example embodiments of the present disclosure.

[0123] The connected nature of system 900 allows for continual improvement of planning, simulating, monitoring, validating, and executing PA procedures, resulting in optimised PA procedures. Although the reference numbers of system 900 start from unit 902, i.e. initial well parameters 902, system 900 can be initiated from any of the units depicted. Moreover, any unit or combination of units depicted in system 900 can be used in performing any of the methods described herein, such as method 1200 of Figures 12A-12C.

[0124] System 900 comprises the following units: initial well parameters 902; planning objectives 904; application data experience 906; diagnostics 908; initial risk assessment 910; simulations 912; operation plan (DOP) 914; risk assessment 916; execution 918; answer product 920; learnings 922; and application database 924. Simulations 912 comprises any of computation fluid dynamics (CFD) module 912-A, conveyance module 912-B, hydraulics module 912-C, or perforation / ablation module 912-D. Execution 918 comprises any of analysis module 918-A, sensor / data stitching module 918-B, road map module 918-C, or operation protocol module 918-D.

[0125] In one example, system 900 obtains initial well parameters 902, e.g. from application database 924, for use in determining planning objectives 904, wherein the planning objectives 904 are indicative of what needs to be accomplished within the well for a successful PA procedure. Application data experience 906 uses the planning objectives 904 and stored data from the application database 924 to identify previous experience in similar well scenarios. The previous experience in similar well scenarios is optionally used to modify planning objectives 904, e.g. in the event that completion of the planning objectives 904 was not successful in similar well scenarios. Diagnostics 908 applies pre-intervention log data, and initialrisk assessment 910 evaluates risk of the planned PA operation, based on the planning objectives 904 and application data experience 906. Simulations 912 calculates and / or performs simulations of the planned PA operation to confirm well intervention operational steps. Simulations 912 are based on at least calculations performed by conveyance module 912-B and hydraulics module 912-C, associated with conveyance simulation based on torque and drag and hydraulics calculations, respectively.

[0126] Calculations from simulations 912 are optionally used by application data experience 906, resulting in modification from diagnostics 908 and initial risk assessment 910 to optimise the planned PA procedure. Simulations 912 simulate the optimised planned PA procedure and, optionally, the looping process is repeated to further optimise the planned PA procedure. Alternatively, such as after a preset number of iterations, an exceeded quality threshold, or having received an automated and / or manual indication to proceed, the looping process is stopped. Operation plan (DOP) 914 determines a detailed operating plan for the planned PA procedure, such as operation plan (DOP) 1002 of Figure 10, which comprises one or more operating steps to be performed for the PA procedure to be completed. Risk assessment 916 evaluates risk associated with completing the operation plan (DOP) 914.

[0127] After the risk assessment 916, the PA procedure can be performed at execution 918. Execution 918 comprises analysis module 918-A, which performs execution analytics using human intelligence or artificial / computational intelligence (e.g. connected to 1012 interface to data analytics of Figure 10), and sensor / data stitching module 918-B, which is responsible for the collection and optional processing of data necessary for monitoring the PA procedure during execution of the PA procedure (e.g. shown as analytics system 1100 of Figure 11 ). The PA procedure is controlled, monitored, or performed by 918 execution according to either: road map 918-C, an operation road map based on human intelligence; or operation protocol 918-D, a protocol based on operating steps of operation plan (DOP) 914 determined using automation such as artificial intelligence, e.g. shown in Figure 10.

[0128] Using data associated with execution 918, answer product 920 combines diagnostics (e.g. from initial well parameters 902 and diagnostics 908), execution and verification (e.g. from execution 918) into an output, such as 1300 of Figure 13A or 13B. The output is indicative of the PA procedure from planning to execution to outcome, providing a comprehensive overview. Observations and teachings, such as successful steps and areas for potential improvements, e.g. where modifications to execution parameters could result in a more a successful PA procedure in the future, are identified by learnings 922, which can be stored in the application database 924. The application database 924 collects and / or stores all relevant data in a database, which is local, non-local, or cloud-based.

[0129] Example methods associated with using units of system 900, i.e. method 1200 of Figures 12A-C, method 1400 of Figure 14, method 1500 of Figure 15, method 1600 of Figure 16, method 1700 of Figure 17, and method 2000 of Figure 20, are described below.

[0130] Figure 10 illustrates an example implementation of system 900 used for monitoring a PA procedure. Specifically, Figure 10 shows a block diagram of an embodiment of a monitoring system 1000 according to example embodiments of the present disclosure.

[0131] Monitoring system 1000 comprises: operation plan (DOP) 1002; operational step 1004; operational instruction 1006; execution 1008; data acquisition 1010; interface to data analytics 1012; diagnostics and prognostics 1014; historical fault data and trends 1014-A; trained model 1014-B; conditions indicators 1014-C; step monitoring 1016; step status 1018; step action 1020; procedure monitoring 1022; operational status 1024; and action indicator 1026.

[0132] The operation plan (DOP) 1002 of monitoring system 1000, e.g. the detailed operating plan for a planned PA procedure determined by operation plan (DOP) 914 of Figure 9, comprises at least one operational step 1004 and corresponding operational instruction 1006 for performing or monitoring the operational step 1004. Operation plan (DOP) 1002 is a protocol or road map defining steps or milestones required to reach a target outcome, such as one or more requirements of a completed PA procedure, e.g. washing a well annulus or installing cement to set a barrier. In the example shown in Figure 10, there are multiple operationalsteps within the operation plan (DOP) 1002, including: first operational step 1004- 1 , second operational step 1004-2, third operational step 1004-3, fourth operational step 1004-4, and Nthoperational step 1004-N. Each operational instruction 1006 is associated with a set of parameters required for performing or monitoring each 1004 operational step. For example, operation plan (DOP) 1002 comprises: first set of parameters 1006-1 , second set of parameters 1006-2, third set of parameters 1006-3, fourth set of parameters 1006-4, and Nthset of parameters 1006-N. As an example, first operational step 1004-1 is to measure the depth of the annulus washing tool 100 of Figure 1 and the first set of parameters 1006-1 comprises any of time, pressure, flow, torque, wait on bit (WOB), acoustic signals, or another parameter.

[0133] The execution 1008 of the operation plan (DOP) 1002 comprises data acquisition 1010, where data associated with the corresponding set of parameters is collected e.g. from a sensor or from an application database such as application database 924 of system 900 in Figure 9. In the example of Figure 10, where operation plan (DOP) 1002 comprises N sets of parameters, execution 1008 comprises monitoring of the first set 1010-1 , monitoring of the second set 1010-2, monitoring of the third set 1010-3, monitoring of the fourth set 1010-4, and monitoring of the Nthset 1010-N. Returning to the above example where first operational step 1004-1 is to measure the depth of the annulus washing tool, monitoring of the first set 1010-1 (e.g. monitoring the first set of parameters 1006- 1 ) results in acquisition of a first dataset comprising the data required to determine the depth of the annulus washing. The first dataset is used for executing the first operational step 1004-1.

[0134] An interface to data analytics 1012 comprises or otherwise connects to diagnostics and prognostics 1014, step monitoring 1016, and / or to procedure monitoring 1022. Interface to data analytics 1012 is either a human interface for manual monitoring and / or intervention of the PA procedure being executed, or a computational interface for automated monitoring and / or optimisation of the PA procedure. For example, 1012 interface to data analytics is analytics system 1100 of Figure 11 .

[0135] Diagnostics and prognostics 1014 identifies potential issues or failure modes, such as previous issues, current issues, or future issues associated with an operational step 1004 or the PA procedure more broadly. Optionally, diagnostics and prognostics 1014 uses historical fault data and trends 1014-A, e.g. obtained from application database 924 of Figure 9. Alternatively, diagnostics and prognostics comprises a trained model 1014-B trained using historical fault data and trends 1014-A to identify conditions indicators 1014-C, where conditions indicators 1014-C are indicative of possible faults or issues. Trained model 1014- B comprises one or more mathematical models, such as any of the modules of simulations 912 of Figure 9, or one or more machine learning modules, such as a linear regression model or an artificial neural network. For example, trained model 1014-B comprises a generative artificial intelligence (Al) algorithm based on standard autoencoder methodology for identifying anomalies in monitoring of the first set 1010-1 or the associated dataset. In another example, trained model 1014- B comprises a convolutional neural network configured to classify monitoring of the first set 1010-1 or the associated dataset as successful, unsuccessful, or in need of further evaluation. Alternatively or additionally, trained model 1014-B is a machine learning module as described in more detail below.

[0136] Step monitoring 1016 comprises a step status 1018 and a corresponding step action 1020 for each operational step 1004 of the operation plan (DOP) 1002. For example, first step status 1018-1 is the status of the first operational step 1004- 1 and first step action 1020-1 is an indicator of whether an action needs to be performed based on the first step status 1018-1. Interface to data analytics 1012 or diagnostics and prognostics 1014 use monitoring of the first set 1010-1 and the first dataset to determine whether the first operation step 1004-1 is proceeding or has been completed as planned, e.g. “as planned”. If the first step status 1018-1 is “as planned”, first step action 1020-1 indicates that no action is required. If the first step status 1018-1 is not “as planned”, first step action 1020-1 indicates that an action is required, such as a corrective action. Optionally, interface to data analytics 1012 provides a suggested action to be initiated based on the first step status 1018-1 , the first operation step 1004-1 , and the first dataset from monitoring of the first set 1010-1. For example, the suggested action is determined usingdiagnostics and prognostics 1014 or obtained from an application database such as application database 924 of Figure 9.

[0137] In an example where operation plan (DOP) 1002 comprises a plurality of operational steps 1004, multiple operational steps can be performed in parallel or in series, and the step status 1018 is combined by procedure monitoring 1022 as operational status 1024. Similarly, procedure monitoring 1022 combines multiple step actions 1020 as an action indicator 1026. The operational status 1024 comprises each step status 1018 and / or a collective operational status, such as whether all step statuses are “as planned”, a number / fraction of step statuses that are “as planned” or not “as planned”, or an indicator of the probability of failure for either each step status or the step statuses collectively. The action indicator 1026 indicates whether an action is required and, optionally, one or more actions required, such as the most relevant action. The most relevant action is determined based on information obtained from the application database described above or using diagnostics and prognostics 1014, such as the action identified as having the greatest corrective consequence, or the action most likely to prevent or reduce the probability of failure.

[0138] Returning to the example where first operational step 1004-1 is to measure the depth of the annulus washing tool, monitoring of the first set 1010-1 is used by 1014 diagnostics and prognostics to identify no potential issues, as the first set of parameters 1006-1 required to measure the depth of the annulus washing tool wash acquired and the depth was likely calculated successfully. As the first dataset of monitoring of the first set 1010-1 comprises the data set out by the first set of parameters 1006-1 , first step status 1018-1 is “as planned” and first step action 1020-1 indicates that no action is required.

[0139] In a further example, first operational step 1004-1 is to measure washing flow rate at a first specified depth and second operational step 1004-2 is to measure washing flow rate at a second specified depth. The first set of parameters 1006-1 and the second set of parameters 1006-2 comprise acoustic signals and pressure at the first specified depth and the second specified depth, respectively. Execution 1008 comprises data acquisition from an acoustic sensor, such as acoustic sensor 115 of Figure 1 , to detect sounds originating from the flow ofwashing fluid, and a pressure sensor to detect pressure. Monitoring of the first set 1010-1 results in a first dataset comprising a first acoustic signal used to calculate a first washing flow rate and a first pressure measurement for the first specified depth, e.g. 1200 litres per minute (LPM) and 57 Bar at a depth of 2069 mMD from the surface of the well. Monitoring of the second set 1010-2 results in a second dataset comprising a second acoustic signal used to calculate a second washing flow rate and a second pressure measurement for the second specified depth, e.g. 700 litres per minute (LPM) and 49 Bar at a depth of 2089 mMD from the surface of the well.

[0140] Diagnostics and prognostics 1014 identifies, based on historical fault data and trends 1014-A, that the first washing flow rate and the first pressure reading are similar to a historical washing flow rate at the first specified depth for a historical PA procedure similar to the current planned and executed PA procedure. Additionally, the trained model 1014-B was trained on data associated with the historical PA procedure and classifies the first washing flow rate as “good” and identifies condition indicators 1014-C associated with a low probability of failure, e.g. the washing rate is as expected. Therefore, first step status 1018-1 is “as planned” and the first step action 1020-1 indicates that no action is required. Diagnostics and prognostics 1014 further identifies, based on historical fault data and trends 1014-A, that the second washing flow rate and the second pressure reading are not similar to a historical washing flow rate at the second specified depth for the historical PA procedure. The trained model 1014-B classifies the second washing flow rate as “poor” and identifies condition indicators 1014-C associated with a high probability of failure, e.g. the washing rate and pressure is lower than expected indicating that the wash cup such as wash cup 109 of Figure 1 is not performing optimally. Second step status 1018-2 is “not as planned” and the second step action 1020-2 indicates that an action is required.

[0141] Procedure monitoring 1022 uses the first step status 1018-1 and the second step status 1018-2 to determine operational status 1024, which is “not okay”. Based on second step action 1020-2, action indicator 1026 indicates an action is required. Diagnostics and prognostics 1014 identifies, e.g. using simulations 912 of Figure 9, historical data or a machine learning module, that the most relevant corrective actions are deploying a backup wash cap, such as backup wash cap109a of Figure 1 , or increasing the flow of washing fluid. Optionally, interface to data analytics 1012 is used to determine which action should be performed first, which is output to procedure monitoring 1022.

[0142] Figure 11 shows a flowchart illustrating an analytics system for use in monitoring or planning a PA procedure. Specifically, Figure 11 shows a block diagram of an embodiment of an analytics system 1100 according to example embodiments of the present disclosure.

[0143] Analytics system 1100 is an example implementation of execution 918 of system 900, and units of analytics system 1100 can be performed in combination or in place of any of the units described in relation to system 900 of Figure 9 and monitoring system 1000 of Figure 10. Analytics system 1100 is optionally directed towards acquiring and, optionally, processing sensor data, either collected from a sensor network or obtained from an application database, performing diagnostics and / or prognostics based on the acquired data and using the diagnostics and / or prognostics to determine an operation plan for performing the PA procedure or an operational status for monitoring the PA procedure.

[0144] Analytics system 1100 comprises: data acquisition 1102; sensor network 1104; processing and diagnostics 1106; pre-processing 1108; diagnostics and prognostics 1110; operation plan 1112; operational status 1114; state of operation 1116; and probable failure modes 1118.

[0145] Data acquisition 1102 acquires data from a sensor network 1104 or an application database, such as application database 924 of Figure 9. The sensor network 1104 comprises one or more sensors, such as surface sensors 904-A or downhole sensors 1104-B. Surface sensors 904-A comprise any of pressure, flowrate, weight, torque, or rotation sensors. Downhole sensors 1104-B comprise any of pressure, temperature, acoustic, or ultrasonic sensors. For example, data acquisition 1102 acquires data as part of execution 1008 of Figure 10 or as part of execution 918 of Figure 9.

[0146] After data is acquired at data acquisition 1102, the data is obtained by processing and diagnostics 1106. Optionally, the data is processed at preprocessing 1108, which comprises any of the following modules: data fusion 1108- A; data filtering 1108-B; or feature extraction 908-C. Data fusion 1108-A fusesand / or merges a plurality of data points or a plurality of parameters to combine relevant data, which results in increased integration and improved analysis across different data types and formats. Additionally, by merging data, the resulting data is potentially more feature-rich, resulting in improved extraction of patterns and correlations that may not be identifiable when analysing or processing unfused data. Data filtering 1108-B filters a plurality of data points or a plurality of parameters to help isolate relevant data from noise or otherwise irrelevant data. This improves efficiency and accuracy in extracting features, patterns, or correlations in the data. Feature extraction 1108-C extracts relevant features from the data, reducing the amount of data and hence computational resources required for analysing the data without losing vital information. It therefore also improves the efficiency and accuracy of both mathematical and machine learning modules that utilise the processed data. Optionally, pre-processing 908 is sensor / data stitching 918-B of Figure 9.

[0147] Analysis of the data is performed by diagnostics and prognostics 1110. Optionally, diagnostics and prognostics 1110 is analysis 918-A of Figure 9 and / or diagnostics and prognostics 1014 of Figure 10. Diagnostics and prognostics 1110 comprises any of the following units: machine learning module 1110-A, such a generative Al algorithm; historical fault data and trends 1110-B, e.g. historical fault data and trends 1014-A of Figure 10; trained model 1110-C, e.g. trained model 1014-B of Figure 10; and / or conditions indicators 1110-D,, e.g. conditions indicators 1014-B of Figure 10. The trained model 1010-C is configured to identify conditions indicators 1010-D based on historical fault data and trends 1110-B. Optionally, trained model 1010-C is machine learning module 1110-A, a separate additional machine learning module, or a different mathematical or computational model.

[0148] Operation plan 1112 comprises at least one of road map 1112-A or protocol 1112-B. The road map 1112-A is a detailed operation plan comprising a plurality of steps to be performed to complete the planned PA procedure, which is generated using diagnostics and prognostics 1110, e.g. road map 918-C of Figure 9. The road map 1112-A further comprises intelligence and knowhow from operators, obtained from an interface of processing and diagnostics 1106 or from the application database e.g. application database 924 of Figure 9. The protocol1112-B , e.g. operation protocol 918-D of Figure 9, is an alternate detailed operation plan further comprising intelligence an knowhow extracted from historical or simulated data via one or more machine learning modules. Machine learning modules, especially deep learning algorithms, extract features and trends from data that may be missed by manual or standard statistical methods. Beneficially, use of machine learning can provide e.g. optimised instructions or accurate parameters otherwise not easily or efficiently identifiable. The road map 1112-A represents an optimised operation plan 1112 based on manual expertise and operator intelligence while the protocol 1112-B represents an optimised operation plan 1112 based on data extrapolation and computer intelligence. Optionally, operation plan 1112 comprises one or more instructions from road map 1112-A and protocol 1112-B to form an operator-enriched operation plan optimised by machine learning.

[0149] Figures 12A-12C show a flowchart illustrating a circular method for optimising barrier setting and verification. Optionally, method 1200 of Figures 12A- 12C is a method for using system 900 of Figure 9 and / or utilises tools depicted in Figures 1-1. Method 1200 is representative of one potential utilisation, and any illustrated steps can be skipped, repeated, or added to where required to perform any of the actions described in relation to systems, apparatus, and methods described above and below.

[0150] Method 1200 comprises: step 1202 barrier setting and verification; step 1204 determining whether a control line is present; step 1206 diagnostics I; step 1208 tubing conveyed perforating (TCP) ; step 1210 install cement base; step 1212 diagnostics I; step 1214 diagnostics II; step 1216 ablation of control line; step 1218 diagnostics III; step 12 install cement base; step 12 identify material in the annulus; step 1224 completion fluids; step 1226 wash annulus I; step 1228 mud solids; step 1230 wash annulus II; step 1232 cemented; step 1234 break up cemented; step 1236 wash annulus III; step 1238 place cement plug in annulus and main bore; step 1240 determine whether to drill out and log; step 1242 dill out; step 1244 cement bond log (CBL)Zultrasonic log; step 1246 place cement plug; step 1248 verify cement plug; step 1250 output barrier answer product; and step 1252 select new barrier depth and replan.

[0151] Step 1202 comprises initiating barrier setting and verification, such as initiating an operation plan e.g. operation plan (DOP) 914 of Figure 9, operation plan (DOP) 1002 of Figure 10, or operation plan 1112 of Figure 11. Optionally, barrier setting and verification as described throughout method 1200 is execution 918 of Figure 9 or execution 1008 of Figure 10.

[0152] Step 1204 comprises determining whether there is a control line present. Determining whether there is a control line present is based on data obtained from a database, e.g. application database 924, that stores well parameters such as initial well parameters 902, or sensor data. For example, determining whether there is a control line present is based on data obtained from one or more downhole sensors, e.g. an ultrasonic sensor.

[0153] Step 1206, diagnostics I, is an optional step of performing a first set of diagnostic processes. The first set of diagnostic processes includes analysing data associated with whether a control line is present, such as cement bond logs, variable density logs, or ultrasonic data, and confirming that the control line is not present. Optionally, diagnostics I comprises confirming that the control line was previously extracted or otherwise ablated to a predetermined standard required for setting a secure barrier.

[0154] Step 1208 comprises performing or reviewing a previously performed TCP technique for the target well. For example, step 1208 comprises using a perforating tool to create holes (perforations) in tubing casing and formation, facilitating the flow of fluids or the placement of cement plugs for the PA procedure.

[0155] Step 1210 comprises initiating the installation of a cement base within the well using standard techniques. The cement base is designed to permanently seal the well and isolate it from surrounding geological formations. Optionally, step 1210 further comprises setting casing, e.g. a steel pipe, into the wellbore and pumping cement into the annulus to provide structural support and / or prevent fluid migration. Further optionally, step 1210 comprises placing one or more cement plugs, such as at specified intervals within the wellbore, to isolate different zones, support the casing, or prevent fluid movement.

[0156] Step 1212 is an optional step of performing a first set of diagnostic processes, as described in relation to step 1206 above. The first set of diagnosticprocesses includes analysing data associated with whether a control line is present, such as cement bond logs, variable density logs, or ultrasonic data, and optionally confirming that the control line is present.

[0157] Step 1214 comprises performing a second set of diagnostic processes for determining the orientation and / or location of the control line. For example, an ultrasonic sensor is used to collect ultrasonic image data, and computer imaging or other techniques are used to extract the orientation / location information.

[0158] Step 1216 comprises ablation of the control line, such as using perforation tools and methods described above.

[0159] Step 1218 comprises performing a third set of diagnostic processes for verifying ablation of the control line. For example, visual or ultrasonic image data is used to determine whether a control line is present when perforations are analysed. Alternatively or additionally, step 1212 or step 1214 is repeated to determine whether a control line is present.

[0160] Step 1220 comprises initiation the installation of a cement base, as disclosed in relation to step 1210.

[0161] Step 1222 comprises determining whether there is material in the annulus and, optionally, identifying the material in the annulus. Determination and / or identification is made using data obtained from the application database or sensor data obtained from one or more sensors of a network of sensors associated with the plug and abandonment procedure. Example sensors include surface sensors, downhole sensors and / or sensors associated with tools as described above.

[0162] Optionally, a machine learning module is used to determine and / or identify material, such as a classifier, e.g. a support vector machine, random forest, naive bayes, K-nearest neighbour, decision tree, logistical regression, gradient boosting machine, a neural network, or multiple combined algorithms. For example, a first machine learning module, e.g. a convolutional neural network, is trained using historical sensor data stored in the application database, such as application database 924 of Figure 9, and configured to classify whether there is material in the annulus from an input of sensor data from one or more downhole sensors. A second machine learning module, e.g. a gradient boosting machine, is trainedusing a variety of sensor data associated with different materials likely found in a well, including simulated sensor data. The second machine learning module is configured to identify the material determined to be in the well.

[0163] If it is determined at step 1222 that no material is in the annulus, or that the annulus does not otherwise require washing, method 1200 proceeds to step 1238. Otherwise, method 1200 proceeds to any of step 1224, step 1228, or step 1232 as required.

[0164] Step 1224 comprises identifying or verifying whether one or more completion fluids are within the annulus. Common types of completion fluids include clear brine fluids, inhibited zinc calcium bromide brines, water-based brines, oil-based completion fluid, glycol-based fluids, or polymer-based fluids. Optionally, step 1224 is performed at step 1222, such as based on data from an application database, sensor data, and / or use of a machine learning module.

[0165] Step 1226, wash annulus I, comprises washing the annulus. Preferably, washing the annulus is performed using acoustic detection of particle flow to confirm washing efficiency, as described above in relation to Figures 1-1. Optionally, initiation of step 1226 depends on step 1222 and / or step 1224, such that the washing procedure is optimised based on the material identified within the annulus. Beneficially, optimising the washing of the annulus based on the identified material, such as the material composition or quantity, increases the efficiency of the PA procedure, potentially decreases the resources required to perform the PA procedure, and helps increase the finished barrier quality.

[0166] Step 1228 comprises identifying or verifying whether mud solids are within the annulus. Mud solids, solid particles suspended in drilling mud, typically originate from the formation being drilled, the additives introduced into the mud, or the wear and tear of drilling equipment. Common types of mud solids include drill cuttings, formation solids, or additive solids. Optionally, step 1228 is performed at step 1222, such as based on data from an application database, sensor data, and / or use of a machine learning module, and is performed as an alternative or in addition to step 1224.

[0167] Step 1230, wash annulus II, comprises washing the annulus as described in relation to step 1226. For example, initiation of step 1230 depends on step 1222and / or step 1228, such that the washing procedure is optimised based on the material identified within the annulus. Moreover, washing mud solids helps prevent issues such as wellbore instability, equipment wear, and formation damage.

[0168] Step 1232 comprises identifying or verifying whether there are any cemented areas within the annulus. For example, a cemented area is a previously installed cement plug. Optionally, step 1232 is performed at step 1222, such as based on data from an application database, sensor data, and / or use of a machine learning module, and is performed as an alternative or in addition to step 1224 or step 1228.

[0169] Step 1234 comprises breaking up any cemented area detected in step 1222 or step 1232. Example methods include using manually or computationally operated tools such as pneumatic breakers, concrete saws, hydraulic bursting, or non-mechanical methods such as use of expansive demolition agents or chemical demolition.

[0170] Step 1236, wash annulus II, comprises washing the annulus as described in relation to step 1226. For example, initiation of step 1236 depends on step 1222 and / or step 1232, such that the washing procedure is optimised based on the material identified within the annulus.

[0171] Optionally, washing in step 1226, step 1230 or step 1236 comprises comparing an amount of particles that can be washed and lifted out of the annulus with the amount of particles in place before washing in initiated. This is based on sensor data, such as ultrasonic or acoustic signals, and is indicative of how washed the annulus volume is. For example, the indicator is a percentage, e.g. 80% cleaned annulus volume (80 particles lifted vs 100 particles in place = 80 / 100 = 0.8 = 80%). Alternatively or additionally, the size of the particles and / or a particle size distribution curve is analysed using the sensor data. Further optionally, acoustic sensors are paired with electronics configured to separate fluid flow from solids particles. Washing comprises detecting and discriminating between flow induced noise and particle impact noise, and the total amount of particles passing the sensor is subsequently calculated and / or computed.

[0172] Step 1238 comprises initiating the placing of one or more cement plugs in the annulus and / or main bore of the well. The cement plug permanently seals oneor more sections of the wellbore, preventing migration of fluids between formations. The placement of cement plugs comprises pumping cement, or a cement slurry, into the wellbore at a desired depth. The cement / slurry then sets and hardens, creating a solid barrier. Optionally, the composition of the cement / slurry is optimised to meet the requirements of the specific PA procedure, based on factors such as temperature, pressure, and wellbore conditions.

[0173] The design and placement of cement plugs play a crucial role in ensuring well integrity, safety, and environmental protection throughout the life cycle of an oil or gas well. For example, properly placed cement plugs contribute to the effective isolation of different zones within the well, preventing unwanted fluid movement and ensuring the overall integrity of the wellbore. By washing and verifying the washing of the annulus, especially if the washing is optimised based on the material detected in the annulus, the placement of a cement plug is improved, resulting in a more effective and higher quality seal.

[0174] Optionally, step 1238 comprises any of steps 1240-1248.

[0175] Step 1240 comprises determining whether to drill out and obtain a log, such as a cement bond log (CBL) or ultrasonic log. If it is determined not to drill out and obtain a log, such as if the cement plug is already in place or if a cement plug is not required at this annulus or specific depth range, method 1200 proceeds to step 1248.

[0176] Step 1242 comprises drilling out a previous cement plug, such as a temporary cement plug or the cement of step 1232 or step 1238, to reopen the wellbore for further operations. The drill out process involves using drilling equipment to break through or remove the cement barrier, allowing access to the wellbore beyond the plug.

[0177] Step 1244 comprises obtaining a log, such as a CBL or ultrasonic log. Measurement while drilling, logging while drilling, or other typical methods for obtaining a log are performed. The log is indicative of formations and / or conditions during or after the drill out process and / or before placement of a cement plug. For example, the log is obtained using borehole imaging tools, such as an ultrasonic sensor, to capture images of the wellbore wall associated with geological features and conditions.

[0178] Step 1246 comprises placing a cement plug, as described in relation to step 1238. For example, a single cement plug is placed at a target depth range at step 1238. The single cement plug is drilled out at step 1242 for a detailed log to be obtained associated with the quality of the single cement plug across the target depth range at step 1244. An additional cement plug is placed to stabilise and seal the single cement plug after the drill out procedure at step 1246. In another example, a first cement plug is placed at a first target depth range at step 1238. A second cement plug is placed at step 1246 at a second target depth range to help zone and secure the well. In a further example, only one cement plug is placed at either step 1238 or step 1246.

[0179] Step 1248 comprises verifying whether the cement plug is placed as required to permanently seal the well. For example, verification is based on pressure tests or data obtained from one or more acoustic sensors. Pressure testing comprises obtaining pressure measurements to assess the integrity of the wellbore, casing, and cement barriers. Acoustic sensors are used to obtain an acoustic track over the target depth range to determine whether there is a leak present in the cement plug. If the cement plug is successfully verified such that no leak is present and the seal is secure, method 1200 proceeds to step 1250. If a leak is detected, method 1200 proceeds to step 1252.

[0180] Step 1250 comprises outputting a barrier answer product, such as output 1300 of Figure 13A and / or the barrier quality value of method 2010. The barrier answer product optionally comprises at least one barrier quality value and / or a displayable barrier image relative to the target depth of the placed barrier and verification of step 1248.

[0181] Step 1252 comprises selecting a new barrier depth and replan. The new barrier depth is determined based on initial well parameters, sensor data, and / or analysis of data used to verify the cement plug in step 1248, and is selected such that a new barrier seals the well e.g. from any leaks detected at step 1248. Further planning of the setting of a new barrier is optionally based on using a pervious operation plan, repurposing or adjusting a previous operation plan, or generating a new operation plan, such as operation plan 1112 of Figure 11 , 1002 of Figure 10, or 914 of Figure 9 above. After a new barrier depth is determined, method 1200proceeds to step 1204, at which points any of the steps 1204 - 1248 are repeated as required.

[0182] Optionally, method 1200 is performed for multiple annuli, such that any of steps 1204 - 1252 are repeated where necessary. For example, at step 1222, it is determined whether there is material present in each annuli and, if present, the material is determined. Based on the material present, any of steps 1224 - 1236 are repeated, and a cement plug is placed in the annuli at step 1238.

[0183] Figure 13A illustrates an example confirmation and verification output from a PA procedure performed using any the methods, apparatus, or systems described herein. Specifically, Figure 13 shows a block diagram of an embodiment of an output 1300 according to example embodiments of the present disclosure.

[0184] Output 1300 is comprises relevant data for assessing and verifying the complete PA procedure, including steps taken in planning, executing, verifying execution, and results of the PA procedure. Output 1300 is therefore suitable for determining the success of a PA procedure and used to optimise further PA procedures, such as through modification to future planning or execution based on the results of previous PA procedures. Beneficially, when performing multiple PA procedures, all relevant data for each PA procedure is accessible for evaluation and comparison, such that the success of larger PA operations can be efficiently determined. By having the most relevant data for each stage of the PA procedure in a machine-readable and / or displayable format, success of the PA procedure is efficiently verified, preventing the need for additional resource-heavy verification operations. Additionally, multiple additional barriers are not required to securely seal a well, or unnecessary large plugs, as the quality of each barrier can be readily evaluated to determine whether the well is successfully sealed. Furthermore, targeted monitoring based on the success of the PA procedure, e.g. associated with the likelihood of leaks, reduces levels of necessary monitoring of the well over time, further increasing efficiency.

[0185] Output 1300 comprises: diagnostic section 1302; execution section 1304; verification section 1306; results section 1308; depth range 1310; a gamma ray track 1312; a cementing track 1314; an acoustic track 1316; a barrier length image 1318; a first barrier quality flag 1320; and a second barrier quality flag 1322.

[0186] Diagnostics section 1302 is associated with information and data used in planning the PA procedure, and comprises a set of one or more diagnostic parameters. One diagnostic parameter is gamma ray track 1312, which is a casing collar locator track over the depth range 1310. The depth range 1310 optionally encompasses a target depth range for optimal placing of a cement barrier, such as a range between depth A and depth B, and the same depth range 1310 is used for all parameters in diagnostics section 1302, and optionally all sections of the output 1300, for easy tracking and comparison between different parameters.

[0187] Diagnostics section 1302 comprises either current or previous data, with data acquired from sensors associated with the planned PA operation or obtained from an application database associated with storing data of previous PA operations. Diagnostic parameters are indicative of information for depth control inside tubing, including tubular condition, bonding and material in the annulus. Diagnostic parameters comprise any of: a gamma ray track 1312, e.g. a casing collar locator track, for depth control; a cement bond log (CBL) for defining a wash indicator (Wl); a variable density log (VDL), or VDL image; an ultrasonic or multifinger log to determine inner casing corrosion or deformation; control line and clamp location / orientation data or image depending on the presence of one or more control lines; or a combination thereof.

[0188] A Wl is indicative of the likelihood of successfully washing an annulus section. Wl is a CBL value, which is typically scaled between 0 and 100 mV into categories such as good, medium and bad sections. A CBL value larger than 15mV typically indicates a good washability, a CBL value between 10 and 15m V indicates a medium washability and a CBL value less than 10mV typically indicates a bad washability, and therefore requires further considerations. When the Wl is good, a further risk assessment is not required, saving time and resources without compromising the planned PA procedure. The CBL value can be converted into Bl (Bond Index to normalize). Optionally, the Wl is based on CBL value and / or CBL index, an ultrasonic log or mass flow rate / permeability, or any other suitable methods.

[0189] The execution section 1304 is associated with the execution, delivery and quality control of the planned PA procedure. Performing a PA procedure is basedon step-by-step execution of protocols or road maps of a PA operation, such as described in relation to Figure 10. For example, operation steps comprise control line ablation and perforation if control lines are present, washing of annuli, and placement of cement. The operational plan divides planned sections into incremental steps to verify and log planned versus actual parameters.

[0190] Execution section 1304 comprises a set of one or more procedure parameters, such as cementing track 1314 for monitoring cementing flow rate and pressure versus predefined optimal parameters / value ranges. Another procedure parameter is a perforation-ablation track for confirmation of ablation depths and calculation of actual ablated length (LAA). LAA is the confirmed ablated length [m] over total perforated length [m], either defined by length or by number of ablation points. Another procedure parameter is a washing pressure track. The washing pressure and flowrates are to stay between the set maximum flow pressure and the minimum flow pressure to ensure good washing efficiency. Optionally, planned or optimised pressure and flowrate ranges are set in the road map / protocol, and are compared to measured pressure and flowrate. Another procedure parameter is a washing particles track e.g. based on acoustic signals. Using acoustic particle flow detection results in efficient and accurate monitoring of the successful lifting of particles from annuli, e.g. estimated particle mass lifted versus originally in place.

[0191] Optionally, execution section 1304 is associated with an Execution Quality Flag (EOF) calculated from the incremental operational steps that build up a barrier required for successful completion of the PA procedure. If one or more procedure parameters associated with monitoring and executing the PA procedure are within an expected range for the specific incremental interval, then the partial-EQF for the step, e.g. E1 , is equal to a predefined value, e.g. 1 . The final EOF is an average of the partial-EQFs for the entire operation procedure. For example, the EOF is calculated by performing a mean average calculation partial-EQF. In an example where execution section 1304 comprises a perforation-ablation track (A), a washing pressure track (VF), a washing particles track (P), and a cementing track (C), the EQF ( / ) is calculated as:

[0193] where a, p, y, and 8 are normalisation factors tailored to the planned or optimised ranges set in the operational plan of the PA procedure, and where n is the total number of depth sections i across depth range 1310.

[0194] Verification section 1306 comprises acoustic track 1316, a verification parameter for verifying whether a barrier placed during execution of the PA procedure is successful or unsuccessful, e.g. “pass” where no leak is detected or “no pass” where a leak is detected. Additionally or alternatively, verification section 1306 comprises any of the following verification parameters: a CBL / ultrasonic log or associated interpretation or ultrasonic data; a pressure or temperature sensor track, where the pressure and / or temperature sensor is below or above the barrier; or a barrier gas permeability test track.

[0195] Results section 1308 comprises a barrier length image 1318, associated with the actual length of the placed barrier and verification that there is no leak detected at depth sections throughout the barrier, e.g. based on one or more verification parameters of verification section 1306. The barrier length image 1318, in this example, comprises a first barrier quality flag 1320 and a second barrier quality flag 1322. The first barrier quality flag 1320 is associated with a poorer quality, e.g. “not good”, as there is acoustic data indicative of a poorer quality barrier at associated depth range C. The second barrier quality flag 1322 is associated with a higher quality, e.g. “good”, as there is no acoustic data indicative of a poorer quality barrier. The verification parameters of verification section 1306 indicate that a leak present in the depth range D is very unlikely, as no leak was detected.

[0196] A barrier quality value is a classification or a numerical value, and is calculated based on the quality flags of the barrier length image 1318. For example, a modal average of the quality flag for each depth section, e.g. every meter, results in a barrier quality value associated with the second barrier quality flag 1322, e.g. “good”. In another example, where “good” is equal to 1 and “not good” is equal to 0, a mean average results in a barrier quality value of 0.8, which is optionally further classified, e.g. as “good”. Optionally, results section 1308 comprises a barrier quality image indicative of the barrier quality value for the barrier as a whole. Further optionally, the barrier length image 1318 or barrierquality image of results section 1108 is generated based on combining the execution section 1304 with the verification section 1308. For example, the EQF is of the execution section 1304 is multiplied with whether a leak is detected based on verification parameters of the verification section 1308.

[0197] Figure 14 is a flowchart illustrating a method for controlling a PA procedure. Method 1400 is optionally a method for utilising system 900 of Figure 9, or any of the systems or apparatus described herein, and any of the steps of method 1400 are optionally performed in isolation, in addition, or as a replacement to any of the method steps described herein.

[0198] Method 1400 comprises: step 1402, obtain a set of diagnostic parameters; step 1404, determine operational steps; step 1406 perform a simulation of the operational steps; step 1408, generate an operation plan for a PA procedure; step 1410, monitor the PA procedure to determine an operation status; step 1412 determine a barrier quality value to validate the PA procedure; and step 1414, store data associated with the PA procedure.

[0199] Step 1402 comprises obtaining a set of diagnostic parameters from a network of sensors or an application database. The network of sensors optionally comprises one or more surface sensors or one or more downhole sensors, which are configured to measure any of pressure, flow rate, acoustic data, or ultrasonic data. The set of diagnostic parameters comprises initial well parameters, cement bond logs, or variable density logs over a target depth range.

[0200] For example, step 1402 comprises obtaining initial well parameters 902 from application database 924. The set of diagnostic parameters are associated with information about the well and / or the condition of the well where the PA procedure is to take place, e.g. data obtained during previous operation of the well before the PA procedure was determined or initiated. Further example parameters include hydraulics parameters and conveyance parameters for performing simulations at step 1406.

[0201] Optionally, the set of diagnostic parameters are processed once obtained, as described in relation to pre-processing 1108 of Figure 11 . For example: two or more parameters are merged or otherwise fused; one or more parameters arefiltered; or one or more features are extracted from one or more parameters, such as a target feature.

[0202] Step 1404 comprises using the set of diagnostic parameters to plan the PA procedure thereby determining one or more operational steps required to perform the planned PA procedure, e.g. planning objectives 904.

[0203] Step 1406 comprises performing a simulation on the one or more operational steps determined at step 1404, e.g. using simulation 912. Performing the simulation comprises performing hydraulics calculations, e.g. using hydraulics 912-C, and estimating downhole forces, e.g. using conveyance 912-B, based on the set of diagnostic parameters obtained at step 1402. The simulation is optionally initiated or repeated as needed from a data analytics interface, such as if any of the operational steps are altered or the set of diagnostic parameters is updated.

[0204] Optionally, performing the simulation comprises generating a hydraulics module configured to perform hydraulics calculations comprising ball drop calculations, pressure drop calculations, volume calculations, or calculations associated with cementing the well. Alternatively or additionally, step 1406 further comprises: generating a conveyance module configured to estimate downhole forces based on torque and drag calculations; generating a computational fluid dynamics (CFD) module configured to perform further hydraulics calculations comprising any of: cement slumping calculations, washing efficiency calculations, listing of particles calculations, multiple annuli calculations, or calculations associated with tubular, tubing, or casing designs; or generating a perforation ablation module configured to verify correct perforation and ablation or determine mechanical perforation and ablation parameters based on geometry of perforations, perforation depth, or perforation gun setup. Any one or combination of modules can then be used to simulate one or more operational steps. In an alternative example, the simulations are performed using a machine learning module or pre-generated mathematical, numerical, or computational modules.

[0205] Step 1408 comprises analysing the simulation to generate an operation plan comprising the one or more operational steps, e.g. operation plan (DOP) 914. The one or more operational steps are optionally modified based on the simulations performed at step 1406 to optimise the planned PA procedure. Optionally, a riskassessment is performed based on the generated operation plan, e.g. risk assessment 916, which is based on performing further simulations, operation expertise, computational or machine learning modelling, or other typical risk assessment techniques. For example, faults or risks associated with either the PA procedure or the well itself are identified based on the simulation of step 1406.

[0206] Step 1410 comprises monitoring the PA procedure once the PA procedure is initiated. Data collected from the network of sensors is used to determine an operation status of the PA procedure, e.g. using monitoring system 1000. The network of sensors comprises an ultrasonic imaging tool, e.g. for verifying whether a control line has been ablated, and an acoustic sensor, e.g. for verifying whether the annulus has been washed efficiently or whether a leak is present once a barrier has been set.

[0207] Optionally, monitoring the PA procedure comprises using a machine learning module, e.g. machine learning module 2138, to predict and estimated state of the PA procedure based on the set of diagnostic parameters obtained at step 1402. The machine learning module is trained with historical parameters or sensor data associated with historical PA procedures, or simulated parameters and sensor data associated with simulated PA procedures, or a combination thereof. Such training data is obtained from the application database.

[0208] Further optionally, a set of execution parameters, e.g. procedure parameters, measured or otherwise collected from the network of sensors while the PA procedure is being performed, is obtained during monitoring of the PA procedure. A current state of the PA procedure can then be generated based on the set of execution parameters, and an operation status determined based on a comparison between the estimated state and the current state.

[0209] Step 1412 comprises validating the PA procedure based on determining a barrier quality value associated with whether there is a leak in the barrier, e.g. as described in relation to output 1300.

[0210] Step 1414 comprises storing, in the application database, data associated with execution of the PA procedure, such as the operation plan, sensor data, and the barrier quality value. In the example where a machine learning module is trained using data obtained from the application database, the machine learningmodule is optionally retrained or otherwise updated using the new data stored at step 1414 to improve the machine learning module.

[0211] Figure 15 is a flowchart illustrating a method for simulating a PA procedure. Method 1500 is optionally a method for utilising system 900 of Figure 9, or any of the systems or apparatus described herein, and any of the steps of method 1500 are optionally performed in isolation, in addition, or as a replacement to any of the method steps described herein.

[0212] Method 1500 comprises: step 1502, obtain a set of diagnostic parameters; step 1504, generate a hydraulics module; step 1506, generate a conveyance module; step 1508, optionally generate a computational fluid dynamics module; step 1510, optionally generate a perforation ablation module; step 1512, simulate operational steps of a planned PA procedure using the generated modules; and step 1514 generate or verify an operation plan.

[0213] Step 1502 comprises obtaining, from the application database, a set of diagnostic parameters. The set of diagnostic parameters comprises hydraulics parameters and conveyance parameters associated with a planned PA procedure. Optionally, hydraulics parameters and conveyance parameters are obtained from one or more sensors associated with the well, e.g. one or more surface sensors or downhole sensors configured to measure pressure, flowrate, or acoustic / ultrasonic data.

[0214] Step 1504 comprises generating or otherwise obtaining a hydraulics module configured to perform hydraulics calculations. Hydraulics calculations comprise ball drop calculations, pressure drop calculations, volume calculations, calculations associated with cementing the well, or some combination thereof.

[0215] Step 1506 comprises generating or otherwise obtaining a conveyance module configured to estimate downhole forces based on torque and drag calculations.

[0216] Step 1508 is optional and comprises generating or otherwise obtaining a computational fluid dynamics (CFD) module configured to perform further hydraulics calculations. Further hydraulics calculations comprise any of: cement slumping calculations, washing efficiency calculations, listing of particlescalculations, multiple annuli calculations, or calculations associated with tubular, tubing, or casing designs

[0217] Step 1510 is optional and comprises generating or otherwise obtaining a perforation ablation module configured to verify correct perforation and ablation. Alternatively or additionally, the perforation ablation module is configured determine mechanical perforation and ablation parameters based on geometry of perforations, perforation depth, or perforation gun setup.

[0218] Step 1512 comprises using the hydraulics module and the conveyance module, and optionally the CFD and / or perforation ablation module, to simulate one or more operational steps of the planned PA procedure based on the set of diagnostic parameters.

[0219] Step 1514 comprises generating an operation plan comprising operational steps used for planning the PA procedure. If the operation plan is already generated, step 1514 comprises verifying the generated operation plan or modifying the operation plan to optimise the planned PA procedure based on the simulation(s). The operation plan can then be compared to a monitored PA procedure to determine an operational status of the monitored PA procedure.

[0220] Optionally, step 1514 comprises generating, verifying, validating, or modifying an operational envelope comprising a range of operation parameters for performing the PA procedure optimally. Further optionally, step 1514 comprises identifying faults or risks in the well or PA procedure based on simulation of the one or more operational steps

[0221] In one example, a machine learning module comprises at least the hydraulics module and the conveyance module such that the machine learning module simulates operation of the PA procedure. For example, the machine learning module comprises a generative artificial intelligence (Al) algorithm configured to generate the operation plan.

[0222] In another example, a data analytics interface initiates simulation of the one or more operational steps, and is optionally configured to display the simulation of the one or more operational steps and / or store simulation of the one or more operational steps associated with the set of diagnostic parameters in theapplication database. Simulations and associated data can then be used for planning future PA procedures and / or training / updating machine learning modules described herein.

[0223] Figure 16 is a flowchart illustrating a method for monitoring a PA procedure. Method 1600 is optionally a method for utilising monitoring system 1000 of Figure 10, or any of the systems or apparatus described herein, and any of the steps of method 1600 are optionally performed in isolation, in addition, or as a replacement to any of the method steps described herein.

[0224] Method 1600 comprises: step 1602, obtain a first set of parameters; step 1604, generate processed parameters; step 1606, predict an estimated state; step 1608, obtain a second set of parameters; step 1610, generate a current state; and step 1612, determine an operation status.

[0225] Step 1602 comprises obtaining, from a network of sensors comprising at least one acoustic sensor, a first set of parameters associated with the well.

[0226] Step 1604 comprises generating one or more processed parameters by performing one or more processing steps on each parameter of the first set of parameters. The processing steps comprise: step 1604-A, merging two or more parameters of the first set of parameters, resulting in data fusion; step 1604-B, filtering one or more parameters of the first set of parameters; and / or step 1604-C, extracting a target feature from one or more parameters of the first set of parameters.

[0227] Fusing and / or merging a plurality of data points or a plurality of parameters to combines relevant data, resulting in increased integration and improved analysis across different data types and formats. Additionally, by merging data, the resulting data is potentially more feature-rich, resulting in improved extraction of patterns and correlations that may not be identifiable when analysing or processing unfused data. Filtering a plurality of data points or a plurality of parameters helps isolate relevant data from noise or otherwise irrelevant data. This improves efficiency and accuracy in extracting features, patterns, or correlations in the data. Feature extraction reduces the amount of data and hence computational resources required for analysing the data without losing vital information. It therefore alsoimproves the efficiency and accuracy of both mathematical and machine learning modules that utilise the processed data of step 1604.

[0228] Step 1606 comprises predicting an estimated state of the PA procedure based on the one or more processed parameters. Optionally, an operation plan is generated based on the estimated state of the PA procedure, where the operation plan comprises operational steps used for planning the PA procedure and planning the PA procedure can include evaluating risk prior to initiation of the PA procedure, such as using simulation methods described above. The operation plan optionally includes a probability of one or more failure modes occurring.

[0229] Step 1608 comprises obtaining, from the network of sensors, a second set of parameters associated with monitoring the well after the PA procedure is initiated. The second set of parameters comprises one or more acoustic signals obtained from an acoustic sensor.

[0230] Step 1610 comprises generating a current state of the PA procedure based on the second set of parameters.

[0231] Step 1612 comprises determining an operation status based on comparing the estimated state of the PA procedure to the current state of the PA procedure.

[0232] Optionally, any of the steps of method 1600 are performed using a machine learning module, e.g. trained with historical parameters associated with historical PA procedures stored in an application database. Storing the current state of the PA procedure and / or the second set of parameters in the application database allows for the machine learning module to be retrained or updated in the future to optimise future iterations of method 1600. For example, the current state of the PA procedure or the second set of parameters stored in the application database is used in predicting an estimated state of a subsequent PA procedure.

[0233] Figure 17 is a flowchart illustrating a method for diagnosing a PA procedure. Method 1700 is optionally a method for utilising system 2100 of Figure 21 , or any of the systems or apparatus described herein, and any of the steps of method 1700 are optionally performed in isolation, in addition, or as a replacement to any of the method steps described herein.

[0234] Method 1700 comprises: step 1702, obtain sensor data from a network of sensors; step 1704, obtain historical data from an application database; step 1706, extract relevant features from the sensor data; step 1708, identify conditions indicators based on the relevant features; step 1710, determine a predicted state; step 1712, compare the predicted state to a current state; and step 1714, predict whether a failure mode is probable.

[0235] Step 1702 comprises obtaining sensor data from a network of sensors, the network of sensors comprising one or more surface sensors and one or more downhole sensors.

[0236] Step 1704 comprises obtaining historical data from an application database, the historical data comprising historical fault data or trend data associated with previous PA operations.

[0237] Step 1706 comprises extracting one or more relevant features from the sensor data. What is considered a relevant feature is determined based on the sensor data, such as from data (pre-)processing or machine learning based feature extraction. Alternatively, or additionally, relevant features are dependent on the sensor data, such as what the sensor data is, the sensor types used to collect it, or the type, form or shape of the data.

[0238] Step 1708 comprises using an ANN trained using the historical data obtained at step 1704 to identify one or more conditions indicators. The one or more relevant features extracted at step 1706 are used by the ANN to identify the one or more conditions indicators and, optionally, the ANN performs both step 1706 and step 1708, such that relevant features are determined by the ANN during training without requiring input from external programmers or operators. The one or more conditions indicators are associated with performing the PA operation, and the condition of either planned / executed steps of the PA operation or components of the well.

[0239] Step 1710 comprises determining a predicted state of the PA operation based on the one or more conditions indicators. Optionally, the predicted state of the PA operation is also generated by the ANN or a separate ANN also trained using historical data obtained at step 1704. For example, the ANN is input thesensor data at step 1702 and the ANN generates an output of conditions indicators and / or a predicted state of the PA operation.

[0240] Step 1712 comprises predicting whether a failure mode is probable based on comparing the predicted state of the PA operation to a current state of the PA operation. The current state of the PA operation based on either a protocol for the PA operation, the protocol comprising a plurality of planned operational steps or instructions to perform the PA operation, or updated sensor data from the network of sensors, e.g. step 17 is repeated. Optionally, the probability of one or more failure modes is predicted using the ANN or a separate ANN also trained using historical data obtained at step 1704. For example, the ANN is input the sensor data at step 1702 and the ANN predicts whether a failure mode is probable or, e.g., the probability of multiple predetermined failure modes.

[0241] Any of the above referenced ANNs are trained using standard machine learning training methods. For example, the ANN is configured to perform the following training steps until a stop criterion is met: output one or more predicted conditions indicators based on an input of the one or more historical relevant features; calculate, based on a loss function, a difference between the one or more predicted conditions indicators and the one or more historical conditions indicators; and based on the difference, modify hyperparameters of the ANN to reduce the difference. Alternatively, simulated data is used instead of or in addition to historical data.

[0242] Optionally, the stop criterion is based on any of: a number of completed epochs during training of the ANN; the difference meeting or exceeding a stopping threshold; a comparison between the protocol and a historical protocol or a simulated protocol, wherein the comparison is indicative of whether the protocol is suitable; a comparison between the predicted state of the PA operation and a historical state or a simulated state, wherein the comparison is indicative of whether the predicted state is accurate; a comparison between whether the failure mode is probable and the historical data; or any combination thereof.

[0243] Figure 18 and Figure 19 are flowcharts illustrating methods for monitoring a PA procedure, where the PA procedure comprises ablation of a control line. Method 1800 or method 1900 is optionally a method for utilising monitoring system900 of Figure 9, or any of the systems or apparatus described herein, and any of the steps of method 1800 or method 1900 are optionally performed in isolation, in addition, or as a replacement to any of the method steps described herein.

[0244] Method 1800 comprises: step 1802, determine a control line is present; step 1804, obtain a first set of diagnostic parameters; step 1806, determine a position of the control line; step 1808, obtain a second set of diagnostic parameters; step 1810, verify ablation of the control line; and step 1812, initiate installation of cement.

[0245] Step 1802 comprises determining that a control line is present in the well based on data obtained from the application database. If no control line is present, method 1800 optionally proceeds to step 1810. If a control line is determined to be present, such as based on logs associated with historical operation of the well or sensor data indicative of the presence of the control line, method 1800 proceeds to step 1804.

[0246] Step 1804 comprises obtaining a first set of diagnostic parameters before ablation of the control line is initiated. The first set of diagnostic parameters comprises sensor data, or processed sensor data, collected from one or more sensors. For example, the first set of diagnostic parameters comprises initial well parameters, CBL or VBL over a target depth range of the well, or ultrasonic sensor data e.g. ultrasonic images. The first set of diagnostic parameters is obtained from the application database, sensors, or an intermediary interface or system. The one or more sensors optionally comprise or are comprised within a network of sensors, where the network of sensors comprises at least one surface or downhole sensor as described above.

[0247] Step 1806 comprises determining a position of the control line based on the first set of diagnostic parameters. The position is indicative of the location of the control line, the orientation of the control line, or a combination of both location and orientation of the control line. The position is determined relative to the annulus, the horizon, the tool, a boundary or point within of the target depth, or another suitable reference point.

[0248] Step 1808 comprises obtaining a second set of diagnostic parameters comprising data collected from the one or more sensors after perforation of thecontrol line is initiated. Optionally, step 1808 first comprises initiation of perforation of the control line and, further optionally, performing perforation of the control line, e.g. by using the perforation tool described above. The second set of diagnostic parameters comprises one or more ultrasonic signals, such as ultrasonic images, indicative of where the control line is perforated and / or whether the control line is perforated.

[0249] Step 1810 comprises verifying ablation of the control line by using the second set of diagnostic parameters obtained in step 1808. For example, sensor data collected over an ablated length associated with the target depth range is used to verify ablation of the control line over the ablated length. Preferably, when installation of a cement plug is planned or otherwise required after ablation of the control line, the ablated length is at least the length of a cement plug.

[0250] Beneficially, verification of the ablation of the control line results in improved efficiency and a reduction in required resources when completing PA procedures. For example, an ablated control line can result in a more secure cement plug. Therefore, verification before installation of the cement plug prevents e.g. leaks or other imperfections from being detected only after the cement plug has been placed, which would require an additional plug, seal, or barrier to rectify (essentially a repeat of the whole PA procedure at a separate depth) or the removal and replacement of the inadequate cement plug. While operators may try and compensate for such scenarios by increasing the length of the cement plug, thereby lowering the likelihood of leaks, this requires additional resources that are not required when ablation of the control line is verified.

[0251] If ablation of the control line is complete and / or verified, method 1800 proceeds to step 1812.

[0252] Step 1812 comprises initiating installation of cement within the annulus for setting the permanent plug.

[0253] Method 1900 comprises: step 1902, determine a control line is present, e.g. step 1802; step 1904, obtain ultrasonic images from an ultrasonic imaging tool; step 1906, verify ablation of the control line based on the ultrasonic images; step 1908, obtain acoustic signals from an acoustic sensor; step 1910, confirm washingof the well is completed; and step 1912 obtain and store a set of verification parameters.

[0254] Step 1902 comprises determining that a control line is present in the well based a cement bond log or an ultrasonic log, e.g. step 1802 of method 1800. The cement bond log or the ultrasonic log obtained is obtained from an application database and / or based on data collected from one or more sensors.

[0255] If no control line is present, method 1900 optionally proceeds to step 1906. If a control line is determined to be present, such as based on logs associated with historical operation of the well or sensor data indicative of the presence of the control line, method 1900 proceeds to step 1904. Optionally, before method 1900 proceeds to step 1904, ablation of the control line is initiated and / or completed. Ablating the control line optionally comprises perforating a casing, e.g. by using a perforation tool.

[0256] Step 1904 comprises obtaining one or more ultrasonic images from an ultrasonic imaging tool within the well. Optionally, the ultrasonic imaging tool is an ultrasonic sensor attached to a tool such as a perforation tool, drilling tool, or other tool. Further optionally, the ultrasonic images are indicative of whether the casing has been perforated, such that any perforations in the casing are visible in the ultrasonic images. Alternatively, the ultrasonic images can detect the presence and / or position of a control line through the casing, such that visualisation through perforations is not required.

[0257] Step 1906 comprises verifying ablation of the control line over an ablated length based on the ultrasonic images associated perforations of the control line. For example, ultrasonic sensor data collected over an ablated length associated with the target depth range is used to verify ablation of the control line over the ablated length. Preferably, when installation of a cement plug is planned or otherwise required after ablation of the control line, the ablated length is at least the length of a cement plug.

[0258] If perforations in the casing are present, perforations allow a control line to be detected visibly, e.g. if the control line has not been ablated. For example, a position of the control line is determined using data obtained from one or more sensors or the application database, as per step 1806 of method 1800. If a controlline is not detected at that position through one or more perforations in the casing, the control line has been ablated.

[0259] Alternatively, ultrasonic images are used without examining perforations and / or comparisons to a known position of the control line. For example, ultrasonic images of a section of the well, e.g. over the ablated length, are examined for any indication of the control line. If data indicative of the presence of the control line over the section is detected, the control line has not been ablated. If data indicative of the presence of the control line over the section is not detected, the control line has been ablated.

[0260] Beneficially, as described above, verification of the ablation of the control line results in improved efficiency and a reduction in required resources when completing PA procedures. For example, an ablated control line can result in a more secure cement plug. Therefore, verification before installation of the cement plug prevents e.g. leaks or other imperfections from being detected only after the cement plug has been placed, which would require an additional plug, seal, or barrier to rectify (essentially a repeat of the whole PA procedure at a separate depth) or the removal and replacement of the inadequate cement plug. While operators may try and compensate for such scenarios by increasing the length of the cement plug, thereby lowering the likelihood of leaks, this requires additional resources that are not required when ablation of the control line is verified.

[0261] If ablation of the control line is complete and / or verified, method 1900 proceeds to step 1908. Alternatively, any of steps 1902 - 1906 are repeated, method 1900 is suspended, e.g. to examine ablation / perforation tools used to ablate control lines, or method 1900 proceeds without the control line ablated. Preferably, if method 1900 proceeds without an ablated control line verified, steps 1908 - 1912 are modified to compensate for presence of a control line. For example, a change in acoustic signals based associated with a change in washing flow rate may differ, and the efficiency of washing the annulus, based on the presence of the control line.

[0262] Step 1908 comprises obtaining one or more acoustic signals from an acoustic sensor after washing of the annulus is initiated. Optionally, step 1808 also comprises initiating and / or completing one or more steps required for washing ofthe annulus to be performed. The one or more acoustic signals are obtained at a first time point and a second time point corresponding to two different time points during the washing of the annulus.

[0263] For example, the first time point is near the beginning of the washing procedure, e.g. less than ten seconds after washing is initiated, and the second time point is after a predefined interval, e.g. when the washing is estimated to be completed. For example, the predefined interval is based on historical trends or machine learning / mathematical predictions. Optionally, one or more acoustic signals are taken at a plurality of time points during the washing procedure, e.g. step 1908 is repeated, and / or each time point represents an interval of time over which an acoustic signal is collected by the acoustic sensor. Preferably, the acoustic sensor is attached to a washing tool performing the washing procedure.

[0264] Step 1910 comprises confirming whether washing of the well is complete. A change in the one or more acoustic signals detected at the first time point and the one or more acoustic signals detected at a second time point is associated with a change in the washing flow rate. The washing fluid and hence washing fluid flow changes depending on the particles and / or material present in the washing fluid, which results in a change in acoustic signals detected by the acoustic sensor.

[0265] For example, at a first time point when the washing is initiated, the level of particles / material in the washing fluid is relatively high as the washing process removes mud and other built-up material. Consequently, the washing flow rate is relatively low. When washing is complete at a second time point, the level of particles and / or material present in the washing fluid will be reduced as mud or other substances have been at least partially washed out, and the washing flow rate is higher. Therefore, there is a change in an acoustic signal detected at the first time point in comparison to an acoustic signal detected at a second time point.

[0266] Beneficially, confirming whether washing of the well is complete increases efficiency and accuracy of the PA procedure. For example, a washed well can result in a more secure cement plug. Therefore, confirming washing is completed to a predefined standard before installation of the cement plug prevents e.g. leaks or other imperfections from being detected only after the cement plug has been placed, which would require an additional plug, seal, or barrier to rectify (essentiallya repeat of the whole PA procedure at a separate depth) or the removal and replacement of the inadequate cement plug. While operators may try and compensate for such scenarios by increasing the length of the cement plug or the time spend washing the well, thereby lowering the likelihood of leaks, this requires additional resources that are not required when efficient washing of the well is confirmed.

[0267] If it is confirmed that the washing of the well is complete, method 1900 proceeds to step 1910 and, optionally, installation of cement in the annulus is initiated. Alternatively, if it is not confirmed that the washing of the well is complete: any of steps 1902 - 1908 are repeated, e.g. washing of the well is continued and step 1908 is repeated; method 1900 is suspended, e.g. to examine washing tools used to wash the annulus; or method 1900 proceeds without confirmation that the washing of the well is complete. Preferably, if method 1900 proceeds without confirmation that the washing of the well is complete, steps 1908 - 1912 are modified to compensate for an incomplete washing procedure. For example, a different composition of cement is used, or a longer cement plug is installed in the well.

[0268] Step 1910 comprises obtaining a set of verification parameters from an acoustic sensor. The set of verification parameters are indicative of the presence of leaks in the cement plug and are collected after installation of cement in the annulus is initiated. The acoustic sensor of step 1910 is either the same sensor as used in step 1906 or an additional acoustic sensor. Optionally, the set of verification parameters comprises additional data, such as CBL or ultrasonic tracks over the target depth range, pressure or temperature data, or a barrier gas permeability test track. The additional data is obtained from a further sensor, e.g. an ultrasonic sensor or a pressure sensor, or from the application database.

[0269] Optionally, the set of verification parameters are stored in the application database, which can then be used in planning or optimising installation of another cement barrier at a different depth and / or in a different well. For example, if the set of verification parameters indicate a leak is present, a plan for additional cement plug is generated with improved parameters based on the set of verification parameters, preferably in addition to other data collected during the PA procedure.In another example, the set of verification parameters are used to train or update a machine learning module to generate future operation steps associated with performing a successful PA procedure.

[0270] Figure 20 is a flowchart illustrating a method for diagnosing a PA procedure. Method 2000 is optionally a method for generating output 1300 of Figure 13, or for use with any of the systems or apparatus described herein, and any of the steps of method 2000 are optionally performed in isolation, in addition, or as a replacement to any of the method steps described herein.

[0271] Method 2000 comprises: step 2002, obtain a set of diagnostic parameters; step 2004, obtain a set of procedure parameters; step 2006, analyse a set of verification parameters to determine whether there is a leak in the barrier; step 2008, assign a barrier quality value indicative of success; and step 2010, output the barrier quality value.

[0272] Step 2002 comprises obtaining a set of diagnostic parameters from a network of sensors and / or the application database. The set of diagnostic parameters comprises any of gamma ray signals, cement bond logs (CBL) and / or variable density logs (VDL) over the target depth range of the well, other sensor data, or a combination thereof. Data within the set of diagnostic parameters is collected or otherwise measured before the any operations associated with the PA procedure are initiated, such as ablating a control line, washing the well, or setting a cement plug.

[0273] The set of diagnostic parameters are associated with planning the PA procedure, such that the set of diagnostic parameters are used to determine an operation plan, e.g. operation plan 2014, and any operating parameters used for performing the PA procedure.

[0274] For example, the set of diagnostic parameters comprise a casing collar locator gamma ray track, a CDL VDL track, and a control line track indicative of the presence and position of a control line, e.g. if it is determined that a control line is present. Optionally, the set of diagnostic parameters further comprises a multifinger log or an ultrasonic track. The gamma ray track is used for depth control, e.g. determining depth related parameters. The CBL VDL track is used to estimate potential washability, indicative of the chance of successfully washing anannulus section. The control line location and orientation track, optional based on the presence of control lines, is used for planning, and optionally verifying, ablation of the control line.

[0275] Step 2004 comprises obtaining, from the network of sensors, a set of procedure parameters associated with monitoring the execution of the PA procedure. The network of sensors comprises an acoustic sensor to detect acoustic signals generated by a washing fluid flow, and the set of procedure parameters comprises a washing pressure flowrate track. The set of procedure parameters further comprises a cementing track over the target depth range associated with installation of cement within the well.

[0276] The washing pressure flowrate track is used in monitoring washing of the well. The washing pressure and flowrates optimally stay between a set maximum flow pressure and a minimum flow pressure to ensure good washing efficiency, wherein the maximum and minimum flow pressures are determined based on the set of diagnostic parameters obtained at step 2002. Pressure and flowrate are therefore predefined in the operation plan, e.g. a road map or protocol followed to perform the PA procedure. Similarly, the cementing track is used to monitor cementing flow rate and pressure versus set parameters.

[0277] Optionally, the washing pressure flowrate track is accompanied with a washing particles track, which is obtained using acoustic particle flow detection to monitor successful lifting of particles from annuli. The washing particles track is also used to estimate particle mass lifted versus particle mass originally in place, therefore is used in calculating the efficiency of the washing procedure.

[0278] If there is a control line present, the set of procedure parameters further comprises a perforation ablation track. The perforation ablation track is used to confirm ablation depths and in calculation of confirmed ablated lengths over a total perforated length, either defined by length or by number of ablation points.

[0279] Step 2006 comprises obtaining a set of verification parameters after the execution of the planned PA procedure is completed, e.g. after a cement plug is installed in the well. The set of verification parameters are analysed to determine whether there is a leak in the barrier set during execution of the PA procedure. For example, the set of verification parameters comprises an acoustic track. A leak inthe barrier is associated with one or more peaks in the acoustic track where leaking gas or material is detected by the acoustic sensor, and therefore the acoustic track is indicative of whether there is a leak in the barrier.

[0280] Step 2008 comprises assigning the target depth range a barrier quality value indicative of whether the PA procedure was executed successfully. The barrier quality value is based on whether a leak is detected over the target depth range, as a leak is associated with the PA procedure being executed at least partially unsuccessfully, and no leaks detected is associated with the PA procedure being executed successfully.

[0281] Step 20 10 comprises outputting the barrier quality value. The output barrier quality value can then be used for validation of the PA procedure, such as whether it was successful, whether any steps of the PA procedure need repeating, or whether data associated with the PA procedure should be used in planning future PA procedures. For example, modifications to PA procedure operation plans are made based on the output barrier quality value in order to optimise future PA procedures, such as changing the amount of cement used to set a barrier, the length of time washing the well, or other operating parameter. Therefore, the output barrier quality value is used to improve both future and ongoing PA procedures.

[0282] Figure 21 illustrates a system for diagnosing, monitoring, or optimising a PA procedure according to the methods disclosed above, and can be used in combination or in place of any of the system units described herein. Specifically, Figure 21 shows a block diagram of an embodiment of a system 2100 according to example embodiments of the present disclosure.

[0283] System 2100 comprises: an annulus 2101 , e.g. annulus 101 , a casing 2103, e.g. casing 103; a subsea well 2105, e.g. subsea well 105; perforations 2103a, e.g. perforations 103a; a sensor 2115, e.g. acoustic sensor 115, auxiliary acoustic sensor 117, or ultrasonic imaging tool 121 ; a control line 2119, e.g. control line 119; a communication unit 1030; an application database 2132, e.g. application database 924; a computing system 2134; a cloud / edge device 2136; and a machine learning module 2138.

[0284] Machine learning module 2138 optionally comprises an artificial neural network (ANN). Communication unit 2130 obtains sensor data from sensor 2115,which is acquired by computing system 2134 and stored in application database 2132. Computing system 2134 obtains historical data from application database 2132, comprising historical fault data or trend data associated with previous sensor data, which is used to train the ANN. Computing system 2134 communicates with machine learning module 2138 such that sensor data from communication unit 2130 is used by the machine learning module 2138 to identify one or more conditions indicators based on the sensor data. Optionally, relevant features are extracted from the sensor data by computing system 2134 or machine learning module 2138.

[0285] The conditions indicators are used by machine learning module 2138 or computing system 2134 to determine a predicted state of the PA operation or annulus 2101 and. Computing system 2134 determines a current state of the PA operation based on a protocol, e.g. an operational plan or a roadmap obtained from the application database 2132, or sensor data obtained from communication unit 2130. The current state of the PA operation is compared to the planned state of the PA operation to predict whether a failure mode is probable. Optionally, historical data obtained from application database 2132 is also used for failure mode prediction.

[0286] In one example, machine learning module 2138 obtains sensor data from communication unit 2130, which is collected from sensor 2115, and verifies whether control line 2119 has been ablated based on the sensor data.

[0287] Optionally, computing system 2134 comprises any of communication unit 2130, application database 2132, or machine learning module 2138. Further optionally, communication unit 2130, application database 2132, computing system 2134 or machine learning module 2138 communicate with cloud / edge device 2136, such that steps of the methods described herein and operations performed by system 2100 or system 900 of Figure 9 are performed remotely. For example, the tool comprises 2130, which communicates with computing system 2134 via cloud / edge device 2136 for real-time or near real-time acquisition of sensor data from sensor 2115.

[0288] As disclosed above, the methods, apparatus and systems described herein optionally comprise a machine learning module such as machine learning module2138 of Figure 21. For example, the machine learning module comprises an artificial neural network (ANN), which is either a dedicated hardware module for implementing the ANN, also known as a hardware ANN, or a software module for emulating a hardware ANN, also known as an emulated ANN.

[0289] The machine learning module is rule-based and / or deterministic, such as a decision tree or Bayes model e.g. if configured for determining control line orientation or location. Alternatively or additionally, the machine learning module comprises classifiers such as a k-nearest neighbour algorithm, a support vector machine, or a convolutional neural network, e.g. if configured to determine the status of the well and / or PA procedure.

[0290] Preferably, the machine learning module comprises a generative Al algorithm. Generative Al algorithms are configured for the creation of new content, such as numerical data, images, or other forms of data, by extracting features from training data, e.g. learning patterns and structures from existing examples. For example, the generative Al algorithm is a deep neural network configured to generate output that closely resembles the input data it was trained on, such as an autoencoder, a generative adversarial network, or another form of neural network.

[0291] Training data for the generative Al algorithm comprises historical data, data from one or more sensors, data from the application data, simulated data, or a combination thereof. For example, training data comprises data from well sensors, drilling logs, historical performance, or the application database, or a combination thereof. The generative Al algorithm is trained to extract features from the training data such that the algorithm can identify patterns indicative of potential issues, such as equipment malfunction, reservoir changes, or flow irregularities. Optionally, the generative Al algorithm is configured to predict future well behaviour, helping prevent problems before they escalate via mitigation and / or optimisation procedures. Additionally, the predicted future state of the well can be compared to the current state, determined by current or real-time sensor data, allowing for detection of emerging deviations from optimal and planned procedures.

[0292] Optionally, the generative Al algorithm simulates different extraction scenarios and / or is used to indicate optimal operational parameters, e.g. based onanalysis of feature extraction. Beneficially, use of the generative Al algorithm increases efficiency and sustainability in PA procedures, such as from diagnostics determination, procedure optimisation both during planning and execution, or operation status monitoring. Due to the complexities of PA procedures, traditional mathematical-based simulations can require extensive computational resources and rely on known physical and chemical interactions that can be slow and data intensive to calculate. Additionally, such calculations require a prohibitive number of variables from a large number of sensors, further increasing the resources required to accurately predict PA procedures. However, use of one or more generative Al algorithms can reduce the computational resources, sensor resources, time and the amount of data and required in comparison to these traditional methods, without compromising on accuracy. Moreover, as generative Al algorithms can be tailored to specific PA procedures, the outputs are more accurate than use of knowhow-based methods alone, resulting in more accurate detailed operation plans, more efficient diagnostic and fault detection, and better monitoring of the PA procedure.

[0293] In one example, the machine learning module is configured to predict an estimated state of a PA procedure based on data obtained from one or more sensors, such as surface sensors and downhole sensors configured to measure pressure, flowrate, or acoustic / ultrasonic data. The estimated state of a PA procedure is a planned state of the well and / or PA procedure at a predefined time point or operational step of the PA procedure, which is predicted based on historical data, including historical faults and trends. For example, the machine learning module is trained using pre and post PA procedure data collected from a network of sensors from a historical PA procedure at a different well.

[0294] By comparing the estimated state of the PA procedure with a current state, e.g. where the current state is obtained based on real-time monitoring of the PA procedure, differences between the estimated and current state can be identified and an operation status determined. These differences are indicative of the PA procedure not proceeding as planned, as the estimated state of the PA procedure is optionally used to generate an operation plan comprising operational steps used for planning and subsequently performing the PA procedure. Therefore, operation status is associated with whether there is a difference between the estimated andcurrent state. For example, the difference between a first operational step in the estimated and current state exceeds a predetermined quality threshold, indicating that a correction or mitigation action is required. If the difference exceeds predetermined safely threshold, the PA procedure can be halted or otherwise postponed. The difference between a second operation step does not exceed a predetermined quality threshold, indicating that no action is required and the PA procedure is proceeding as planned.

[0295] The machine learning module is input raw sensor data or pre-processed data, depending on: the one or more algorithms comprised within the machine learning module; the size of the machine learning module; the available computational resources; or other operational or computational constraints and considerations. Pre-processing includes any of data fusion to merge two or more data points, data filtering to exclude irrelevant or undesired data points, and feature extraction to optimise the data input to the machine learning module.

[0296] The machine learning module is optionally further configured to perform diagnostics and prognostics on the well associated with the PA procedure, identify conditions indicators, or predict a probability of one or more failure modes, such as the probability of a control line not being fully ablated or probability of a leak being present in a cement plug. For example, the machine learning module comprises a plurality of Al algorithms, and / or one complex generative Al algorithm.

[0297] In another example, the machine learning module comprises a plurality of modules configured to simulate planned operations of a PA procedure. For example, the machine learning module comprises a hydraulics module, a conveyance module, an optional computational fluid dynamics (CFD) module, and / or an optional perforation / ablation module. The hydraulics module is configured to perform a first set of hydraulic calculations, including any of: ball drop calculations, pressure drop calculations, cementing programs, and volume calculations. The CFD module is configured to perform a second set of hydraulic calculations, including any of: cement slumping calculations, washing efficiency (lifting of particles), multiple annuli calculations, and calculations arising from special or atypical tubular designs, such as tubing and casing designs. The conveyance module is configured to perform conveyance simulations based ontorque and drag phenomena, including estimating downhole forces from wireline, coiled tubing, and drillpipe conveyance, and / or determining whether BHA can reach a the target depth, whether downhole operational step programs can be executed, and whether a return to the surface is probable. The perforation / ablation module is configured to verify correct perforation and ablation and / or determine mechanical perforation and ablation parameters, e.g. by verifying geometries of perforation holes or slots, and perforation depth, or based on the setup of e.g. a perforation gun.

[0298] In another example, the machine learning module is configured to generate a barrier quality value based on an input of well diagnostic information such as a depth dependent casing collar locator gamma ray signal / track or a cement bond log (CBL) / variable density (VDL) signal / track.

[0299] Alternatively or in addition to the machine learning module, any of the above methods, apparatus or systems comprise a numerical / computational model for generating an operational plan or optimising parameter ranges within the operational plan. For example, a numerical one-dimensional model is used to estimate mass flow rate using standard thermodynamic principals for, e.g., determining optimal and efficient washing pressure flowrate parameters. Assumptions comprise any of: Darcy’s law is followed; materials are modelled as compressible Newtonian fluids, there is laminar flow, and the well is at steady state; there is unidirectional flow along the well; and boundary conditions at the top and bottom of the well are indicative of an uniform / infinite well. Beneficially, a numerical model provides an efficient and explainable estimation of e.g. mass flow rates. Optionally, a numerical model is used to validate mass flow rates calculated using one or more machine learning modules, or during training of the machine learning module. Machine learning modules do not require multiple pre-defined assumptions to form an accurate approximation at speed and, as such, can produce fast and accurate mass flow rate estimates. Combining with a numerical model increases explainability and confidence in machine learning module based mass flow rate estimates.

[0300] Optionally, the machine learning module comprises a machine learning model for analysing data obtained from an ultrasonic logging tool,e.g. after the ultrasonic logging tool has been used to survey a perforated pipe. For example, the machine learning model is configured to classify ultrasonic data around one or more (preferably each) perforated hole. This can then be used to determine a status of a flat pack, e.g. after an attempt of cutting has been made.

[0301] The machine learning model is preferably trained using a comprehensive dataset of labelled 2D or 3D images comprising ultrasonic data, which are captured from regions surrounding the perforated holes in the well casing. These images are optionally black and white and have specific dimensions that balance the need for detailed information with the constraints of the training hardware. The dataset preferably comprises images obtained from controlled workshop environments, where various configurations of flat pack placement are tested, as well as images from full- scale test wells that simulate real-life conditions. Additionally or alternatively, data captured from well intervention procedures is used to train the machine learning mode, and / or simulated data. The training data is preferably augmented through techniques such as normalization, flipping, mirroring, and shifting to enhance the model's robustness and performance.

[0302] The model optionally undergoes supervised training, e.g. when labelled training sets are available for training and implementation, utilizing optimization algorithms and loss functions tailored to the classification task. The training process involves multiple epochs and employs early stopping criteria to prevent overfitting. The model is preferably retrained periodically to incorporate new data, ensuring that it remains accurate and up-to-date. The training process is designed to be efficient, allowing for complete retraining when necessary without significant time investment.

[0303] The architecture of the machine learning model is preferably based on convolutional neural networks (CNNs), which are well-suited for image classification tasks, although the skilled person will understand suitable alternative classifiers or architectures can be used, such as those described below. For example, the model comprises multiple layers that progressively extract features from the input images, culminating in a dense layer that outputs classification results. The model is designed to handle normalized3D images and provides confidence levels for various classes related to the status of the components observed through the perforated holes.

[0304] The machine learning model is preferably integrated into one or more specialized software applications used for analysing downhole ultrasound data. These applications automate the classification process, increasing efficiency, significantly reducing the time required for manual inspection, and reducing computational requirements. The model is deployed in a format that allows for easy updates and replacements, ensuring that the system can adapt to improvements and new developments in machine learning techniques. The integration of the model into the software enhances the efficiency and accuracy of well intervention operations more broadly, providing valuable insights that aid in decision-making and operational planning.

[0305] In a specific example implementation, the machine learning model is trained on a dataset of between one thousand and ten thousand labelled 3D images, e.g. 1400 to 2800 images, with multiple dimensions of, e.g. 32x48x128x1. The images, e.g. ultrasonic data, are captured from regions around perforated holes in a petroleum well, e.g. using an ultrasonic image tool. The dataset comprises over 1000 images from controlled workshop testing and less than 1000 images from a full-scale test well. The images are normalized and augmented to improve model performance.

[0306] The model is trained using an Adam optimizer with a static learning rate and categorical cross-entropy as the loss function. The training process spans approximately 100 epochs, with early stopping criteria based on validation loss. The model architecture consists of a 4-layered 3D CNN, which processes the input images and outputs confidence levels for five classes related to the status of the flat pack and casing.

[0307] The model is integrated into an application for analysing downhole ultrasound data. This integration automates the classification of perforated holes, reducing the time required for manual inspection from several hours to approximately 10-15 minutes. The model is deployed as a standard machine-readable file, allowing for easy updates and replacements. By leveraging advanced machine learning techniques, the invention provides a robust and efficient solution for monitoring well intervention operations,enhancing the safety and effectiveness of plug and abandonment procedures.

[0308] Alternative or additional machine learning models are provided below, which are non-limiting examples. Recurrent Neural Networks (RNNs) are a class of neural networks designed to recognize patterns in sequences of data. They are particularly useful for time-series data or any data where the context from previous steps is important. For ultrasonic data, RNNs can be used to analyse sequences of images or signals, capturing temporal dependencies that might be relevant for identifying the position of the flatpack. Long Short-Term Memory Networks (LSTMs) are a specialized type of RNN designed to overcome the limitations of RNNs, such as the vanishing gradient problem. LSTMs are capable of learning long-term dependencies, making them suitable for analysing sequences of ultrasonic data where the position of the flatpack might be influenced by historical context within the data. Support Vector Machines (SVMs) are a type of supervised learning algorithm that can be used for classification tasks. SVMs work by finding the hyperplane that best separates different classes in the feature space. For ultrasonic data, SVMs can be used to classify the position of the flatpack by transforming the data into a higher-dimensional space where the classes become more separable. Random Forests are an ensemble learning method that combines multiple decision trees to improve classification accuracy. Each tree in the forest is trained on a random subset of the data, and the final classification is determined by majority voting. Random Forests can handle high-dimensional data and are robust to overfitting, making them a viable alternative for classifying flatpack positions from ultrasonic data. Gradient Boosting Machines (GBMs) are another ensemble learning technique that builds multiple decision trees sequentially, with each tree correcting the errors of the previous one. GBMs are effective for classification tasks and can be used to identify the position of the flatpack by iteratively improving the model's performance on the ultrasonic data. K-Nearest Neighbors (KNN) is a simple, non-parametric algorithm used for classification. KNN works by finding the k-nearest data points in the feature space and assigning the class based on the majority class of these neighbors. For ultrasonic data, KNN can be used to classifythe flatpack position by comparing the input data to a labeled dataset of known positions. Principal Component Analysis (PCA) is a dimensionality reduction technique that can be used to transform high-dimensional data into a lower-dimensional space while preserving most of the variance. Combined with clustering algorithms such as K-means, PCA can be used to identify patterns in ultrasonic data and classify the position of the flatpack based on the transformed features. Autoencoders are a type of neural network used for unsupervised learning. They work by encoding the input data into a lower-dimensional representation and then decoding it back to the original dimensions. The encoded representation can be used for classification tasks. For ultrasonic data, autoencoders can be trained to learn the underlying structure of the data and then used to classify the flatpack position based on the learned features. Decision Trees are a simple yet powerful classification method that splits the data into subsets based on feature values. Each node in the tree represents a decision point, and the leaves represent the final classification. Decision Trees can be used to classify the position of the flatpack by recursively partitioning the ultrasonic data based on the most informative features. Bayesian Networks are probabilistic graphical models that represent the dependencies among variables. They can be used for classification by calculating the posterior probabilities of different classes given the input data. For ultrasonic data, Bayesian Networks can model the probabilistic relationships between the features and the flatpack position, providing a robust classification framework. These alternatives offer a range of approaches for classifying or identifying the position of the flatpack from ultrasonic data, each with its own strengths and suitable applications.

[0309] Optionally, in addition to any of the normalization, flipping, mirroring, and shifting techniques mentioned, other data augmentation methods can be employed to further enhance the robustness of the machine learning model. These methods may include rotation, scaling, or adding noise to the images. Such techniques help in creating a more diverse training dataset, which can improve the model's ability to generalize to new, unseen data.

[0310] Optionally, standard evaluation and validation processes are used to assess the performance of the machine learning model. This can involvesplitting the dataset into training, validation, and test sets, and using metrics such as accuracy, precision, recall, F1 -score, and confusion matrices to evaluate the model's performance. Cross-validation techniques, such as k- fold cross-validation, can also be employed to ensure the model's robustness and reliability.

[0311] Optionally, the model is optimized for faster inference times to ensure it can be deployed effectively in real-time environments, such as on-site well intervention operations. Techniques such as model quantization, pruning, and the use of specialized hardware accelerators (e.g., GPUs or TPUs) can be employed to reduce latency and improve processing speed. The software architecture should support asynchronous data processing and parallel execution to handle high-throughput data streams efficiently. The system is preferably capable of processing and classifying ultrasonic data in near real-time, providing immediate feedback to operators and enabling timely decision-making during well intervention procedures. This real-time capability ensures that the model can be used effectively in dynamic and time-sensitive operational contexts.

[0312] Optionally, the machine learning model is configured to integrate seamlessly with existing well intervention systems and workflows. This includes ensuring compatibility with various ultrasonic logging tools and data formats. The model is preferably adaptable to different data preprocessing pipelines and capable of handling various data resolutions and formats, ensuring that the model can be deployed without requiring significant modifications or upgrades.

[0313] Preferably, the machine learning model is configured to continuously learn and improve over time. Automated data collection pipelines are established to gather new ultrasonic data and associated labels from ongoing operations. This data is preferably curated and pre-processed to ensure quality and relevance. Periodic retraining schedules are defined, allowing the model to be updated with the latest data while minimizing downtime. Techniques such as transfer learning and incremental learning can be employed to efficiently update the model without requiring complete retraining from scratch. By implementing continuous learning mechanisms,the model can adapt to changing conditions and improve its performance over time, ensuring that it remains accurate and up-to-date.

[0314] Optionally, data encryption is applied both at rest and in transit to prevent unauthorized access and tampering.

[0315] Figure 22 shows an example computing system. Specifically, Figure 22 shows a block diagram of an embodiment of a computing system 2200 according to example embodiments of the present disclosure.

[0316] Computing system 2200 can be configured to perform any of the operations disclosed herein. Computing system 2200 includes one or more computing devices 2202. Computing device 2202 of computing system 2200 comprises one or more processors 2204 and memory 2206. device(s) 2202 of computing system 2200 comprise one or more processors 2204 and memory 2206. One or more processors 2204 can be any general-purpose processor(s) configured to execute a set of instructions. For example, one or more processors 2204 can be one or more general-purpose processors, one or more field programmable gate array (FPGA), and / or one or more application specific integrated circuits (ASIC). In one embodiment, one or more processors 2204 include one processor. Alternatively, one or more processors 2204 include a plurality of processors that are operatively connected. One or more processors 2204 are communicatively coupled to memory 2206 via address bus 2208, control bus 2210, and data bus 2212. Memory 2206 can be a random-access memory (RAM), a read-only memory (ROM), a persistent storage device such as a hard drive, an erasable programmable read-only memory (EPROM), and / or the like. Computing device(s) 2202 further comprise input / output (I / O) interface 2214 communicatively coupled to address bus 2208, control bus 2210, and data bus 2212.

[0317] Memory 2206 can store information that can be accessed by one or more processors 2204. For instance, memory 2206 (e g., one or more non-transitory computer-readable storage mediums, memory devices) can include computer- readable instructions (not shown) that can be executed by one or more processors 2204. The computer-readable instructions can be software written in any suitable programming language or can be implemented in hardware. Additionally, or alternatively, the computer-readable instructions can be executed in logically and / or virtually separate threads on one or more processors 2204. For example,memory 2206 can store instructions (not shown) that when executed by one or more processors 2204 cause one or more processors 2204 to perform operations such as any of the operations and functions for which computing system 2200 is configured, as described herein. In addition, or alternatively, memory 2206 can store data (not shown) that can be obtained, received, accessed, written, manipulated, created, and / or stored. The data can include, for instance, the data and / or information described herein in relation to Figures 1 to 6. In some implementations, computing device(s) 2202 can obtain from and / or store data in one or more memory device(s) that are remote from the computing system 2200.

[0318] Computing system 2200 further comprises storage unit 2216, network interface 2218, input controller 2220, and output controller 2222. Storage unit 2216, network interface 2218, input controller 2220, and output controller 2222 are communicatively coupled to central control unit or computing devices 2202 via I / O interface 2214.

[0319] Storage unit 2216 is a computer readable medium, preferably a non- transitory computer readable medium, comprising one or more programs, the one or more programs comprising instructions which when executed by one or more processors 2204 cause computing system 2200 to perform the method steps of the present disclosure. Alternatively, storage unit 2216 is a transitory computer readable medium. Storage unit 2216 can be a persistent storage device such as a hard drive, a cloud storage device, or any other appropriate storage device.

[0320] Network interface 2218 can be a Wi-Fi module, a network interface card, a Bluetooth module, and / or any other suitable wired or wireless communication device. In an embodiment, network interface 2218 is configured to connect to a network such as a local area network (LAN), or a wide area network (WAN), the Internet, or an intranet.

[0321] Figure 22 illustrates one example computer system 2200 that can be used to implement the present disclosure. Other computing systems can be used as well. Computing tasks discussed herein as being performed at and / or by one or more functional unit(s) (e.g., as described in relation to Figure 2) can instead be performed remote from the respective system, or vice versa. Such configurations can be implemented without deviating from the scope of the present disclosure.The use of computer-based systems allows for a great variety of possible configurations, combinations, and divisions of tasks and functionality between and among components. Computer-implemented operations can be performed on a single component or across multiple components. Computer-implemented tasks and / or operations can be performed sequentially or in parallel. Data and instructions can be stored in a single memory device or across multiple memory devices.

[0322] Regarding the above disclosure, references to items in the singular should be understood to include items in the plural, and vice versa, unless explicitly stated otherwise or clear from the context. Additionally, grammatical conjunctions are intended to express any and all disjunctive and conjunctive combinations of conjoined clauses, sentences, words, and the like, unless otherwise stated or clear from the context. Thus, the term “or” should generally be understood to mean “and / or” and so forth. The use of any and all examples, or exemplary language (“e.g.,” “such as,” “including,” or the like) provided herein, is intended merely to better illuminate the embodiments and does not pose a limitation on the scope of the embodiments or the claims.

[0323] Methods described herein may relate to a computer storage product with a non-transitory computer-readable medium (also can be referred to as a non- transitory processor-readable medium) having instructions or computer code thereon for performing various computer-implemented operations, such that the methods are performed. The computer-readable medium (or processor-readable medium) is non-transitory in the sense that it does not include transitory propagating signals per se (e.g., a propagating electromagnetic wave carrying information on a transmission medium such as space or a cable). The media and computer code (also can be referred to as code) may be those designed and constructed for the specific purpose or purposes. Examples of non-transitory computer-readable media include, but are not limited to, magnetic storage media such as hard disks, floppy disks, and magnetic tape, optical storage media such as Compact Disc / Digital Video Discs (CD / DVDs), Compact Disc-Read Only Memories (CD-ROMs), and holographic devices; magneto-optical storage media such as optical disks; carrier wave signal processing modules; and hardware devices that are specially configured to store and execute program code, such asApplication-Specific Integrated Circuits (ASICs), Programmable Logic Devices (PLDs), Read-Only Memory (ROM) and Random-Access Memory (RAM) devices. Other embodiments described herein relate to a transitory computer program product, which can include, for example, the instructions and / or computer code discussed herein.

[0324] Some embodiments and / or methods described herein can be performed by software (executed on hardware), hardware, or a combination thereof. Hardware modules include, for example, a general-purpose processor, a field programmable gate array (FPGA), and / or an application specific integrated circuit (ASIC). Software modules (executed on hardware) can be expressed in a variety of software languages (e.g., computer code), including C, C++, Java, Ruby, Visual Basic, Python, and / or other object-oriented, procedural, or other programming language and development tools. Examples of computer code include, but are not limited to, micro-code or micro-instructions, machine instructions, such as produced by a compiler, code used to produce a web service, and files containing higher-level instructions that are executed by a computer using an interpreter. For example, embodiments can be implemented using imperative programming languages (e.g., C, Fortran, etc.), functional programming languages (Haskell, Erlang, etc.), logical programming languages (e.g., Prolog), object-oriented programming languages (e.g., Java, C++, etc.) or other suitable programming languages and / or development tools. Additional examples of computer code include, but are not limited to, control signals, encrypted code, and compressed code. Optionally, the embodiments and / or methods described herein are implemented using an operating system such as Robot Operating System (ROS).

[0325] The following statements encompass example embodiments of systems and methods disclosed herein and do not define the scope of the invention, which is instead defined in the appended claims. Any of the example embodiments can be combined, and dependencies are not limited to those explicitly listed.Statements of Invention1. A method of validating a subsea well annulus (101 ) washing operation, comprising the following steps: a) with an annulus washing tool (100) arranged at a perforated section of a well pipe (103), providing a washing fluid flow (110) from wash fluid ports (e.g. nozzles) (111 ) through perforations (103a) in the well pipe, into the annulus (101 ), and back through perforations into the inner bore (102) of the well pipe; b) with an acoustic sensor (115) being part of the annulus washing tool (100), detecting acoustic signals generated by the washing fluid flow (110); c) detecting a change of acoustic signals detected in step b).2. A method according to statement 1 , wherein step b) further includes detecting, with an auxiliary acoustic sensor (117) that is part of the annulus washing tool (100) and arranged above the said acoustic sensor (115), acoustic signals of the washing fluid flow flowing through the inner bore (102) of the well pipe (103).3. A method according to statement 2, wherein it further comprises the following step: d) based on at least the acoustic signals detected with the auxiliary sensor (117), calculating an amount of material that has been washed out of the annulus (101 ) with the washing fluid flow (110).4. Method according to any one of statements 1 , 2, or 3, wherein it further comprises-storing detected acoustic signals in a computer-readable memory unit;-with a computer, performing step c) with acoustic signals stored in the computer-readable memory unit.5. An annulus washing tool (100) for cleaning of a subsea well annulus, comprising wash fluid ports (e.g. nozzles) (111 ) configured to provide a washing fluid flow (110), characterized in that the annulus washing tool (100) further comprises an acoustic sensor (115).6. An annulus washing tool (100) according to statement 5, characterized in that it further comprises an auxiliary acoustic sensor (117) arranged a distance above the said acoustic sensor (115).7. An annulus washing tool (100) according to statement 6 or statement 102, characterized in that it comprises or is connected to a computer-readable memory unit and a computing unit programmed to detect a change of acoustic signals.8. A method of setting a permanent plug in a subsea well (105) having a well pipe (103) and a control line (119) located in an annulus (101 ) outside the well pipe, wherein the method comprises a) with a perforation tool (123), providing perforations (103a) in the well pipe (103) for removal of the control line (119) along a portion of the well pipe (103) by severing the control line when providing said perforations (103a); b) washing the annulus (101 ) and installing cement in the annulus and the bore of the well pipe; wherein the method further comprises with an ultrasonic imaging tool (121 ) arranged in the bore (102) of the well pipe (103), recording ultrasonic images of the perforations (103a); inspecting the ultrasonic images and verifying removal of the control line (119).9. A method for controlling a Plug and Abandonment (PA) procedure for a well, wherein the PA procedure comprises washing and setting a permanent plug in the well to form a barrier, the method comprising: obtaining a set of diagnostic parameters from a network of sensors or an application database, wherein the set of diagnostic parameters comprises initial well parameters, cement bond logs or variable density logs over a target depth range of the well before the PA procedure; using the set of diagnostic parameters to plan the PA procedure, determining one or more operational steps; performing a simulation of the one or more operational steps, wherein performing the simulation comprises performing hydraulics calculations and estimating downhole forces based on the set of diagnostic parameters; analysing the simulation to generate an operation plan comprising the one or more operational steps; once the PA procedure is initiated according to the operation plan, monitoring the PA procedure using data collected from the network of sensors to determine an operation status of the PA procedure, wherein the network of sensors comprise an ultrasonic imaging tool and an acoustic sensor; validating the PA procedure based on determining a barrier quality value, wherein the barrier quality is associated with whether there is a leak in the barrier; and storing, in the application database, data associated with execution of the PA procedure.10. The method of statement 9, further comprising processing the set of diagnostic parameters, processing steps comprising:merging two or more parameters of the set of diagnostic parameters, resulting in data fusion; filtering one or more parameters of the set of diagnostic parameters; or extracting a target feature from one or more parameters of the set of diagnostic parameters.11 . The method of statement 9, wherein monitoring the PA procedure to determine an operation status of the PA procedure comprises: using a machine learning module to predict an estimated state of the PA procedure based on the set of diagnostic parameters, wherein the machine learning module is trained with historical parameters associated with historical PA procedures obtained from the application database; obtaining an set of execution parameters measured by the network of sensors while the PA procedure is performed; generating a current state of the PA procedure based on the set of execution parameters; and determining an operation status based on comparing the estimated state of the PA procedure to the current state of the PA procedure.12. The method of statement 11 , wherein the machine learning module comprises a generative artificial intelligence (Al) algorithm.13. The method of statement 9, wherein the set of diagnostic parameters comprises hydraulics parameters and conveyance parameters.14. The method of statement 13, wherein performing the simulation comprises: generating a hydraulics module configured to perform hydraulics calculations comprising ball drop calculations, pressure drop calculations, volume calculations, or calculations associated with cementing the well; generating a conveyance module configured to estimate downhole forces based on torque and drag calculations; andusing the hydraulics module and the conveyance module to simulate one or more operational steps.15. The method of statement 14, wherein performing the simulation further comprises: generating a computational fluid dynamics (CFD) module configured to perform further hydraulics calculations comprising any of: cement slumping calculations, washing efficiency calculations, listing of particles calculations, multiple annuli calculations, or calculations associated with tubular, tubing, or casing designs; and using the CFD module to simulate one or more operational steps.16. The method of statement 14 or 15, further comprising: generating a perforation ablation module configured to verify correct perforation and ablation or determine mechanical perforation and ablation parameters based on geometry of perforations, perforation depth, or perforation gun setup; and using the perforation ablation module to simulate one or more operational steps.17. The method of statement 9, wherein a data analytics interface initiates simulation of the one or more operational steps, and is configured to display the simulation of the one or more operational steps.18. The method of statement 9, further comprising storing simulation of the one or more operational steps associated with the set of diagnostic parameters in the application database.19. The method of statement 9, further comprising identifying faults or risks in the well or PA procedure based on simulation of the one or more operational steps.20. The method of statement 9, further comprising determining whether a leak is detected in the barrier after execution of the PA procedure.21 . The method of statement 9, wherein the network of sensors comprises surface sensors or downhole sensors, and the network of sensors configured to measure pressure, flowrate, or acoustic / ultrasonic data.22. A computer-readable medium storing instructions for performing the method of any of statements 9-21 .23. A system comprising: a processor configured to communicate with an application database; a network of sensors comprising an ultrasonic imaging tool and an acoustic sensor; and a memory storing instructions that, when executed, cause the system to perform the method of any of statements 9-21 .24. A method for simulating operation of a Plug and Abandonment (PA) procedure for a well, wherein the PA procedure comprises washing and installing cement in the well, the method comprising: obtaining, from an application database, a set of diagnostic parameters, wherein the set of diagnostic parameters comprises hydraulics parameters and conveyance parameters associated with a planned PA procedure; generating a hydraulics module configured to perform hydraulics calculations comprising ball drop calculations, pressure drop calculations, volume calculations, or calculations associated with cementing the well; generating a conveyance module configured to estimate downhole forces based on torque and drag calculations; and using the hydraulics module and the conveyance module to simulate one or more operational steps of the planned PA procedure based on the set of diagnostic parameters.25. The method of statement 24, further comprising:generating a computational fluid dynamics (CFD) module configured to perform further hydraulics calculations comprising any of: cement slumping calculations, washing efficiency calculations, listing of particles calculations, multiple annuli calculations, or calculations associated with tubular, tubing, or casing designs; and using the CFD module to simulate one or more operational steps of the planned PA procedure.26. The method of statement 24, further comprising: generating a perforation ablation module configured to verify correct perforation and ablation or determine mechanical perforation and ablation parameters based on geometry of perforations, perforation depth, or perforation gun setup; and using the perforation ablation module to simulate one or more operational steps of the planned PA procedure.27. The method of statement 24, wherein simulation of the one or more operational steps is used to generate or verify an operation plan comprising operational steps used for planning the PA procedure.28. The method of 27, wherein simulation of the one or more operational steps is used to generate or verify an operational envelope comprising a range of operation parameters for performing the PA procedure optimally.29. The method of statement 27, further comprising comparing the operation plan to a monitored PA procedure to determine an operational status of the monitored PA procedure.30. The method of statement 27, wherein a machine learning module comprises the hydraulics module and the conveyance module such that the machine learning module simulates operation of the PA procedure.31 . The method of statement 30, wherein the machine learning module comprises a generative artificial intelligence (Al) algorithm configured to generate the operation plan.32. The method of statement 24, wherein a data analytics interface initiates simulation of the one or more operational steps, and is configured to display the simulation of the one or more operational steps.22. The method of statement 24, further comprising storing simulation of the one or more operational steps associated with the set of diagnostic parameters in the application database.34. The method of statement 24, further comprising identifying faults or risks in the well or PA procedure based on simulation of the one or more operational steps.35. The method of statement 24, wherein hydraulics parameters and conveyance parameters are obtained from one or more sensors associated with the well.36. The method of statement 35, wherein the one or more sensors comprise surface sensors or downhole sensors, and the one or more sensors are configured to measure pressure, flowrate, or acoustic / ultrasonic data.37. A computer-readable medium storing instructions for performing the method of any of statements 24-36.38. A system comprising: a processor configured to communicate with an application database; and a memory storing instructions that, when executed, cause the system to perform the method of any of statements 24-36.39. A method for monitoring operation of a Plug and Abandonment (PA) procedure for a well, the method comprising: obtaining, from a network of sensors comprising at least one acoustic sensor, a first set of parameters associated with the well;generating one or more processed parameters by performing one or more processing steps on each parameter of the first set of parameters, the processing steps comprising: merging two or more parameters of the first set of parameters, resulting in data fusion; filtering one or more parameters of the first set of parameters; or extracting a target feature from one or more parameters of the first set of parameters; predicting an estimated state of the PA procedure based on the one or more processed parameters; obtaining, from the network of sensors, a second set of parameters associated with monitoring the well after the PA procedure is initiated, wherein the second set of parameters comprises one or more acoustic signals; generating a current state of the PA procedure based on the second set of parameters; and determining an operation status based on comparing the estimated state of the PA procedure to the current state of the PA procedure.40. The method of statement 39, further comprising generating an operation plan based on the estimated state of the PA procedure, wherein the operation plan comprises operational steps used for planning the PA procedure.41 . The method of statement 40, wherein planning the PA procedure comprises evaluating risk prior to initiation of the PA procedure.42. The method of statement 41 , wherein planning the PA procedure further comprises simulating one or more operational steps of the operation plan.43. The method of 41 , wherein the PA procedure is initiated and executed according to the operation plan.44. The method of statement 39, wherein the operation status comprises a probability of one or more failure modes occurring.45. The method of statement 39, wherein the processing steps or predicting an estimated state is performed using a machine learning module, and wherein the machine learning module is trained with historical parameters associated with historical PA procedures stored in an application database.46. The method of statement 45, wherein the machine learning module comprises a generative artificial intelligence (Al) algorithm.47. The method of clam 45 or 46, further comprising storing the current state of the PA procedure or the second set of parameters in the application database.48. The method of statement 47, wherein the current state of the PA procedure or the second set of parameters stored in the application database is used for updating the machine learning module.49. The method of statement 47, wherein the current state of the PA procedure or the second set of parameters stored in the application database is used in predicting an estimated state of a subsequent PA procedure.50. The method of statement 39, wherein the network of sensors further comprises one or more surface sensors configured to measure any of: pressure, flowrate, weight, torque, or rotation data associated with the well or PA procedure.51 . The method of statement 39, wherein the network of sensors further comprises one or more downhole sensors configured to measure any of: pressure, temperature, acoustic, or ultrasonic data associated with the well or PA procedure.52. A computer-readable medium storing instructions for performing the method of any of statements 39-51.53. A system comprising:a network of sensors comprising at least one acoustic sensor, wherein the network of sensors are configured to detect data associated with monitoring planning or operation of a Plug and Abandonment (PA) procedure for a well; a processor configured to obtain data from the network of sensors and communicate with an application database; and a memory storing instructions that, when executed, cause the system to perform the method of any of statements 39-51 .54. A system for diagnosing a Plug & Abandonment (PA) operation comprising an artificial neural network (ANN), wherein the system is configured to perform the following steps: obtain sensor data from a network of sensors, the network of sensors comprising one or more surface sensors and one or more downhole sensors; obtain historical data from an application database, the historical data comprising historical fault data or trend data associated with previous PA operations; extract one or more relevant features from the data from the sensor data, wherein relevant features are determined based on the sensor data; use the ANN, trained using the historical data, to identify one or more conditions indicators based on the one or more relevant features, wherein the one or more conditions indicators are associated with performing the PA operation; determine a predicted state of the PA operation based on the one or more conditions indicators; and predict whether a failure mode is probable based on comparing the predicted state of the PA operation to a current state of the PA operation, the current state of the PA operation based on either:a protocol for the PA operation, the protocol comprising a plurality of planned operational steps or instructions to perform the PA operation; or updated sensor data from the network of sensors.55. A method for training the ANN of statement 54, the method comprising: obtaining the historical data from the application database, wherein the historical data comprises one or more historical relevant features and one or more historical conditions indicators associated with a historical PA operation or a simulated PA operation; and providing the historical data to the ANN, wherein the ANN is configured to perform the following steps until a stop criterion is met: output one or more predicted conditions indicators based on an input of the one or more historical relevant features; calculate, based on a loss function, a difference between the one or more predicted conditions indicators and the one or more historical conditions indicators; and based on the difference, modify hyperparameters of the ANN to reduce the difference.56. The method of statement 55, wherein the stop criterion is based on: a number of completed epochs during training of the ANN; the difference meeting or exceeding a stopping threshold; a comparison between the protocol and a historical protocol or a simulated protocol, wherein the comparison is indicative of whether the protocol is suitable; a comparison between the predicted state of the PA operation and a historical state or a simulated state, wherein the comparison is indicative of whether the predicted state is accurate;a comparison between whether the failure mode is probable and the historical data; or any combination thereof.57. A method for monitoring a Plug and Abandonment (PA) procedure for a well, wherein the PA procedure comprises washing an annulus of the well and setting a permanent plug in the well, the method comprising: determining that a control line is present in the well based on a cement bond log or an ultrasonic log; after ablation of the control line is initiated, obtaining one or more ultrasonic images from an ultrasonic imaging tool within the well; verifying ablation of the control line over an ablated length based on the ultrasonic images associated with perforations of the control line; after washing of the annulus is initiated, obtaining one or more acoustic signals from a first acoustic sensor within the well at a first time point and a second time point; confirming whether washing of the well is complete based a change in the one or more acoustic signals detected at the first time point and the one or more acoustic signals detected at the second time point; and obtaining a set of verification parameters from a second acoustic sensor after installation of cement in the annulus is initiated, wherein the set of verification parameters are indicative of the presence of leaks in the cement plug.58. The method of statement 57, wherein the cement bond log or the ultrasonic log is obtained from an application database.59. The method of statement 58, further comprising storing the set of verification parameters in the application database.60. The method of statement 59, wherein the set of verification parameters stored in the application database is obtained for planning or optimising installation of a cement barrier at a different depth within the well or installation of a cement barrier in a different well.61 . The method of statement 57, wherein the ultrasonic log comprises ultrasonic data collected from an ultrasonic sensor.62. The method of statement 61 , wherein the ultrasonic sensor is the ultrasonic imaging tool.63. The method of statement 57, further comprising: if washing of the well is complete, initiating installation of cement in the annulus.64. The method of statement 57, further comprising: if washing of the well is not complete, obtaining one or more acoustic signals at a third time point; and confirming whether washing of the well is complete based a change in the one or more acoustic signals detected at the second time point and the one or more acoustic signals detected at the third time point.65. The method of statement 57, further comprising: if washing of the well is not complete, modifying one or more operational steps associated with installing cement in the annulus to at least partially compensate for incomplete washing of the well; and initiating installation of cement in the annulus according to the one or more operational steps.66. The method of statement 57, further comprising: if the control line is ablated, initiating washing of the well.67. The method of statement 57, further comprising:if the control line is not ablated, initiating ablation of the control line; obtaining one or more further ultrasonic images from the ultrasonic imaging tool; and verifying whether the control line is ablated.68. The method of statement 57, further comprising: if the control line is not ablated, modifying one or more operational steps associated with washing the well or installing cement in the annulus to at least partially compensate for presence of the control line; and initiating washing of the well or installation of cement in the annulus according to the one or more operational steps.69. A method for monitoring a Plug and Abandonment (PA) procedure for a well, wherein the PA procedure comprises washing an annulus of the well and setting a permanent plug in the well, the method comprising: determining that a control line is present in the well based on data obtained from an application database; obtaining a first set of diagnostic parameters collected by one or more sensors before ablation of the control line, wherein the one or more sensors comprises at least one ultrasonic imaging tool, and wherein the first set of diagnostic parameters comprises initial well parameters, cement bond logs or variable density logs over a target depth range of the well, or ultrasonic sensor data; determining a position of the control line based on the first set of diagnostic parameters, wherein the position comprises a location or an orientation of the control line relative to the annulus; after perforation of the control line is initiated, obtaining a second set of diagnostic parameters from the one or more sensors comprising one or more ultrasonic signals indicative of whether the control line is perforated;verifying ablation of the control line over an ablated length associated with the target depth range based on the second set of diagnostic parameters; and if ablation of the control line is complete and verified, initiating installation of cement within the annulus for setting the permanent plug.100. A computer-readable medium storing instructions for performing the method of any of statements 57-99.101 . A system comprising: a processor configured to communicate with an application database; a network of sensors comprising an ultrasonic imaging tool and an acoustic sensor; and a memory storing instructions that, when executed, cause the system to perform the method of any of statements 57-99.102. A method for validating a Plug and Abandonment (PA) procedure for a well, wherein the PA procedure comprises forming a barrier over a target depth range, the method comprising: obtaining a set of diagnostic parameters comprising gamma ray signals, cement bond logs or variable density logs over the target depth range of the well collected before the PA procedure, wherein the set of diagnostic parameters are associated with planning the PA procedure; obtaining, from a network of sensors comprising an acoustic sensor, a set of procedure parameters collected during the PA procedure, wherein the set of procedure parameters are associated with monitoring execution of the PA procedure, and wherein the set of procedure parameters comprises: a washing pressure flowrate track over the target depth range associated with detecting acoustic signals generated by a washing fluid flow, anda cementing track over the target depth range associated with installation of cement within the well; obtaining, from the network of sensors after the PA procedure is executed, a set of verification parameters comprising an acoustic track over the target depth range, and analysing the set of verification parameters to determine whether there is a leak in the barrier; based on whether a leak is detected over the target depth range, assigning the target depth range a barrier quality value indicative of whether the PA procedure was executed successfully; and outputting the barrier quality value for validation of the PA procedure.103. The method of statement 102, further outputting: the set of diagnostic parameters for planning the PA procedure; the set of procedure parameters for generating an operation plan or monitoring the PA procedure; or the set of verification parameters for verifying execution of the PA procedure.104. The method of statement 102, wherein the set of diagnostic parameters further comprises a multifinger ultrasonic track or a control line track over the target depth range, and wherein the network of sensors further comprises an ultrasonic imaging tool.105. The method of statement 102, wherein the set of procedure parameters further comprises a washing particles track over the target depth range.106. The method of statement 102, wherein, if a control line is present, the set of procedure parameters further comprises a perforation ablation track over the target depth range.107. The method of statement 102, wherein the set of verification parameters further comprises any of:a drill out and cement bond line track over the target depth range; measurements from a pressure or temperature sensor below the target depth range; or measurements from a gas permeability test of the barrier.108. The method of statement 102, wherein the barrier quality value comprises a plurality of quality values, each of quality value associated with a predetermined lengths within the target depth range, and wherein the plurality of quality values are determined based on the set of verification parameters for each predetermined length.109. The method of statement 102, wherein any of the set of diagnostic parameters, the set of procedure parameters, the set of verification parameters or the barrier quality value is stored in an application database.110. The method of statement 109, wherein stored data in the application database is used for planning, simulating, or monitoring a subsequent PA procedure.111. The method of statement 102, wherein the barrier quality value is output to a data analytics interface.112. A method for validating a Plug and Abandonment (PA) procedure for a well, wherein the PA procedure comprises forming a barrier over a target depth range, the method comprising: obtaining a set of diagnostic parameters comprising gamma ray signals, cement bond logs or variable density logs over the target depth range of the well collected before the PA procedure: using the set of diagnostic parameters for planning the PA procedure, wherein planning the PA procedure comprises determining one or more operational steps followed during execution of the PA procedure;obtaining, from a network of sensors comprising an acoustic sensor, a set of procedure parameters collected during the PA procedure, wherein the set of procedure parameters comprises: a washing pressure flowrate track over the target depth range associated with detecting acoustic signals generated by a washing fluid flow, and a cementing track over the target depth range associated with installation of cement within the well; using the set of procedure parameters to monitor execution of the PA procedure; obtaining, from the network of sensors after the PA procedure is executed, a set of verification parameters comprising an acoustic track over the target depth range; analysing the set of verification parameters to determine whether there is a leak in the barrier, wherein one or more signal peaks in the acoustic track are associated with one or more leaks; assigning the target depth range a barrier quality value based on whether a leak is detected over the target depth range, wherein the barrier quality value is indicative of whether the PA procedure was executed successfully; and outputting the barrier quality value for validation of the PA procedure.113. The method of statement 112, wherein the barrier quality value is used to improve future or ongoing PA procedures.114. The method of statement 112, wherein a machine learning module is used for determining the one or more operational steps based on a set of historical diagnostic parameters and a historical barrier quality value associated with a historical PA procedure.115. A computer-readable medium storing instructions for performing the method of any of statements 102-114.116. A system comprising: a network of sensors comprising an acoustic sensor within the well; a processor configured to obtain data from the one or more sensors and communicate with an application database; and a memory storing instructions that, when executed, cause the system to perform the method of any of statements 102-114.149. A method of validating a well annulus (101 ) washing operation, comprising the following steps: a) with an annulus washing tool (100) arranged at a perforated section of a well pipe (103), providing a washing fluid flow (110) into the annulus (101 ) of the well pipe; b) with an acoustic sensor (115, 117) arranged inside an inner bore (102) of the well pipe, detecting acoustic signals generated by the washing fluid flow (110).150. A method according to statement 149, wherein the acoustic sensor (115) is part of the annulus washing tool (100) and is, during step a), located at the location of the perforations (103a).151. A method according to statement 149 or statement 150, wherein the acoustic sensor comprises an upper acoustic sensor (117) being part of the annulus washing tool (100) above wash fluid nozzles (111 ) of the annulus washing tool; or an upper acoustic sensor (117) being supported on a string that supports the annulus washing tool (100), at a position above the annulus washing tool; wherein the upper acoustic sensor (117) during step a) is arranged inside the inner bore (102), above the location of the perforations (103a).152. A method according to one of the preceding statements 149-151 , comprising c) based on detected acoustic signals, calculating an amount of material(113) that has been washed out of the annulus (101 ) with the washing fluid flow (110).153. A method according to statement 152, further comprising the following steps: d) calculating an amount of material present in the portion of the annulus (101) before the washing operation in step a); e) calculating an annulus wash index by dividing the amount of material (113), as calculated in step c), by the amount of material as calculated in step d).154. An annulus washing assembly, comprising an annulus washing tool (100) comprising wash fluid nozzles (111) configured to provide a washing fluid flow (110), wherein the annulus washing assembly further comprises an acoustic sensor (115, 117).155. An annulus washing assembly according to statement 154, wherein the annulus washing tool (100) comprises the acoustic sensor (115).156. An annulus washing assembly according to statement 154 or statement 155, further comprising an upper acoustic sensor (117) arranged above the wash fluid nozzles (111).157. An annulus washing assembly according any one of statements 154 to 156, wherein the annulus washing tool (100) further comprising an upper wash cup (107) and a lower wash cup (109), wherein the wash fluid nozzles (111 ) are arranged between the upper and lower wash cups (107, 109), and wherein the acoustic sensor (115, 117) is arranged above the wash fluid nozzles (111 ).158. An annulus washing assembly according to any one of statements 154 to157, wherein the annulus washing assembly comprises or is connected to acomputer-readable memory unit or a computing unit programmed to calculate an annulus wash index.166. A method for simulating operation of a Plug and Abandonment (PA) procedure for a well, wherein the PA procedure comprises washing and installing cement in the well, the method comprising: obtaining, from an application database, a set of diagnostic parameters; generating a hydraulics module; generating a conveyance module; and using the hydraulics module and the conveyance module to simulate one or more operational steps of the planned PA procedure based on the set of diagnostic parameters.167. Method according to statement 166, wherein the set of diagnostic parameters comprises hydraulics parameters and conveyance parameters associated with a planned PA procedure.168. Method according to statements 166 or 167, wherein the hydraulics module is configured to perform hydraulics calculations comprising ball drop calculations, pressure drop calculations, volume calculations, or calculations associated with cementing the well.169. Method according to one of statements 166 to 168, wherein the conveyance module is configured to estimate downhole forces based on torque and drag calculations.189. A method for validating a Plug and Abandonment (PA) procedure for a well, wherein the PA procedure comprises forming a barrier over a target depth range, the method comprising: obtaining a set of diagnostic parameters, wherein the set of diagnostic parameters are associated with planning the PA procedure; obtaining, from a network of sensors, a set of procedure parameters collected during the PA procedure, wherein the set of procedure parameters are associated with monitoring execution of the PA procedure,obtaining, from the network of sensors after the PA procedure is executed, a set of verification parameters; and analysing the set of verification parameters to determine whether there is a leak in the barrier; based on whether a leak is detected over the target depth range, assigning the target depth range a barrier quality value indicative of whether the PA procedure was executed successfully; and outputting the barrier quality value for validation of the PA procedure.190. Method according to statement 189, wherein obtaining a set of diagnostic parameters comprising gamma ray signals, cement bond logs or variable density logs over the target depth range of the well collected before the PA procedure.191 . Method according to statement 189 or statement 190, wherein the network of sensors comprising an acoustic sensor.192. Method according to one of statements 189-191 , wherein the set of procedure parameters comprises: a washing pressure flowrate track over the target depth range associated with detecting acoustic signals generated by a washing fluid flow, and a cementing track over the target depth range associated with installation of cement within the well.193. Method according to one of statements 189-192, wherein the set of verification parameters comprising an acoustic track over the target depth range.197. A computer-readable medium storing instructions for performing the method of any of statements above.198. A system comprising: a network of sensors, wherein the network of sensors are configured to detect data associated with monitoring planning or operation of a well intervention procedure (e.g. a Plug and Abandonment (PA) procedure) for a well;a processor configured to obtain data from the network of sensors and communicate with an application database; and a memory storing instructions that, when executed, cause the system to perform the method of any of statements above.

[0484] Any of the above statements can be combined and the skilled person would understand such examples do not limit the potential embodiments of the present disclosure. Although one or more concepts have been explained in relation to the above embodiment, it is to be understood that many other possible modifications and variations can be made without departing from the spirit and scope.

[0485] The skilled person will also understand that any use of “or” throughout the statements of invention or description herein encompasses use of “or”, “and / or”, and “and”. For example, the term "or" within the discourse is construed to encompass both "and" and "and / or" owing to its inherent inclusivity. Within linguistic reasoning, "or" denotes an inclusive disjunction, allowing for the consideration of scenarios wherein either one condition holds true, the other condition holds true, or both conditions hold true concurrently. This interpretation inherently incorporates the conjunction "and", permitting the acknowledgment of scenarios wherein multiple conditions coexist. Additionally, the term "and / or" explicitly acknowledges the possibility of either condition being singularly true or both conditions being true simultaneously, thus aligning with the broader meaning of "or" within the context of this disclosure. Consequently, "or" functions as a flexible connector within the statements of invention, accommodating both exclusive and inclusive interpretations to suit the nuanced requirements of embodiments described herein.Having described example embodiments of the invention it will be apparent to those skilled in the art that other embodiments incorporating the concepts may be used. These and other examples illustrated above are intended by way of example only and the actual scope of the invention is to be determined from the following claims.

Claims

CLAIMS1 . A method for validating a well intervention procedure for a well, wherein the well intervention procedure comprises forming a barrier over a target depth range, the method comprising: obtaining a set of diagnostic parameters, wherein the set of diagnostic parameters are associated with planning the well intervention procedure; obtaining, from a network of sensors, a set of procedure parameters collected during the well intervention procedure, wherein the set of procedure parameters are associated with monitoring execution of the well intervention procedure, obtaining, from the network of sensors after the well intervention procedure is executed, a set of verification parameters; and analysing the set of verification parameters to determine whether there is a leak in the barrier; based on whether a leak is detected over the target depth range, assigning the target depth range a barrier quality value indicative of whether the well intervention procedure was executed successfully; and outputting the barrier quality value for validation of the well intervention procedure.

2. Method according to claim 1 , wherein obtaining a set of diagnostic parameters comprising obtaining initial well parameters over the target depth range of the well.

3. Method according to claim 1 or claim 2, wherein obtaining a set of diagnostic parameters comprising obtaining cement bond logs over the target depth range of the well.

4. Method according to one of claims 1-3, wherein obtaining a set of diagnostic parameters comprising obtaining variable density logs over the target depth range of the well.

5. Method according to one of claims 1-4, wherein obtaining a set of diagnostic parameters comprising obtaining hydraulics parameters and conveyance parameters.

6. Method according to one of claims 1-5, wherein the network of sensors comprising an acoustic sensor.

7. Method according to one of claims 1-6, wherein the set of procedure parameters comprises a washing pressure flowrate track over the target depth range.

8. Method according to one of claims 1-7, wherein the set of procedure parameters comprises a cementing track over the target depth range.

9. Method according to one of claims 1-8, wherein the set of verification parameters comprising an acoustic track over the target depth range.

10. Method according to one of claims 1-9, wherein the network of sensors further comprises an ultrasonic logging tool.11 . Method according to one of claims 1-10, wherein the set of verification parameters comprising an acoustic track over the target depth range associated with detecting acoustic signals generated by a washing fluid flow.

12. The method according to one of claims 1-11 , wherein, if a control line is present, the set of procedure parameters comprises a perforation ablation track over the target depth range.

13. The method according to one of claims 1-12, wherein the procedure parameters for the perforation ablation track over the target depth range comprises:- direct control line detection through holes or windows in the tubing I casing using ultrasonic images or ultrasonic measurements, and I or- control line detection through the tubing / casing using an acoustic sensor.

14. The method according to one of claims 1-13, wherein the set of diagnostic parameters further comprises any of:- a multifinger caliper track over the target depth range, or- ultrasonic measurement track over the target depth range, or- a control line track over the target depth range.

15. The method according to one of claims 1-14, wherein the set of procedure parameters further comprises a washing particles track over the target depth range.

16. The method according to one of claims 1-15, further outputting one or more of: the set of diagnostic parameters for planning the well intervention procedure; the set of procedure parameters for generating an operation plan or monitoring the well intervention procedure; or the set of verification parameters for verifying execution of the well intervention procedure.

17. The method of according to one of claims 1-16, wherein the set of verification parameters further comprises any of:- a drill out and cement bond line track over the target depth range;- measurements from a pressure or temperature sensor below the target depth range or below the barrier;- validation of barrier integrity derived from one or more acoustic signals across the barrier; or- measurements from a gas permeability test of the barrier.

18. The method according to one of claims 1-17, wherein the barrier quality value comprises a plurality of quality values, each of the quality values associated with a predetermined length within the target depth range.

19. The method according to one of claims 1-18, and wherein the plurality of quality values are determined based on the set of verification parameters for each predetermined length.

20. The method according to one of claims 1-19, wherein the barrier quality value is quality parameter track is calculated based on values calculated for theexecution of the barrier setting and verification values for the set barrier along the barrier length.21 . The method according to one of claims 1-20, wherein the barrier quality value is a number of tracks for the parameters obtained before, during and after the barrier setting.

22. The method according to one of claims 1-21 , wherein the barrier quality value represents a visualization of the parameters.

23. The method according to one of claims 1-22, wherein the well intervention procedure is a Plug and Abandonment (PA) procedure.

24. A method for simulating operation of a well intervention procedure for a well, wherein the well intervention procedure comprises washing and installing cement in the well, the method comprising: obtaining, from an application database, a set of diagnostic parameters; obtaining a hydraulics module; obtaining a conveyance module; and using the hydraulics module and the conveyance module to simulate one or more operational steps of the planned well intervention procedure based on the set of diagnostic parameters.

25. Method according to claim 24, wherein the set of diagnostic parameters comprises hydraulics parameters and conveyance parameters associated with a planned well intervention procedure.

26. Method according to claim 24 or 25, wherein the hydraulics module is configured to perform hydraulics calculations comprising ball drop calculations, pressure drop calculations, volume calculations, or calculations associated with cementing the well.

27. Method according to one of claims 24 - 26, wherein the conveyance module is configured to estimate downhole forces based on torque and drag calculations.

28. The method of one of claims 24 - 27, further comprising: obtaining a computational fluid dynamics (CFD) module configured to perform further hydraulics calculations comprising any of: cement slumping calculations, washing efficiency calculations, listing of particles calculations, multiple annuli calculations, or calculations associated with tubular, tubing, or casing designs; and using the CFD module to simulate one or more operational steps of the planned well intervention procedure.

29. The method of one of claims 24 - 28, further comprising: generating a perforation ablation module configured to verify correct perforation and ablation or determine mechanical perforation and ablation parameters based on e.g. geometry of perforations, perforation depth, or perforation gun setup; and using the perforation ablation module to simulate one or more operational steps of the planned well intervention procedure.

30. The method of one of claims obtaining 24 - 29, wherein simulation of the one or more operational steps is used to generate or verify an operation plan comprising operational steps used for planning the well intervention procedure.31 . The method of one of claims 24 - 30, wherein simulation of the one or more operational steps is used to generate or verify an operational envelope comprising a range of operation parameters for performing the well intervention procedure optimally.

32. The method of one of claims 24 - 31 , further comprising comparing the operation plan to a monitored well intervention procedure to determine an operational status of the monitored well intervention procedure.

33. The method of one of claims 24 - 32, wherein a machine learning module comprises the hydraulics module and the conveyance module such that the machine learning module simulates operation of the well intervention procedure.

34. The method of claim 24 - 33, wherein the machine learning module comprises a generative artificial intelligence (Al) algorithm configured to generate the operation plan.

35. The method of one of claims 24 - 34, wherein a data analytics interface initiates simulation of the one or more operational steps, and is configured to display the simulation of the one or more operational steps.

36. The method of one of claims 24 - 35, further comprising storing simulation of the one or more operational steps associated with the set of diagnostic parameters in the application database.

37. The method of one of claims 24 - 36, further comprising identifying faults or risks in the well or well intervention PA procedure based on simulation of the one or more operational steps.

38. The method of one of claims 24 - 37, wherein hydraulics parameters and conveyance parameters are obtained from one or more sensors associated with the well.

39. The method according to one of claims 24-38, wherein the well intervention procedure is a Plug and Abandonment (PA) procedure.

40. Method for a barrier setting in a wellbore, the method comprising:- planning the barrier setting,- simulating the planned barrier setting,- creating an operation plan for the barrier setting;- performing a risk assessment for the operation plan;- executing the operation plan setting the barrier in the wellbore; and- verifying the barrier in the wellbore.41 . Method for a barrier setting in a wellbore, the method comprising:- executing an operation plan setting the barrier in the wellbore; and- verifying the barrier in the wellbore.

42. Method according to claim 40 or claim 41 , further comprising calculating a quality of the barrier in the wellbore.

43. Method according to claim 42, further comprising calculating the barrier quality over the barrier length.

44. Method according to claim 43, wherein the barrier length corresponds to an ablated length of the wellbore, preferably the ablated length of the wellbore versus total ablated length.

45. Method according to one of claims 40-44, wherein executing the operation plan further comprises ablating / severing a control line in an annulus of the wellbore and confirming ablation / severing of the control line.

46. Method according to one of claims 40-45, further comprising calculating a barrier quality for a length of the barrier.

47. Method according to one of claims 40-46, further comprising displaying a barrier quality for a length of the barrier.

48. Method according to one of claims 40-47, further comprising performing diagnostics of the well before simulation.

48. Method in a barrier setting in a wellbore, the method comprising:- detecting a control line location and orientation in the wellbore;- ablating the control line over at least a length of the wellbore;- verifying ablation of the control line over the ablated length of the wellbore.

49. Method according to claim 48, wherein verifying the ablated length of the wellbore comprises verifying the ablated length versus total ablated length.

50. Method according to claim 48 or 49, wherein determining location and orientation of the control line is based on ultrasonic imaging or ultrasonic measurements or an acoustic log.51 . Method according to one of claims 48-50, further comprising verifying ablation of the control line by performing downhole diagnostics.

52. Method according to one of claims 48-51 , further comprising verifying ablation of the control line by ultrasonic logging.

53. Method according to one of claims 48-52, further comprising verifying ablation of the control line using an imaging tool.

54. Method in a barrier setting in a wellbore, the method comprising:- washing an annulus; and- detecting by use of acoustics, a particle flow in the annulus during washing.

55. Method according to claim 54, further comprising calculating a total mobilized mass of particles in the annulus based on the detected particle flow.

56. Method for establishing a quality of a barrier setting in a wellbore, the method comprising calculating at least one quality parameter for the barrier based on at least one parameter obtained during the barrier setting and at least one verification parameter obtained in a verification of the barrier.

57. Method according to claim 56, wherein the at least on parameter obtained during the barrier setting comprising cementing flow rate and pressure.

58. Method according to one of claims 54-57, wherein the at least one parameter obtained during the barrier setting is related to ablation of a control line along the barrier.

59. Method according to one of claims 54-58, wherein the at least one parameter obtained during the barrier setting is washing pressure.

60. Method according to one of claims 54-59, wherein the at least one parameter obtained during the barrier setting is related to monitoring of washing particles during the washing of the wellbore.61 . Method according to one of claims 54-60, wherein the at least one quality parameter of the barrier is calculated for a number of intervals over the barrier length.

62. Method according to one of claims 54-61 , wherein the at least one quality parameter is displayed in intervals along the depth of the wellbore.

63. A product for qualification of a barrier setting in a well comprising:- a number of tracks for parameters obtained before, during and after the barrier setting in intervals along a depth of the wellbore, and- a calculated quality parameter track representing a quality of the barrier in intervals along the depth of the wellbore.

64. The product according to claim 63, wherein the parameters obtained before the barrier setting comprising parameters related to at least one of gamma ray log, casing collar locator log, cement bond log, variable density log.

65. The product according to claim 63 or 64, wherein the parameters obtained before the barrier setting comprising parameters related control line and clamp location and orientation.

66. The product according to one of claims 63-65, wherein the parameters obtained during the barrier setting comprising parameters related to cementing flow rate and pressure, preferably cementing flow rate and pressure versus set parameters.

67. The product according to one of claims 63-66, wherein the parameters obtained during the barrier setting comprising parameters related to detectedparticles during a washing operation, preferably estimated particle mass lifted from an annulus versus particles in place in the annulus before washing.

68. The product according to one of claims 63-67, wherein the parameters obtained during the barrier setting comprising parameters related to ablation of control line over the length of the barrier.

69. The product according to one of claims 63-68, wherein the parameters obtained after the barrier setting comprising parameters related to acoustic monitoring of the barrier along the barrier length to verify the barrier.

70. The product according to one of claims 63-69, wherein the quality parameter track is calculated based on values calculated for the execution of the barrier setting and verification values for the set barrier along the barrier length.71 . The product according to one of claims 63-706, wherein the number of tracks for the parameters obtained before, during and after the barrier setting and the calculated quality parameter track represent a visualization of the parameters.

72. Method for a barrier setting in a wellbore, the method comprising:- logging the wellbore;- ablating a control line in the wellbore;- confirming ablation of the control line;- washing the wellbore;- placing a fluidized plugging material in the wellbore creating a barrier in the wellbore; and- verifying the barrier.

73. Method for a barrier setting in a wellbore, the method comprising:- logging the wellbore;- washing the wellbore;- placing a fluidized plugging material in the wellbore creating a barrier in the wellbore; and- verifying the barrier.

74. Method according to claim 72 or 73, further comprising determining that a control line is present in the wellbore based on an ultrasonic log.

75. Method according to claim 74, wherein the ultrasonic log comprising:- obtaining ultrasonic images or ultrasonic measurements from an ultrasonic tool within the well after ablation of the control line is initiated; or- verifying ablation of the control line over an ablated length based on the ultrasonic images or ultrasonic measurements associated with perforations of the control line; or- a phased-array oriented beam.

76. Method according to claim 72 or 73, further comprising determining that a control line is present in the wellbore based on use of an acoustic sensor.

77. Method for determining a presence of a control line in a wellbore using an ultrasonic log.

78. Method for confirming ablation of a control line in the wellbore using am ultrasonic log.

79. Method according to claim 77 or 78, wherein ultrasonic log comprising:- obtaining ultrasonic images or ultrasonic measurements from an ultrasonic tool within the well after ablation of the control line is initiated; or- verifying ablation of the control line over an ablated length based on the ultrasonic images or ultrasonic measurements associated with perforations of the control line; or- a phased-array oriented beam.

80. A method for controlling a well intervention procedure for a well, wherein the well intervention procedure comprises washing and setting a permanent plug in the well to form a barrier, the method comprising: obtaining a set of diagnostic parameters from a network of sensors or an application database;using the set of diagnostic parameters to plan the well intervention procedure, determining one or more operational steps; performing a simulation of the one or more operational steps, analysing the simulation to generate an operation plan comprising the one or more operational steps; monitoring the well intervention procedure using data collected from the network of sensors to determine an operation status of the well intervention procedure; validating the well intervention procedure based on determining a barrier quality value, wherein the barrier quality is associated with whether there is a leak in the barrier; and storing, in the application database, data associated with execution of the well intervention procedure.81 . Method according to claim 80, wherein the set of diagnostic parameters comprises at least one of initial well parameters, cement bond logs or variable density logs over a target depth range of the well before the well intervention procedure.

82. Method according to claim 80 or 81 , wherein performing the simulation comprises performing hydraulics calculations and estimating downhole forces based on the set of diagnostic parameters.

83. Method according to one of claims 80-82, further comprising monitoring the well intervention procedure once the well intervention procedure is initiated according to the operation plan.

84. Method according to one of claims 80-83, wherein the network of sensors comprise an ultrasonic imaging tool and / or an acoustic sensor.

85. Method according to one of claims 80-84, wherein the well intervention procedure is a Plug and Abandonment (PA) procedure.

86. A method for monitoring operation of a well intervention procedure for a well, the method comprising: obtaining, from a network of sensors a first set of parameters associated with the well; generating one or more processed parameters by performing one or more processing steps on each parameter of the first set of parameters, the processing steps comprising: merging two or more parameters of the first set of parameters, resulting in data fusion; filtering one or more parameters of the first set of parameters; or extracting a target feature from one or more parameters of the first set of parameters; predicting an estimated state of the well intervention procedure based on the one or more processed parameters; obtaining, from the network of sensors, a second set of parameters associated with monitoring the well after the well intervention procedure is initiated; generating a current state of the well intervention procedure based on the second set of parameters; and determining an operation status based on comparing the estimated state of the well intervention procedure to the current state of the well intervention procedure.

87. Method according to claim 86, further comprising at least one acoustic sensor.

88. Method according to claim 86 or 87, wherein the second set of parameters comprises one or more acoustic signals.

89. A system for diagnosing a well intervention operation comprising an artificial neural network (ANN), wherein the system is configured to perform the following steps: obtain sensor data from a network of sensors, the network of sensors comprising one or more surface sensors and one or more downhole sensors;obtain historical data from an application database, the historical data comprising historical fault data or trend data associated with previous well intervention operations; extract one or more relevant features from the data from the sensor data, wherein relevant features are determined based on the sensor data; use the ANN, trained using the historical data, to identify one or more conditions indicators based on the one or more relevant features, wherein the one or more conditions indicators are associated with performing the well intervention operation; determine a predicted state of the well intervention operation based on the one or more conditions indicators; and predict whether a failure mode is probable based on comparing the predicted state of the well intervention operation to a current state of the well intervention operation, the current state of the well intervention operation based on either: a protocol for the well intervention operation, the protocol comprising a plurality of planned operational steps or instructions to perform the well intervention operation; or updated sensor data from the network of sensors.

90. System according to claim 89, wherein the network of sensors comprising one or more surface sensors and one or more downhole sensors.91 . System according to claim 89 or 90, wherein the historical data comprising historical fault data or trend data associated with previous well intervention operations.

92. System according to one of claims 89-91 , wherein relevant features are determined based on the sensor data.

93. System according to one of claims 89-92, wherein the one or more conditions indicators are associated with performing the well intervention operation.

94. A method for monitoring a well intervention procedure for a well, wherein the well intervention procedure comprises washing an annulus of the well and setting a permanent plug in the well, the method comprising: determining that a control line is present in the well; verifying ablation of the control line over an ablated length; after washing of the annulus is initiated, obtaining one or more acoustic signals from a first acoustic sensor within the well; confirming whether washing of the well is complete based on a change in the one or more detected acoustic signals; and obtaining a set of verification parameters after installation of cement in the annulus is initiated, wherein the set of verification parameters are indicative of the presence of leaks in the cement plug.

95. Method according to claim 94, further comprising determining that a control line is present in the well based on a cement bond log or an ultrasonic log or use of an acoustic sensor.

96. Method according to claim 94 or 95, further comprising after ablation of the control line is initiated, obtaining one or more ultrasonic images or measurements from an ultrasonic tool within the well.

97. Method according to one of claims 94-96, further comprising verifying ablation of the control line over an ablated length based on the ultrasonic images or measurements associated with perforations of the control line.

98. Method according to one of claims 94-97, further comprising after washing of the annulus is initiated, obtaining one or more acoustic signals from a first acoustic sensor within the well at a first time point and a second time point.

99. Method according to one of claims 94-98, further comprising confirming whether washing of the well is complete based on a change in the one or more acoustic signals detected at the first time point and the one or more acoustic signals detected at the second time point.

100. Method according to one of claims 94-99, further comprising obtaining a set of verification parameters from a second acoustic sensor after installation of cement in the annulus is initiated.101 . A method for monitoring a well intervention procedure for a well, wherein the well intervention procedure comprises washing an annulus of the well and setting a permanent plug in the well, the method comprising: determining that a control line is present in the well based on data obtained from an application database; obtaining a first set of diagnostic parameters collected by one or more sensors; determining a position of the control line based on the first set of diagnostic parameters, wherein the position comprises a location and / or an orientation of the control line; verifying ablation of the control line over an ablated length associated with the target depth range based on a second set of diagnostic parameters; and if ablation of the control line is complete and verified, initiating installation of cement within the annulus for setting the permanent plug.

102. Method according to claim 101 , further comprising obtaining a first set of diagnostic parameters collected by one or more sensors before ablation of the control line, wherein the one or more sensors comprises at least one ultrasonic tool.

103. Method according to one of claims 101 -102, wherein the first set of diagnostic parameters comprises initial well parameters, cement bond logs or variable density logs over a target depth range of the well, or ultrasonic sensor data.

104. Method according to one of claims 101 -103, wherein after perforation of the control line is initiated, obtaining a second set of diagnostic parameters from the one or more sensors comprising one or more ultrasonic signals indicative of whether the control line is perforated.

105. A method for validating a well intervention procedure for a well, wherein the well intervention procedure comprises forming a barrier over a target depth range, the method comprising: obtaining a set of diagnostic parameters; using the set of diagnostic parameters for planning the well intervention procedure, wherein planning the well intervention procedure comprises determining one or more operational steps followed during execution of the well intervention procedure; obtaining, from a network of sensors, a set of procedure parameters collected during the well intervention procedure; using the set of procedure parameters to monitor execution of the well intervention procedure.

106. Method according to claim 105, wherein the set of verification parameters comprising an acoustic track over the target depth range.

107. Method according to claim 105 or 106, wherein one or more signal peaks in the acoustic track are associated with one or more leaks.

108. Method according to any of claims 86 - 107, wherein the well intervention procedure is a Plug and Abandonment (PA) procedure.

109. A computer-readable medium storing instructions for performing the method of any of claims 1 - 108.

110. A system comprising: a network of sensors, wherein the network of sensors are configured to detect data associated with monitoring planning or operation of a well intervention procedure for a well; a processor configured to obtain data from the network of sensors and communicate with an application database; and a memory storing instructions that, when executed, cause the system to perform the method of any of claims 1 - 108.

111. System according to claim 110, wherein the well intervention procedure is a Plug and Abandonment (PA) procedure.