A downhole multi-parameter cooperative oil and gas drilling leakage layer positioning method and system

Abstract

The application belongs to the technical field of oil and gas drilling, and provides a downhole multi-parameter cooperative oil and gas drilling lost layer positioning method and system, which comprises the following steps: acquiring annulus pressure data at the bottom of a well in the drilling process; identifying a lost circulation according to the acquired annulus pressure data at the bottom of the well and a primary lost circulation identification network, and acquiring an annulus pressure drop at the bottom of the well; when the lost circulation is identified, a drilling tool is lifted to acquire a drilling fluid flow rate and a temperature along the well depth; according to the acquired drilling fluid flow rate, the temperature and a secondary lost layer positioning network, a lost probability distribution corresponding to each well depth point along the well depth direction is obtained; and the position of a lost layer is determined according to the obtained lost probability distribution, so as to realize the downhole multi-parameter cooperative oil and gas drilling lost layer positioning.

Description

Technical Field

[0001] This invention belongs to the field of oil and gas drilling technology, specifically relating to a method and system for locating lost zones in oil and gas drilling using a multi-parameter coordinated approach. Background Technology

[0002] The statements in this section are merely background information related to the present invention and do not necessarily constitute prior art.

[0003] The phenomenon of drilling fluid flowing into the formation due to excessively high formation permeability or large porosity is called well leakage, which is one of the most common downhole complexities during drilling. Once leakage occurs, the seepage of drilling fluid into the formation not only causes drilling fluid loss and operational interruption but can also lead to serious accidents such as wellbore instability and blowouts. Therefore, quickly and accurately identifying leakage and locating the leakage zone is crucial for subsequent plugging design, construction decisions, and risk control.

[0004] During oil and gas well drilling, timely identification of lost circulation (LOC) is crucial for determining whether drilling fluid loss has occurred downhole, providing a basis for adjusting pump flow rate and controlling wellbore pressure. It also allows for the timely determination of the location of the lost circulation zone, supporting the selection of plugging materials and the development of plugging measures. Therefore, the timeliness of LOC identification and the accuracy of LOC location jointly determine the effectiveness of on-site LOC treatment.

[0005] Currently, well leakage detection at drilling sites largely relies on comprehensive logging parameters (such as mud pit volume, inlet and outlet flow rates, etc.); however, logging parameters are affected by wellbore circulation lag, resulting in poor timeliness of monitoring results and a certain lag in identification, which increases the risk of wellbore instability and well control. Existing methods for locating lost circulation zones in oil and gas drilling mostly rely on single-parameter anomalies exhibited by the wellbore and formation during the loss process, such as wellbore temperature, pressure, drilling fluid velocity, and changes in formation structure after the loss occurs. However, the temperature field is greatly affected by wellbore heat transfer conditions, circulation conditions, and heat conduction. When the degree of loss is weak or the loss range is small, the temperature anomaly response characteristics are often not significant enough, resulting in insufficient location accuracy and stability. The standpipe pressure variation method is highly dependent on wellbore structural parameters and fluid parameters, and these parameters are difficult to estimate accurately, leading to low location accuracy. The annular pressure anomaly analysis response is easily affected by pump pressure fluctuations and noise interference from complex operating conditions, resulting in weak anti-interference capabilities. The transient pressure wave method has high requirements for sensor sampling accuracy and signal quality, and is currently mostly in the experimental verification stage. Factors such as local fluid disturbances in the well, changes in solid particle concentration, and measurement noise may also cause abnormal flow velocity fluctuations, affecting the robustness of the location results.

[0006] Loss of formation can lead to changes in formation structure, which can manifest as specific anomalies in well logging responses. However, these are mostly post-drilling or post-operation measurement and analysis methods, which are insufficient to meet the needs for real-time identification and rapid handling of lost formations during drilling.

[0007] Therefore, existing field well leakage identification methods mostly rely on surface logging parameters, which are not direct enough in responding to downhole leakage, and the identification results often have a certain lag. The location of the leakage layer mostly relies on the local characteristics of a single parameter or the change threshold of a small number of parameters, which is easily affected by complex downhole operating conditions, resulting in inaccurate location of the leakage layer and affecting the effectiveness of subsequent plugging measures. Summary of the Invention

[0008] To address the aforementioned issues, this invention proposes a multi-parameter collaborative method and system for locating lost circulation zones in oil and gas drilling. The method identifies lost circulation based on the decrease in bottom hole annular pressure during drilling, promptly identifies lost circulation, and extracts the annular pressure drop after identification. When lost circulation is detected, drilling fluid velocity, velocity gradient, drilling fluid temperature, and temperature gradient are used as characteristics for lost circulation zone location. Furthermore, a physical loss function is introduced during the location process to constrain the annular pressure response consistency of the location results, thus achieving accurate location of the lost circulation zone.

[0009] According to some embodiments, the first aspect of the present invention provides a downhole multi-parameter coordinated method for locating lost circulation zones in oil and gas drilling, employing the following technical solution: A downhole multi-parameter coordinated method for locating lost circulation zones in oil and gas drilling includes: Acquire bottom hole annular pressure data during the drilling process; Based on the acquired bottom hole annular pressure data and the first-level well leakage identification network, well leakage is identified, and the amount of bottom hole annular pressure drop is obtained. When well leakage is detected, the drill string is pulled up to obtain the drilling fluid flow rate and temperature along the well depth; Based on the obtained drilling fluid flow rate, temperature, and secondary lost circulation zone location network, the loss probability distribution corresponding to each well depth point along the well depth direction is obtained. The location of the lost circulation layer is determined based on the obtained loss probability distribution, enabling multi-parameter coordinated location of lost circulation layers in oil and gas drilling.

[0010] As a further technical limitation, the first-level well leakage identification network adopts a bidirectional gated cyclic unit network. Based on the bidirectional gated cyclic unit network, the pressure drop characteristics of the acquired bottom hole annulus pressure data are extracted. Based on the extracted pressure drop characteristics, it is determined whether oil and gas drilling has lost power, thus completing the well leakage identification.

[0011] As a further technical limitation, the secondary leakage layer location network adopts a bidirectional gated circulation unit network. Based on the bidirectional gated circulation unit network, the drilling fluid flow rate and temperature of well leakage are extracted, and the leakage interval is located according to the extracted multi-channel features.

[0012] As a further technical limitation, the secondary lost zone location network is trained using a total loss function that includes a location loss function and an annular pressure response consistency loss function. The location loss function is used to constrain the consistency between the lost zone probability distribution and the actual lost zone label, and the annular pressure response consistency loss function is used to constrain the consistency between the theoretical annular pressure drop corresponding to the lost zone and the actual annular pressure drop, so that the lost zone simultaneously meets the requirements of local flow velocity, temperature anomaly characteristics and overall annular pressure response consistency, thereby obtaining the lost zone location and realizing downhole multi-parameter coordinated oil and gas drilling lost zone location.

[0013] As a further technical limitation, in the process of obtaining drilling fluid flow rate and temperature along the well depth, the drill string is raised to obtain drilling fluid temperature and flow rate along the well depth; during the process of raising the drill string, a fixed depth stop measurement method is adopted, that is, the drill string is raised a certain distance and then stopped for a period of time. After the flow rate and drilling fluid temperature data are collected at the current measurement point, it is raised to the next measurement point.

[0014] As a further technical limitation, in the process of obtaining drilling fluid flow, based on the Doppler ultrasonic velocimeter and the Doppler effect, the flow velocity of solid particles in the drilling fluid is calculated by measuring the echo frequency shift caused by solid particles in the drilling fluid in the wellbore, so as to determine the drilling fluid flow velocity.

[0015] According to some embodiments, a second aspect of the present invention provides a downhole multi-parameter coordinated oil and gas drilling lost circulation zone location system, employing the following technical solution: A downhole multi-parameter coordinated oil and gas drilling lost circulation zone location system includes: The first acquisition module is configured to acquire bottom hole annular pressure data during the drilling process; The first identification module is configured to identify well leakage and obtain the amount of well leakage drop based on the acquired bottom hole annular pressure data and the first-level well leakage identification network. The second acquisition module is configured to, when well leakage is detected, lift the drill string to acquire the drilling fluid flow rate and temperature along the well depth; The second identification module is configured to obtain the leakage probability distribution corresponding to each well depth point along the well depth direction based on the obtained drilling fluid flow rate, temperature and secondary lost circulation layer location network. The location module is configured to determine the location of the lost layer based on the obtained loss probability distribution, thereby achieving multi-parameter coordinated location of lost layers in oil and gas drilling.

[0016] According to some embodiments, a third aspect of the present invention provides a computer-readable storage medium, employing the following technical solution: A computer-readable storage medium having a program stored thereon, which, when executed by a processor, implements the steps in the downhole multi-parameter coordinated oil and gas drilling lost zone location method as described in the first aspect of the present invention.

[0017] According to some embodiments, the fourth aspect of the present invention provides an electronic device, which adopts the following technical solution: An electronic device includes a memory, a processor, and a program stored in the memory and running on the processor, wherein the processor executes the program to implement the steps in the downhole multi-parameter coordinated oil and gas drilling lost zone location method as described in the first aspect of the present invention.

[0018] According to some embodiments, the fifth aspect of the present invention provides a computer program product, which adopts the following technical solution: A computer program product includes software code, wherein the program in the software code performs the steps of the downhole multi-parameter coordinated oil and gas drilling lost zone location method as described in the first aspect of the present invention.

[0019] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention utilizes downhole annular pressure to directly identify leakage occurrences, reducing identification lag caused by the cyclic hysteresis of surface logging parameters. The annular pressure drop obtained in the first-stage identification phase is introduced as a physical consistency constraint into the second-stage leakage layer location network, so that the location results simultaneously satisfy the characteristics of local flow velocity and temperature anomalies and the overall wellbore flow response law, thereby suppressing false anomalies caused by local disturbances or measurement noise and improving the accuracy, reliability and interpretability of leakage layer location. Attached Figure Description

[0020] The accompanying drawings, which form part of this embodiment, are used to provide a further understanding of this embodiment. The illustrative embodiments and their descriptions are used to explain this embodiment and do not constitute an improper limitation of this embodiment.

[0021] Figure 1 This is a flowchart of the downhole multi-parameter coordinated oil and gas drilling lost zone location method in Embodiment 1 of the present invention; Figure 2 This is a schematic diagram of the steps in the downhole multi-parameter coordinated oil and gas drilling lost zone location method in Embodiment 1 of the present invention; Figure 3 This is a schematic diagram of the gated loop unit in Embodiment 1 of the present invention; Figure 4 This is a schematic diagram of the bidirectional gated loop unit network in Embodiment 1 of the present invention; Figure 5 This is a schematic diagram of the structure of the primary well leakage detection network in Embodiment 1 of the present invention; Figure 6This is a schematic diagram of the structure of the secondary leakage layer location network in Embodiment 1 of the present invention; Figure 7 This is a schematic diagram illustrating the division of the flow velocity sequence within a window in Embodiment 1 of the present invention; Figure 8 This is a schematic diagram of the annular pressure during the drilling process in Embodiment 1 of the present invention; Figure 9 This is a schematic diagram of drilling fluid flow rate and temperature data during the lifting process in Embodiment 1 of the present invention. Figure 10 This is a structural block diagram of the downhole multi-parameter coordinated oil and gas drilling lost zone location system in Embodiment 2 of the present invention. Detailed Implementation

[0022] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0023] It should be noted that the following detailed descriptions are exemplary and intended to provide further illustration of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0024] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of exemplary embodiments according to the invention. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0025] In this invention, terms such as "upper," "lower," "left," "right," "front," "back," "vertical," "horizontal," "side," and "bottom" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. These terms are used only to facilitate the description of the structural relationships of the various components or elements of this invention and do not specifically refer to any component or element in this invention. They should not be construed as limiting the invention.

[0026] In this invention, terms such as "fixed connection," "connected," and "linked" should be interpreted broadly, indicating a fixed connection, an integral connection, or a detachable connection; a direct connection or an indirect connection through an intermediate medium. Those skilled in the art can determine the specific meaning of these terms in this invention based on the specific circumstances, and they should not be construed as limitations on the invention.

[0027] Where there is no conflict, the embodiments and features in the embodiments of the present invention can be combined with each other.

[0028] Example 1 Embodiment 1 of this invention introduces a method for locating lost circulation zones in oil and gas drilling using a multi-parameter coordinated approach in downhole.

[0029] like Figure 1 The method for locating lost circulation zones in oil and gas drilling using a multi-parameter coordinated approach, as shown, includes: Acquire bottom hole annular pressure data during the drilling process; Based on the acquired bottom hole annular pressure data and the first-level well leakage identification network, well leakage is identified, and the amount of bottom hole annular pressure drop is obtained. When well leakage is detected, the drill string is pulled up to obtain the drilling fluid flow rate and temperature along the well depth; Based on the obtained drilling fluid flow rate, temperature, and secondary lost circulation zone location network, the loss probability distribution corresponding to each well depth point along the well depth direction is obtained. The location of the lost circulation layer is determined based on the obtained loss probability distribution, enabling multi-parameter coordinated location of lost circulation layers in oil and gas drilling.

[0030] Existing well leakage identification methods often rely on surface logging parameters, resulting in response lags. Furthermore, the location of lost circulation zones often depends on single-parameter anomalies or fixed thresholds for multiple parameters, making them susceptible to fluctuations in downhole measurement data and noise interference. This can lead to misjudgments and missed detections of leakage zones. Figure 2 As shown, this embodiment first identifies lost circulation (LOB) by utilizing the decrease in annular pressure at the bottom of the well during drilling, thus improving the timeliness of LOB identification. Once LOB is identified, the location of the lost circulation zone is determined. Drilling fluid velocity, velocity gradient, drilling fluid temperature, and temperature gradient anomalies are used as characteristics for locating the lost circulation zone, identifying the LOB interval. Furthermore, since downhole velocity and temperature measurements may be affected by local fluid disturbances, changes in solid particle concentration, and sensor measurement noise, velocity fluctuations or temperature gradient anomalies may occur in non-lost circulation areas, potentially leading to misjudgments of the LOB interval. Therefore, this embodiment utilizes the annular pressure decrease obtained in the first-stage identification phase to construct an annular pressure response consistency loss function, constraining the training process of the second-stage lost circulation zone location network. This ensures that the identified LOB intervals simultaneously meet the requirements of local parameter anomalies and overall annular pressure response consistency, thereby improving the reliability of the location results.

[0031] As one or more implementation methods, this embodiment uses a downhole measurement unit installed on the drill collar to collect annular pressure, drilling fluid velocity, and temperature in real time during the drilling process. During the drilling phase, the measured annular pressure data is used to identify lost circulation. After a lost circulation is detected, the location of the lost circulation zone is output using drilling fluid temperature and velocity data collected along the well depth by raising the drill string.

[0032] It should be noted that, in order to ensure the accuracy and stability of temperature and flow rate data measurement, a fixed-depth stop measurement method is adopted for data collection during the lifting process. That is, the drill string is stopped for a period of time every 5m, and after the flow rate and drilling fluid temperature data are collected at the current measuring point, it is lifted to the next measuring point.

[0033] Drilling fluid velocity was obtained using a Doppler ultrasonic velocimeter. Based on the Doppler effect, the velocity of solid particles in the drilling fluid was calculated by measuring the echo frequency shift caused by solid particles in the drilling fluid within the wellbore, thus determining the drilling fluid velocity. Temperature was obtained using a Pt1000 platinum resistance temperature sensor. The measurement principle is that the resistance value changes regularly with temperature. Pressure data was obtained using a pressure measuring instrument, which can be a silicon-titanium sapphire pressure sensor.

[0034] Doppler velocimeters measure flow based on the Doppler effect. When ultrasonic waves are emitted into flowing drilling fluid, solid particles and bubbles in the fluid scatter the waves, generating echo signals. Due to the relative motion of the scattering particles with respect to the ultrasonic probe, the echo frequency shifts compared to the transmission frequency. By extracting this frequency shift and combining it with the installation angle of the ultrasonic probe, the annular drilling fluid velocity can be calculated. The Doppler ultrasonic transceiver probe is installed at a 45° incident angle, and its flow velocity... The calculation formula is: ; in, This represents the propagation speed of ultrasound in drilling fluid. This is the frequency at which the ultrasonic wave is emitted. The frequency shift is caused by the relative motion between the solid particles and the instrument. The angle of incidence is denoted as .

[0035] Because downhole data acquisition is susceptible to noise interference, the raw measurement data needs to be denoised. In this embodiment, the denoising method employs the 3σ criterion to reduce the interference of abnormal data on subsequent network recognition results. Since different monitoring parameters have different feature scales and dimensions, and deep neural networks are sensitive to the scale of input features, these input feature data are normalized using a minimax method. The normalization formula is as follows: ; in, This represents the normalized data. Represents the original data; , These represent the minimum and maximum values ​​in the data, respectively.

[0036] It should be noted that the bottom hole annular pressure is composed of hydrostatic column pressure and annular pressure loss. Since annular pressure loss is closely related to flow rate, the annular pressure will exhibit abnormal characteristics after leakage occurs.

[0037] Assuming the drilling fluid is in a laminar flow state, the formula for the bottom hole annulus pressure during drilling can be simplified as follows: ; After leakage occurs, the flow rates above and below the leakage layer change, causing changes in annular pressure loss. The annular pressure loss above the leakage layer is: ; The annular pressure loss below the leakage layer is: ; in, The density of the annular drilling fluid. It is the acceleration due to gravity. For the depth of the well, This represents the flow rate of the drilling fluid during normal circulation. L represents the leakage amount, and L represents the leakage location. The diameter of the wellbore. For fluid viscosity, This is the outer diameter of the drill string.

[0038] The downhole annular pressure at drill depth H is primarily determined by the hydrostatic column pressure and the cumulative annular pressure loss from the wellhead to that depth. Under normal circulation conditions, the annular flow rate is approximately... Downhole annular pressure increases with well depth. After leakage occurs, the annular flow rate in the well section above the leakage point increases from... Reduced to As a result, the annular pressure loss in that well section decreases. Therefore, after leakage occurs, the annular pressure at the drill bit tends to decrease.

[0039] To effectively extract the correlation features between different stages of the annular pressure sequence during drilling, the first-stage well leakage detection network in this embodiment adopts a bidirectional gated cyclic unit network, namely a Bi-GRU network, to extract features from the annular pressure sequence and determine whether leakage has occurred based on the extracted pressure drop features.

[0040] like Figure 3 As shown, the Gated Recurrent Unit (GRU) is an improved structure of the Recurrent Neural Network (RNN). By introducing a gating mechanism, it alleviates the gradient vanishing problem that occurs in RNNs when modeling long sequences. Bi-GRU can utilize information from both historical and future states simultaneously, enhancing the model's ability to learn temporal context information.

[0041] exist Figure 3 In the middle, x t Enter information for the current moment; h t-1 This represents the hidden state from the previous time step. The hidden state acts as the neural network's memory, containing information about the data seen by previous nodes; h t This indicates the hidden state that will be passed on to the next time step; For the candidate hidden state, r t To reset the door, z t For updating the gate; σ is the sigmoid function, which can transform data into values ​​in the range of 0-1; tanh is the tanh function, which can transform data into values ​​in the range of [-1,1]. To update the weight matrix corresponding to the gate; This represents the weight matrix corresponding to the reset gate; The weight matrix represents the candidate hidden state; its calculation formula is: ; ; ; ; like Figure 4 As shown, the Bi-GRU in this embodiment consists of two independent GRU sub-networks: one processes the input sequence in the forward direction of time (forward GRU) to learn information from the past to the present time; the other processes the input sequence in the reverse direction of time (backward GRU) to learn information from the future to the present time. This bidirectional structure can simultaneously capture the temporal dependencies between historical and future states, improving the ability to capture dynamic features of time series.

[0042] exist Figure 4 middle, and Let represent the hidden states from left to right and from right to left, respectively. GRU represents a GRU unit, and x represents an element in the input sequence. The hidden states in the two directions are concatenated to obtain the final hidden state. ,Right now .

[0043] like Figure 5 As shown, the first-level well leakage detection network is used for real-time leakage detection during drilling. To characterize the decreasing trend and local abrupt changes in annular pressure when leakage occurs, this embodiment further introduces annular pressure gradient features based on the original annular pressure sequence; the input size is... ,in This represents the length of the time series.

[0044] The first-level well leakage detection network has a length of The annular pressure time window is used as input. After relevant features are extracted by Bi-GRU, a fully connected layer is used to output the normal and leakage classification scores corresponding to the time window. The scores are then converted into two probabilities by the softmax function. When the leakage category probability is higher, it is determined that well leakage has occurred in the current time window.

[0045] When the leakage probability continuously exceeds a preset threshold, and the measured annular pressure decreases gradually or the pressure gradient enters a relatively stable state, the difference between the reference annular pressure under normal operating conditions and the measured annular pressure is calculated based on the drill bit depth corresponding to the detection window, and this difference is defined as the annular pressure decrease caused by leakage; that is... ;in, The reference annular pressure under normal operating conditions at well depth H is obtained by correcting the mechanism model and pre-leakage measured data. This represents the bottom hole annulus pressure measured after leakage occurs. This pressure drop serves as the global hydraulic response characteristic output in the first-level identification stage, and is used for subsequent second-level leakage layer location constraints.

[0046] After leakage occurs, the flow state in the annulus above and below the leakage layer changes. The velocity decreases most drastically at the leakage layer, with the decrease being greater closer to the leakage point. Conversely, the fluid velocity tends to stabilize further away from the leakage layer. The drilling fluid velocity above the leakage layer is... Below the leakage layer, the flow rate remains essentially constant, therefore its velocity is... ;in, This represents the normal circulating flow rate. Leakage amount Let be the cross-sectional area of ​​the annular space.

[0047] After a loss of circulation occurs, the flow rate in the annulus above the lost circulation zone decreases, while the flow rate in the annulus below the lost circulation zone remains approximately normal, creating a velocity difference between the areas above and below the lost circulation zone. This velocity change alters the axial convection heat-carrying capacity of the drilling fluid, and the drilling fluid flowing into the formation at the lost circulation location causes heat loss. Therefore, compared to normal circulation, the annular drilling fluid temperature distribution changes under lost circulation conditions, manifesting as a change in the slope of the temperature profile or a localized anomaly in the temperature gradient near the lost circulation zone. Based on this, this embodiment comprehensively utilizes velocity, velocity gradient, temperature, and temperature gradient characteristics to locate the lost circulation zone.

[0048] Once the primary well leakage detection network identifies a leakage, it initiates the drill string lifting measurement process to collect drilling fluid velocity and temperature data along the well depth. The secondary leakage layer location network is used for leakage layer location. Its input is a multi-channel depth window collected during the lifting process. Input features include the original drilling fluid velocity value, velocity gradient, original temperature value, and temperature gradient sequence.

[0049] Suppose that a total of M well depth sampling points are obtained during the lifting measurement process, and the well depth sequence is represented as follows: ; the i-th well depth point The corresponding multi-channel feature vector is represented as ;in, Indicates well depth Drilling fluid flow rate at the location, This represents the velocity gradient at that point. This indicates the drilling fluid temperature at that point. This represents the temperature gradient corresponding to that point.

[0050] Let the length of the sliding depth window be N, then the sampling points are based on the well depth. Construct the i-th sliding depth window, i.e. ;in, For the target well depth The corresponding multi-channel feature vector.

[0051] The secondary lost-loop location network uses this sliding window as input, extracts the flow velocity and temperature variation characteristics within the window using Bi-GRU, and outputs the target well depth within the window. Probability of belonging to the leakage layer As the sliding window continues to slide, the leakage probability corresponding to each well depth can be obtained sequentially, forming a leakage probability sequence distributed along the well depth. By comparing it with a pre-set threshold σ, the leakage well section is output.

[0052] To evaluate the model's performance in locating missed intervals, this embodiment compares the predicted missed intervals with the actual missed intervals; the model's performance is comprehensively evaluated by introducing the Intersection over Union (IoU) accuracy. ;in, To improve positioning accuracy, To determine the overlap length between the predicted interval and the actual interval, Let be the length of the union of the two.

[0053] The secondary leakage layer location network structure in this embodiment is as follows: Figure 6 As shown.

[0054] After leakage occurs, the flow distribution in the wellbore annulus changes. Assume the drill bit depth is... The depth of the lost layer is ,and The annular flow rate below the leakage layer is approximately maintained at the normal circulation flow rate. Above the lost circulation zone, due to some drilling fluid entering the formation, the annular return flow rate is reduced from... Reduced to This flow rate change causes a difference in drilling fluid velocity between the well sections above and below the lost circulation zone, and at the same time reduces the annular pressure loss in the well section above the lost circulation zone, resulting in a decrease in the bottom hole annular pressure.

[0055] Therefore, the drop in bottomhole annular pressure recorded by the first-stage well leakage detection network can be considered as the overall pressure response caused by the reduction in annular flow and pressure loss in the well section above the leakage zone after leakage occurs. If the second-stage leakage zone location network identifies a velocity anomaly in a certain well depth range, it can estimate the leakage based on the velocity difference between the upstream and downstream of that range, and calculate the theoretical annular pressure drop it would cause in conjunction with the well depth of that range. If the calculated theoretical pressure drop is close to the annular pressure drop recorded by the first-stage well leakage detection network, it indicates that the velocity anomaly in that well depth range corresponds to the pressure drop during drilling, and that range is more likely to be a true leakage zone. If the difference is large, it indicates that although there is a velocity or temperature anomaly in that range, it is unlikely to cause the pressure drop observed during drilling, and it is more likely to be a false anomaly caused by local disturbance or measurement error.

[0056] Based on this, this embodiment constructs annular pressure response consistency constraints, so that the output results of the secondary lost layer location network not only conform to the abnormal characteristics of flow velocity and temperature, but also satisfy the detection of annular pressure drop response by the primary well lost layer identification network, thereby improving the accuracy and reliability of lost layer location.

[0057] For the i-th sliding depth window, the velocity sequence within the window is divided as follows: Figure 7 The average flow velocity of the two sub-intervals shown is calculated separately, and the equivalent flow rate corresponding to the window is estimated based on the difference in average flow velocity between the two sub-intervals; that is... ; ;in, For the cross-sectional area of ​​the annulus, , These represent the average flow velocity in the two sub-intervals before and after the current sliding window, respectively.

[0058] Since the decrease in annular pressure at the drill bit after leakage occurs mainly stems from the reduction in annular flow rate in the well section above the leakage zone, resulting in decreased annular frictional pressure loss, the candidate depth is... Substituting the equivalent flow rate into the annular pressure model, we calculate the theoretical annular pressure drop that the candidate point could cause, and compare it with the annular pressure drop obtained in the first-stage identification phase. Construct the annular pressure response consistency loss function; that is... ; ;in, This represents the predicted annular pressure drop at the current drill bit position. This represents the drop in bottom hole annular pressure detected by the primary well leakage detection network. It is a very small positive number. To avoid the influence of the physical constraint loss function on the learning of normal points, we introduce... As a gating mechanism, the physical constraint loss term is only enabled when the probability of leakage at the current point is relatively high; otherwise, it is not enabled.

[0059] The localization loss of the second-level leaky layer localization network can be achieved using binary classification cross-entropy loss, i.e. ;in, Let represent the true label of the i-th well depth point. If this point belongs to the true leakage interval, then... ,otherwise .

[0060] Therefore, the total loss function of the secondary leaky layer localization network in this embodiment is: ;in, The physical consistency loss weight is used to adjust the influence of annular pressure response constraints on network training. Through this loss function, the second-level leakage layer localization network, while learning local anomaly features of flow velocity and temperature, is also constrained by the actual annular pressure drop identified by the first-level well leakage detection network, ensuring that the predicted leakage interval can simultaneously meet the requirements of local parameter anomalies and overall annular pressure response consistency.

[0061] The leakage identification results of the first-level well leakage identification network not only serve as the starting condition for the second-level leakage layer location network, but the pressure drop output by the network also serves as physical constraint information in the training of the second-level location network. When the theoretical pressure drop at a certain well depth differs significantly from the annular pressure drop detected by the first-level well leakage identification network, the corresponding physical consistency loss increases, the trend of high leakage probability output by the network is suppressed, the misjudgment caused by false anomalies in flow velocity and temperature data is reduced, and the reliability and accuracy of leakage layer location are improved.

[0062] Case Analysis To verify the application effect of the method proposed in this embodiment, this embodiment constructs a mechanism simulation dataset of annular pressure, drilling fluid velocity, and wellbore temperature based on the actual wellbore structure, drilling fluid physical parameters, circulation displacement, and set leakage conditions of a certain well. Among them, the annular pressure data is calculated based on the hydrostatic column pressure and annular pressure loss model, the drilling fluid velocity data is calculated based on the annular flow conservation relationship before and after leakage and the annular cross-sectional area, and the temperature data is calculated based on the wellbore-formation transient heat transfer model. The above simulation data is used to simulate the multi-parameter response obtained by the downhole measurement unit during drilling and lifting measurement to verify the effectiveness of the method in this embodiment for identifying leakage occurrence and locating leakage layers.

[0063] Temperature data were generated based on a wellbore-formation transient heat transfer model. The model simplifies the radial structure of the wellbore into equivalent media including drilling fluid inside the drill string, drill string wall, annular drilling fluid, solid media near the well wall, near-formation, and far-formation media, and discretizes them along the well depth. The drilling fluid region mainly considers axial convective heat transfer and radial heat transfer with adjacent solid media, while the solid media region mainly considers conductive heat transfer.

[0064] Based on the law of energy conservation, the unified temperature control equation is: ; in, Let i be the temperature of the i-th computational region or equivalent medium element. For the density of the medium, For specific heat capacity, Thermal conductivity, The axial flow velocity is... Radial coordinates, The coordinates are in the direction of well depth. For heat source items, This is the axial thermal correction factor. For annular drilling fluids, the axial convection term is mainly considered; for drill string wall, wellbore wall, and formation solid media, the radial and axial thermal terms are mainly considered.

[0065] By using the finite volume method to discretely solve the above control equations, the annular drilling fluid temperature distribution under normal circulation and leakage conditions can be calculated, and the temperature gradient characteristics along the well depth direction can be obtained as the input of the secondary leakage layer location network.

[0066] During drilling, the downhole measurement unit continuously collects annular pressure data at the drill bit. For example... Figure 8 As shown, when drilling reached approximately 3690m, the annular pressure at the drill bit changed from a slowly increasing trend with well depth to a decreasing trend, indicating an abnormal change in the annular flow state. This pressure anomaly mainly stemmed from the reduced annular return flow rate above the leakage point after the leakage occurred, leading to a decrease in annular friction and pressure loss in that section, which in turn caused a drop in annular pressure at the drill bit. The model thus identified this as well leakage. After determining that well leakage had occurred, the annular pressure drop under the same well depth conditions was obtained based on the reference pressure curve under normal operating conditions, serving as the pressure response consistency constraint information for subsequent leakage zone location.

[0067] After the leakage is detected by the primary well leakage detection network, the drill string lifting measurement process is initiated to collect drilling fluid flow rate and temperature data along the well depth direction. For example... Figure 9As shown, around 2890m, the drilling fluid velocity transitions from a lower level above the lost circulation point to a higher level below it, exhibiting a significant abrupt velocity change. Simultaneously, the temperature gradient shows a clear inflection point in this range. Based on multi-channel characteristics such as velocity, velocity gradient, temperature, and temperature gradient within this well section, the secondary lost circulation zone location network outputs approximately 2880–2910m as a suspected lost circulation zone. Comparison shows that the actual lost circulation location, set at approximately 2890m, falls within the suspected lost circulation zone identified by the network, indicating that this method can effectively capture multi-parameter anomalies near the lost circulation zone.

[0068] Downhole flow velocity and temperature data may be affected by local fluid disturbances, sensor noise, and changes in wellbore heat transfer conditions, leading to weak abnormal fluctuations in some non-leakage sections. Relying solely on local flow velocity or temperature changes may result in misjudgments.

[0069] In this embodiment, the velocity difference within the window is converted into an equivalent flow rate change and substituted into the annular pressure loss model. The theoretical annular pressure drop at the drill bit that this well section could theoretically cause is calculated. When the theoretical pressure drop of a certain well section is inconsistent with the actual pressure drop recorded in the first-level identification stage, it indicates that although there are local parameter fluctuations in this well section, they are difficult to explain the global annular pressure response observed during drilling. Therefore, the confidence level of this well section is suppressed, spurious anomalies caused by local noise are suppressed, and the reliability and accuracy of the lost circulation zone location results are improved.

[0070] Example 2 Embodiment 2 of the present invention introduces a downhole multi-parameter coordinated oil and gas drilling lost zone location system.

[0071] like Figure 10 The illustrated downhole multi-parameter coordinated oil and gas drilling lost circulation zone location system includes: The first acquisition module is configured to acquire bottom hole annular pressure data during the drilling process; The first identification module is configured to identify well leakage and obtain the amount of well leakage drop based on the acquired bottom hole annular pressure data and the first-level well leakage identification network. The second acquisition module is configured to, when well leakage is detected, lift the drill string to acquire the drilling fluid flow rate and temperature along the well depth; The second identification module is configured to obtain the leakage probability distribution corresponding to each well depth point along the well depth direction based on the obtained drilling fluid flow rate, temperature and secondary lost circulation layer location network. The location module is configured to determine the location of the lost layer based on the obtained loss probability distribution, thereby achieving multi-parameter coordinated location of lost layers in oil and gas drilling.

[0072] The detailed steps are the same as those of the downhole multi-parameter coordinated oil and gas drilling lost zone location method provided in Example 1, and will not be repeated here.

[0073] Example 3 Embodiment 3 of the present invention provides a computer-readable storage medium.

[0074] A computer-readable storage medium having a program stored thereon, which, when executed by a processor, implements the steps in the downhole multi-parameter coordinated oil and gas drilling lost zone location method as described in Embodiment 1 of the present invention.

[0075] The detailed steps are the same as those of the downhole multi-parameter coordinated oil and gas drilling lost zone location method provided in Example 1, and will not be repeated here.

[0076] Example 4 Embodiment 4 of the present invention provides an electronic device.

[0077] An electronic device includes a memory, a processor, and a program stored in the memory and running on the processor. When the processor executes the program, it implements the steps in the downhole multi-parameter coordinated oil and gas drilling lost zone location method as described in Embodiment 1 of the present invention.

[0078] The detailed steps are the same as those of the downhole multi-parameter coordinated oil and gas drilling lost zone location method provided in Example 1, and will not be repeated here.

[0079] Example 5 Embodiment 5 of the present invention provides a computer program product.

[0080] A computer program product includes software code, wherein the program in the software code performs the steps of the downhole multi-parameter coordinated oil and gas drilling lost zone location method as described in Embodiment 1 of the present invention.

[0081] The detailed steps are the same as those of the downhole multi-parameter coordinated oil and gas drilling lost zone location method provided in Example 1, and will not be repeated here.

[0082] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code. The solutions in the embodiments of the present invention can be implemented using various computer languages, such as the object-oriented programming language Java and the interpreted scripting language JavaScript.

[0083] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0084] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0085] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0086] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention.

[0087] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

[0088] The above description is merely a preferred embodiment of this practice and is not intended to limit the scope of this practice. Various modifications and variations can be made to this practice by those skilled in the art. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of this practice should be included within the protection scope of this practice.

Claims

1. A method for locating lost circulation zones in oil and gas drilling using a multi-parameter coordinated approach in downhole operations, characterized in that: include: Acquire bottom hole annular pressure data during the drilling process; Based on the acquired bottom hole annular pressure data and the first-level well leakage identification network, well leakage is identified, and the amount of bottom hole annular pressure drop is obtained. When well leakage is detected, the drill string is pulled up to obtain the drilling fluid flow rate and temperature along the well depth; Based on the obtained drilling fluid flow rate, temperature, and secondary lost circulation zone location network, the loss probability distribution corresponding to each well depth point along the well depth direction is obtained. The location of the lost circulation layer is determined based on the obtained loss probability distribution, thereby achieving multi-parameter coordinated location of lost circulation layers in oil and gas drilling. The secondary lost layer location network is trained using a total loss function that includes a location loss function and an annular pressure response consistency loss function. The location loss function is used to constrain the consistency between the lost layer probability distribution and the actual lost layer label. The annular pressure response consistency loss function is used to constrain the consistency between the theoretical annular pressure drop and the actual annular pressure drop corresponding to the lost layer, so that the lost layer simultaneously meets the requirements of local flow velocity, temperature anomaly characteristics and overall annular pressure response consistency, thereby obtaining the lost layer location and realizing the downhole multi-parameter coordinated oil and gas drilling lost layer location. The first-stage well leakage detection network uses a bidirectional gated cyclic unit network to extract features from the annular pressure sequence and determine whether leakage has occurred based on the extracted pressure drop features. The secondary lost layer location network takes the sliding depth window as input, and the bidirectional gated loop unit network extracts the change characteristics of flow velocity and temperature within the window, outputs the probability that the target well depth point in the window belongs to the lost layer. As the sliding window slides continuously, the lost probability corresponding to each well depth point is obtained in sequence, forming a lost probability sequence distributed along the well depth, and outputting the lost well section. The total loss function of the second-level lost layer localization network is: ;in, The physical consistency loss weight is used to adjust the degree of influence of the annular pressure response constraint on network training; The localization loss of the second-level leaky layer localization network is calculated using binary classification cross-entropy loss, i.e. N is the length of the sliding depth window. The probability of the leakage layer. Let represent the true label of the i-th well depth point. If this point belongs to the true leakage interval, then... ,otherwise ; The annular pressure response consistency loss function is, i.e. , The predicted annular pressure drop at the current drill bit position is, i.e. ; This represents the drop in bottom hole annular pressure detected by the primary well leakage detection network. It is a very small positive number; For gating mechanisms; and These represent the equivalent flow rates corresponding to the two sub-intervals before and after the current sliding window, respectively. ; ;in, For the cross-sectional area of ​​the annulus, , These represent the average flow velocity in the two sub-intervals before and after the current sliding window, respectively; The diameter of the wellbore. For fluid viscosity, This is the outer diameter of the drill string.

2. The downhole multi-parameter coordinated method for locating lost circulation zones in oil and gas drilling as described in claim 1, characterized in that, In the process of obtaining drilling fluid flow rate and temperature along the well depth, the drill string is raised to obtain drilling fluid temperature and flow rate along the well depth. During the process of raising the drill string, a fixed depth stop measurement method is adopted, that is, the drill string is raised a certain distance and then stopped for a period of time. After the flow rate and drilling fluid temperature data are collected at the current measurement point, the drill string is raised to the next measurement point.

3. The downhole multi-parameter coordinated method for locating lost circulation zones in oil and gas drilling as described in claim 1, characterized in that, In the process of obtaining drilling fluid flow, based on the Doppler ultrasonic velocimeter and the Doppler effect, the flow velocity of solid particles in the drilling fluid is calculated by measuring the echo frequency shift caused by solid particles in the drilling fluid in the wellbore, so as to determine the drilling fluid flow velocity.

4. A downhole multi-parameter coordinated oil and gas drilling lost circulation zone location system, employing the downhole multi-parameter coordinated oil and gas drilling lost circulation zone location method as described in any one of claims 1-3, characterized in that, include: The first acquisition module is configured to acquire bottom hole annular pressure data during the drilling process; The first identification module is configured to identify well leakage and obtain the amount of well leakage drop based on the acquired bottom hole annular pressure data and the first-level well leakage identification network. The second acquisition module is configured to, when well leakage is detected, lift the drill string to acquire the drilling fluid flow rate and temperature along the well depth; The second identification module is configured to obtain the leakage probability distribution corresponding to each well depth point along the well depth direction based on the obtained drilling fluid flow rate, temperature and secondary lost circulation layer location network. The location module is configured to determine the location of the lost layer based on the obtained loss probability distribution, thereby achieving multi-parameter coordinated location of lost layers in oil and gas drilling.

5. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the steps of the downhole multi-parameter coordinated oil and gas drilling lost zone location method as described in any one of claims 1-3.

6. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the program, it implements the steps of the downhole multi-parameter coordinated oil and gas drilling lost zone location method as described in any one of claims 1-3.

7. A computer program product, comprising software code, characterized in that, The program in the software code executes the steps of the downhole multi-parameter coordinated oil and gas drilling lost zone location method as described in any one of claims 1-3.

Images