Instrument diagnosis by combining multi-variable validation and mix of measured and predicted data

By using multi-variable validation with predicted and measured data, the technology addresses inaccuracies in downhole operations, enhancing event detection and operational reliability through accurate anomaly detection.

US20260357694A1Pending Publication Date: 2026-11-12HALLIBURTON ENERGY SERVICES INC
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
US19/202198
Authority / Receiving Office
US · United States
Patent Type
Applications(United States)
Current Assignee / Owner
Filing Date
2025-05-08
Publication Date
2026-11-12

AI Technical Summary

Technical Problem

In downhole environments, the complexity of operations and numerous sensors lead to inaccurate or erroneous data, making it difficult for operators to detect events such as sensor failures or recalibration issues, which can affect ongoing operations and monitoring.

Method used

The technology employs multi-variable validation using both measured and predicted data through model instances to detect deviation events by comparing predicted data with actual measurements, allowing for more accurate detection of operational anomalies.

Benefits of technology

This approach enhances the accuracy of event detection in downhole operations, enabling improved decision-making and timely remedial actions, such as sensor recalibration or system adjustments, thereby improving operational reliability.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure US20260357694A1-D00000_ABST
    Figure US20260357694A1-D00000_ABST
Patent Text Reader

Abstract

Aspects of the subject technology relate to systems, methods, and computer-readable media for detecting events associated with operations in a downhole environment. A first model instance can be applied to measured data of a first variable to predict first data a second variable. A second model instance can be applied to predicted data of the first variable to predict second data for the second variable. As follows, the output of the model instances can be compared to the measured data of the second variable to determine whether a deviation event has occurred.
Need to check novelty before this filing date? Find Prior Art

Description

BACKGROUNDTechnical Field

[0001] The present disclosure generally relates to system monitoring and diagnosis for downhole environments through multi-variable validation, and more specifically to system monitoring and diagnosis through multi-variable validation using both measured and predicted data.Introduction

[0002] During operations in downhole environments, variables can be measured and monitored to characterize the operations in the downhole environment. Operations in downhole environments can be complex and involve numerous sensors and systems. As follows, several different variables in association with the downhole environment can be measured and monitored to characterize the operations in the downhole environment.BRIEF DESCRIPTION OF THE DRAWINGS

[0003] The various advantages and features of the present technology will become apparent by reference to specific implementations illustrated in the appended drawings. A person of ordinary skill in the art will understand that these drawings only show some examples of the present technology and would not limit the scope of the present technology to these examples. Furthermore, the skilled artisan will appreciate the principles of the present technology as described and explained with additional specificity and detail through the use of the accompanying drawings in which:

[0004] FIG. 1A illustrates a schematic diagram of an example wellbore operating environment, according to some examples of the present disclosure;

[0005] FIG. 1B illustrates a schematic representation of a well environment in a production phase, according to some examples of the present disclosure;

[0006] FIGS. 2A and 2B illustrate a flowchart of an example method of identifying events occurring in association with downhole operation through multi-variable validation with predicted and measured data, according to some examples of the present disclosure;

[0007] FIG. 3A illustrates a plot of predicted downhole sensor data that is generated through application of a first model instance in comparison to measured downhole sensor data, according to some examples of the present disclosure;

[0008] FIG. 3B illustrates a plot of predicted downhole sensor data that is generated through application of a second model instance in comparison to the measured downhole sensor data, according to some examples of the present disclosure;

[0009] FIG. 4A and 4B illustrate a flowchart of another example method of identifying events occurring in association with downhole operation through multi-variable validation with predicted and measured data, according to some examples of the present disclosure;

[0010] FIG. 5A illustrates a plot of predicted downhole sensor data that is generated through application of a first model instance in comparison to measured downhole sensor data, according to some examples of the present disclosure;

[0011] FIG. 5B illustrates a plot of predicted downhole sensor data that is generated through application of a second model instance in comparison to the measured downhole sensor data, according to some examples of the present disclosure;

[0012] FIG. 5C illustrates a plot of predicted downhole sensor data that is generated through application of a third model instance in comparison to measured downhole sensor data, according to some examples of the present disclosure;

[0013] FIG. 5D illustrates a plot of predicted downhole sensor data that is generated through application of a fourth model instance in comparison to the measured downhole sensor data, according to some examples of the present disclosure;

[0014] FIG. 6 is an example of a deep learning neural network that can be used to implement all or a portion of the systems and techniques described herein, according to some examples of the present disclosure;

[0015] FIG. 7 illustrates an example processor-based system with which some aspects of the subject technology can be implemented, according to some examples of the present disclosure.DETAILED DESCRIPTION

[0016] The detailed description set forth below is intended as a description of various configurations of the subject technology and is not intended to represent the only configurations in which the subject technology can be practiced. The appended drawings are incorporated herein and constitute a part of the detailed description. The detailed description includes specific details for the purpose of providing a more thorough understanding of the subject technology. However, it will be clear and apparent that the subject technology is not limited to the specific details set forth herein and may be practiced without these details. In some instances, structures and components are shown in block diagram form to avoid obscuring the concepts of the subject technology.

[0017] As discussed previously, during operations in downhole environments, variables can be measured and monitored to characterize the operations in the downhole environment. Operations in downhole environments can be complex and involve numerous sensors and systems. As follows, several different variables in association with the downhole environment can be measured and monitored to characterize the operations in the downhole environment. Such downhole operations can include production, stimulation, and drilling activities.

[0018] Many different sensors and instruments, including downhole and surface sensors and instruments, can be used to measure variables for monitoring and characterizing the downhole operations. As follows, an operator can analyze graphical representations of measured variables to monitor and characterize the downhole operations. Due to the large number of sensors and systems and complexities associated with operating in downhole environments, measured data can be inaccurate or otherwise erroneous. Such inaccuracies correspond to events occurring during operation in the downhole environment. Examples of such events can include failed sensors and instruments, failed recalibration of sensors and instruments, down plugging events, and other applicable occurrences that affect or lead to the gathering of erroneous data.

[0019] In some scenarios, an operator can easily detect, e.g. visually, the occurrence of an event in the course of monitoring and characterizing downhole operations. However, in other scenarios it can be difficult for an operator to detect, e.g. visually, such events. Failure to detect an event can lead to operators having a false assumption that measured data is accurate. Further, failure to detect an event can result in small changes that are difficult for an operator to detect, e.g. visually.

[0020] Failure to recognize such events can affect current downhole operations and the monitoring and characterization of such operations. Specifically, operations can be performed downhole that are initiated based on erroneous data. Additionally, past missed event detections can be overlooked, which can ultimately affect current downhole operations and the monitoring and characterization of such operations. Specifically, in relying on erroneous measurements, operators can update their models with this inaccurate data and continue to run cycles with the incorrectly updated models. Further, in failing to identify the inaccuracy in the measured data, operators can miss opportunities to recalibrate systems and sensors.

[0021] The disclosed technology addresses the foregoing by accessing a set of measured data of different variables associated with downhole operations. In turn, model instances can be applied based on one of the variables to predict data of a second variable. In turn, the model instances can be validated based on the predicted data of the second variable to detect the occurrence of events associated with the downhole operations. In using multiple variables to validate model instances and detect an occurrence of an event, the technology abstains from assuming that either or both measured data is accurate and predicted data has deviated. In turn, this can lead to more accurate prediction of events from both past and current measured data, in particular, when the events are difficult for an operator to detect. As follows, this can lead to improved decision making in monitoring and controlling downhole operations.

[0022] FIG. 1A illustrates a schematic diagram of an example wellbore operating environment 100. As depicted in FIG. 1A, the operating environment 100 includes a wellbore 114 that penetrates a formation 102 for the purpose of recovering hydrocarbons, storing hydrocarbons, injecting water or carbon dioxide, or the like in formation 104. In certain embodiments, the purpose of the operating environment 100 is carbon capture & storage (CCS) and includes equipment associated with that purpose (not shown in FIG. 1A). In certain embodiments, the purpose of the operating environment 100 is geothermal energy capture and includes equipment associated with that purpose (not shown in FIG. 1A).

[0023] As depicted in FIG. 1A, formations 102 and 104 are subterranean formations, although it is noted that formations 102 and 104 may be subsea formations. In certain locations, there are a plurality of underground formations 102, 104. The wellbore 114 may extend substantially vertically away from the Earth’s surface 106 over a vertical wellbore portion, or may deviate at any angle from the Earth’s surface 106 over a deviated or horizontal wellbore portion 116. In alternative operating environments, portions or substantially all of the wellbore 114 may be vertical, deviated, horizontal, and / or curved. The wellbore 114 may be drilled into the formations 102, 104 using any suitable drilling technique. As shown, a drilling or servicing rig 110 disposed at the surface 106 (which may be the surface of the Earth, a seafloor surface, or a sea surface) comprises a derrick 112 from which a tubular string 120 (e.g., a drill string, a tool string, a segmented tubing string, a jointed tubing string, or any other suitable conveyance, or combinations thereof) is positioned within or partially within the wellbore 114. The tubular string 120 may include two or more concentrically positioned strings of pipe or tubing (e.g., a first work string may be positioned within a second work string). The drilling or servicing rig 110 may include a motor driven winch and other associated equipment for lowering the tubular string into the wellbore 114. Alternatively, a mobile workover rig, a wellbore servicing unit (e.g., coiled tubing units), or the like may be used to lower the work string into the wellbore 114. In such an environment, the tubular string 120 may be utilized in drilling, stimulating, completing, or otherwise servicing the wellbore, or combinations thereof. A drilling or servicing rig 110 may also comprise other equipment. In certain types of operations, a fluid 122 is forced down the tubular string 120 and out through perforations 124 to fracture the formations 104 that surround the perforations 124.

[0024] While FIG. 1A depicts a stationary drilling rig 110, one of ordinary skill in the art will readily appreciate that mobile workover rigs, stimulation units (such as fracturing pumps), wellbore servicing units (such as coiled tubing units), and the like may be employed. It is noted that while the figures or portions thereof may exemplify horizontal or vertical wellbores, the principles of the presently disclosed apparatuses, methods, and systems, may be similarly applicable to horizontal wellbore configurations, conventional vertical wellbore configurations, deviated wellbore configurations, and any combinations thereof. The horizontal, deviated, or vertical nature of any figure is not to be construed as limiting the wellbore to any particular configuration or formation.

[0025] The disclosure now turns to a description of FIG. 1B, which illustrates a schematic representation of a well environment 130 in a production phase. The well environment 130 can represent an applicable environment in which a substance is pumped through the wellbore 132 toward the surface. For example, the well environment 130 can represent a hydrocarbon production environment in which hydrocarbons are pumped through the wellbore 132 toward the surface. In another example, the well environment 130 can represent a geothermal environment in which hot water is pumped through the wellbore 132 toward the surface.

[0026] The well environment 130 includes a production system 134 disposed in relation to the wellbore 132. The production system 134 includes a surface control system 136. The production system 134 also includes components disposed downhole in the wellbore 132. Specifically, the production system 134 includes a motor 150, a seal section 152, a gas separator 154, a pump 156, a gauge 158, and a power cable 160. The components of the production system 134, in combination, function to perform various tasks related to pumping a substance through the wellbore 132 toward the surface. In particular, the surface control system 136 functions to control and interact with the various downhole components for performing various tasks related to pumping a substance through the wellbore 132 towards the surface.

[0027] The gauge 158 functions to generate downhole data of one or more monitored parameters. Specifically, the downhole data can include applicable data that is capable of being measured downhole. When a first component or first point is described as being before a second component or second point, the first component or point can be positioned further in a wellbore than a second component or point. For example, the gauge 158 can include a pressure gauge that is configured to identify a wellbore pressure before the pump 156, e.g. before a pump intake or gas separator 154. Further the gauge 158 can function to measure parameters for preventing or reducing formation damage cause by over-production through the wellbore 132. The gauge 158 can communicate with the surface control system 136 in generating downhole data. Specifically, the gauge 158 can provide the downhole data as telemetry data to the surface control system 136, where the downhole data can be used in controlling production operation of the production system 134.

[0028] An applicable telemetry system can be used in transmitting data in accordance with the technology described herein. Specifically, an applicable telemetry system can be used in transmitting measured variable data, e.g. to the surface. Further, an applicable telemetry system can be used in transmitting control signals downhole for performing one or more wellbore operations downhole. For example, one or more of a distributed acoustic telemetry system, a wireline system, a mud pulse telemetry system, and a fiber optic system can be used in transmitting measured variable data and control signals.

[0029] Further, a telemetry system can be implemented through downhole memory tools that store captured downhole data. Such memory tools can be implemented as part of sensors that gather / generate the downhole data. The downhole memory tools can then be retrieved from the downhole environment to bring the captured data to the surface through an applicable mechanism, e.g. slickline.

[0030] The motor 150 functions to drive the pump 156. Specifically, the motor 150 can receive power from the surface through the power cable 160 to drive the pump 156 in lifting production substance towards the surface. The motor 150 can be an applicable motor that is capable of driving the pump 156, such as an electrical submersible pump (“ESP). Correspondingly, the pump 156 can be an applicable pump that is capable of pumping production substances toward the surface of the wellbore 132, such as an ESP. The seal section 152 is disposed between the motor 150 and the intake of the pump 156. The seal section 152 functions to isolate the motor 150 from downhole fluids. The seal section 152 also can function to equalize pressure in the wellbore 132 with pressure in the motor 150.

[0031] The gas separator 154 is positioned between the pump 156 and the sealing section 152 and motor 150 combination. The gas separator 154 can serve, at least in part, as an intake for the pump 156. In particular, the gas separator 154 can function to separate gas from fluid in the wellbore and allow for the entry of the separated fluid into the pump 156. In turn, the pump 156 can pump the separated fluid towards the surface as part of a production substance. The separated fluid that is fed to the pump 156 can include portions of the separated gas that are dissolved into the fluid to form a more homogenized solution.

[0032] The technology described herein can be implemented in the wellbore environments shown in FIGS. 1A and 1B. Specifically, the technology described herein can be used to identify events in a drilling wellbore environment or a producing wellbore environment. The disclosure now continues with a discussion of multi-variable validation with predicted and measured data to identify events occurring in association with downhole operations.

[0033] FIGS. 2A and 2B illustrate a flowchart 200 of an example method of identifying events occurring in association with downhole operation through multi-variable validation with predicted and measured data, according to some examples of the present disclosure. The method shown in FIGS. 2A and 2B is provided by way of example, as there are a variety of ways to carry out the method. Additionally, while the example method is illustrated with a particular order of steps, those of ordinary skill in the art will appreciate that FIGS. 2A and 2B and the modules shown therein can be executed in any order and can include fewer or more modules than illustrated. Each module shown in FIGS. 2A and 2B represents one or more steps, processes, methods or routines in the method. The modules will be discussed with respect to the example environments described herein.

[0034] At module 202, a set of measured data of a first variable associated with operation in a downhole environment is accessed. The first variable can include an applicable variable associated with operation in a downhole environment. Specifically, the first variable can include data that is capable of being gathered by a surface sensor to the downhole environment. Alternatively, the first variable can include data that is capable of being gathered by a subsurface sensor in the downhole environment. For example, measurements of a surface pressure can be accessed. In another example, measurements of an intake pressure at a downhole pump can be accessed.

[0035] At module 204, predicted data of the first variable are generated. The predicted data of the first variable can be generated through an applicable technique. Specifically, the predicted data of the first variable can be generated by a model that is either a physics-based model or a data-driven model. The data-driven model can be trained on applicable data, e.g. data from the current downhole environment or other downhole environments. For example, if a surface pressure (exemplary first variable) is measured by a surface instrument and accessed at module 202, then a surface pressure can be predicted at module 204. As another example, if an intake pressure (another exemplary first variable) is measured and accessed at module 202, then an intake pressure can be predicted at module 204.

[0036] Variables that can be measured and predicted in accordance with the technology described herein, include applicable variables that can be measured and predicted in association with wellbore operations performed in or near a downhole environment. For example, variables can comprise the production fluid constituents flow rates, liquid flow rate, a solid, sand, or gas migration rate, or other measurements describing materials conveyance, fluid composition, a frequency component of a control signal for controlling either or both the speed and the torque of a motor in the wellbore operation, an output current of a variable speed drive used in the wellbore operation, an output voltage of a variable speed drive used in the wellbore operation, a tubing pressure of tubing used in the downhole environment during the wellbore operation. a casing pressure between a wellbore and casing or between two casing strings in the downhole environment, an annulus pressure between production tubing and a casing string, and a flowline pressure within a flowline used during the wellbore operation.

[0037] Variables can be associated with wellbore operations during an applicable phase of a wellbore. Specifically, a wellbore operation can comprise an aspect of or an action performed during a completion phase, a production phase, or a workover phase of a wellbore. For example, a wellbore operation can comprise operation of an electrical submersible pump during a well production phase. In another example, a wellbore operation can comprise a cabling workover operation such as a fishing operation, a valve actuation (e.g. opening or closing) operation, a gauge cutting operation, a plug setting or removing operation, a valve and memory log device deployment or removal operation. In yet another example, a wellbore operation can comprise a logging operation, e.g. wireline or log-while-drilling operation during a production phase or during a completion phase.

[0038] Variables can be in association with functioning of an electrical submersible pump in a downhole environment. Specifically, a variable can comprise a pump intake pressure, a pump intake temperature, a motor temperature, a pump discharge pressure, rotational speed of a pump, vibration in a y-direction in association with operation of the electrical submersible pump, vibration in a x-direction in association with operation of the electrical submersible pump, vibration in a z-direction in association with operation of the electrical submersible pump, or a combination thereof. An x direction, a y direction, and a z direction can be defined with respect to a vertical wellbore. For example, x and y directions can define a horizontal plane while a z direction can extend perpendicular from a horizontal plane defined by the x and y direction in a vertical direction.

[0039] At module 206, a set of measured data of a second variable associated with operation in the downhole environment is accessed. The second variable can include an applicable variable associated with operation in a downhole environment. Specifically, the second variable can include data that is capable of being gathered by a surface sensor in association with one or more wellbore operations performed in the downhole environment or in relation to the downhole environment. For example, the second variable can be measurements made at surface production equipment during a production operation that is performed in the downhole environment during a production phase of the well. Alternatively, the second variable can include data that is capable of being gathered by a subsurface sensor in the downhole environment. For example, a second variable can comprise a downhole pressure.

[0040] The second variable and the first variable can be associated with opposite locations, e.g. distal and proximal locations relative to a surface, in the downhole environment. Specifically, the first variable can be a surface variable that is measured at the surface of the downhole environment. As follows, the second variable can be a subsurface variable that is measured in the downhole environment. Alternatively, the first variable can be a subsurface variable that is measured in the downhole environment. As follows, the second variable can be a surface variable that is measured at the surface of the downhole environment.

[0041] At module 208, a first model instance is applied to the measured data of the first variable to generate predicted data of the second variable from the first model instance. An applicable model can be used for the first model instance. Examples of models include but are not limited to, a physics-based model, a linear regression model, a time series model, e.g. an autoregressive integrated moving average (ARIMA) or seasonal ARIMA (SARIMA), a gradient boosting machine, a random forest model, and a support vector machines (SVM). FIG. 6 discusses an example of implementing a deep learning neural network to predict data based on input and can be applied as a model instance in the various embodiments discussed herein. While a deep learning neural network is shown as an example for implementing the model instance, in various embodiments a neural network or shallow neural network can also be applied as the model instance. Specifically, a neural network with less than 3 hidden layers can be applied, or in various embodiments, a neural network with more than 3 hidden layers can be applied.

[0042] The models that are applied throughout this disclosure can depend on various applicable factors, such as characteristics of the data that is applied to the model, characteristics of data that is predicted by the model, requirements for the task of predicting the data, and available computational resources. For example, linear regression models provide a simple prediction with a linear relationship between variables and are easy to implement and interpret. However, such models can be limited to linear relationships. Time series models make predictions based on historical time series data and are effective for capturing temporal dependencies. However, such models can function on stationary data and be complex to tune. Gradient boosting machines provide a general prediction for complex and non-linear relationships and are capable of handling different types of data. However, such models can be prone to overfitting and can require specific tuning. Random forest models can model many features and non-linear relationships and are capable of handling overfitting due to their robustness. Such models are more robust against overfitting than many other models due to their ensemble nature. However, such models can be slower to train and more difficult to interpret. Support vector machines can predict complex relationships and outliers in high-dimensional spaces. However, such models utilize large amounts of computational resources.

[0043] At module 210, the first model instance is validated by comparing the predicted data of the second variable from the first model instance to the measured data of the second variable. Specifically, predicted data of the second variable that is generated based on measured data of the first variable can be compared to the measured data of the second variable. More specifically, a downhole variable that is predicted based on measured data of a surface variable can be compared to measured data of the downhole variable. For example, downhole pressure that is predicted based on measured surface pressure can be compared to measured downhole pressure.

[0044] At module 212, a second model instance is applied to the predicted data of the first variable (from module 204) to generate predicted data of the second variable from the second model instance. This is in contrast to the first model instance, where measured data of the first variable (from module 202) is applied to generate predicted data of the second variable. While the accuracy of the measured data may affect the first model instance, since predicted data is applied in the second model instance, its precision can be unaffected by the accuracy of the measured data. In turn and as will be discussed in greater detail later, this discrepancy in model performance based on whether predicted or measured data is applied to the model can be used to detect events associated with operation in the downhole environment. An applicable model can be applied as the second model instance.

[0045] At module 214, the second model instance is validated by comparing the predicted data of the second variable from the second model instance to the measured data of the second variable. Specifically, predicted data of the second variable that is generated through application of the second model instance based on predicted data of the first variable can be compared to the measured data of the second variable. More specifically, a downhole variable that is predicted based on predicted data of a surface variable can be compared to measured data of the downhole variable in validating the second model instance.

[0046] Applicable techniques for comparing different sets of data can be used in comparing data according to the technology described herein. Specifically, when comparing measured and predicted data of a variable, an applicable technique can be applied to quantify the comparison. For example, a simple difference between the measured and predicted data at a given time can serve as a comparison of the measured and predicted data. In another example, the measured and predicted data can be compared relative to a threshold. As follows, values of the measured data relative to the threshold can be compared to values of the predicted data relative to the threshold value as part of comparing the measured and predicted data.

[0047] At module 216, it is determined whether a deviation event associated with gathering the measured data of the first variable has occurred by comparing the validation of the first model instance with the validation of the second model instance. A deviation event can include an applicable event that causes measured data to deviate from normal or otherwise expected measurements. For example, a deviation event can comprise a malfunctioning sensor is gathering measured data, a damaged sensor is gathering measured data, a down plugging event, or a recalibration event associated with a sensor. A deviation event can be indicative of other events that can occur during downhole operations. For example, a deviation event can be indicative of a downhole system failure. As follows, operations in the downhole environment can be controlled in response to detected deviation events.

[0048] An expected output of the models can comprise a measured output for a variable that is predicted by the models. For example, both the first model instance and the second model instance can output values for the predicted data of the second variable. As follows, the output of the predicted data of the second variable for each model can be compared to measured data of the second variable to determine an amount of divergence between the expected output and the actual output for the model instances.

[0049] Divergence between an expected output of the model instances and an actual output of the model instances can be used as a basis for determining whether a deviation event has occurred. For example, the first model instance can use measured data to generate the output (first output) and the second model instance can use predicted data to generate the output (second output). If the difference between the output of the model that is generated from the measured data (first output) and the expected output (actual measured data of the output) is greater than the difference between the output of the model that is generated from the predicted data (second output) and the expected output (actual measured data of the output), then it can be determined that a divergence event has occurred. Further, deviation between expected and actual output across model instances can be compared in determining whether a deviation event has occurred. For example, if the deviation between the expected output and the actual output for the first model instance is greater than the deviation between the expected output and the actual output for the second model instance, then it can be determined that a deviation event has occurred. Divergence between outputs of model instances and expected outputs in comparison to a threshold can be used to identify whether a deviation event has occurred.

[0050] The deviation event that is capable of being detected at module 216 can be undetectable by a human. Specifically, a human can be incapable of perceiving the measured data of the first variable, the predicted data of the first variable, the measured data of the second variable, the predicted data of the second variable for the first model instance, the predicted data of the second variable for the second model instance, or a combination thereof to determine the occurrence of the deviation event. More specifically, a human can be incapable of detecting divergence between the expected output and the actual output for the model instances in determining whether the deviation event has occurred. Alternatively, the deviation event capable of being detected at module 216 can be detectable by a human. Specifically, a human can be capable of detecting divergence between the expected output and the actual output for the model instances in determining whether the deviation event has occurred.

[0051] Deviation events can be detected through an applicable technique, including but not limited to a comparison of values to identify divergence. Specifically, loss functions or cost functions can be applied to identify deviations events based on the model validations. For example, mean absolute error techniques, root mean squared error technique, and dynamic time warping techniques can be applied to identify divergence events.

[0052] Regardless of whether the deviation event is detected by a human or not, upon recognition that the deviation event has occurred, a remedy in operation in the downhole environment can be performed. An applicable remedy in operation in the downhole environment can be performed, such as recalibrating a sensor configured to gather the measured data of the first variable, repairing damage to the sensor, replacing a damaged sensor, remedying a down plugging event, making a change in the system (e.g. adjusting a surface choke or electrical submersible pump variable speed controller) to adjust the operational performance for avoidance of continuing deviation events, or a combination thereof. Such remedial measures can be performed as part of a well intervention operation, a well completion operation, or a well production operation. For example, remedial measures can comprise changing operating frequency, changing a proportional-integral-derivative (PID) controller setpoint, changing subroutine parameters, changing surface valving to increase or decrease pressure, performing a tubing or annular flush, changing a chemical treatment plan, or pulling a string out of a wellbore.

[0053] The method represented by the flowchart 200 shown in FIGS. 2A and 2B can be applied to past data to detect previously occurring deviation events. Specifically, the method can be applied within certain timeframes in the past to detect the occurrence of deviation events during such timeframes. For example, the measured data of the first variable and the measured data of the second variable can be accessed from a past timeframe. As follows, the first model instance and the second mode instance can be applied and validated to detect the occurrence of one or more deviation events during the past timeframe.

[0054] FIG. 3A illustrates a plot 300 of predicted downhole sensor data that is generated through application of a first model instance in comparison to measured downhole sensor data. The plot 300 includes predicted downhole sensor data, measured downhole sensor data, and measured surface sensor data. The first model instance is applied to measured surface sensor data to generate the predicted downhole sensor data.

[0055] FIG. 3B illustrates a plot 350 of predicted downhole sensor data that is generated through application of a second model instance in comparison to the measured downhole sensor data. The plot 350 includes predicted downhole sensor data, measured downhole sensor data, and measured surface sensor data. The second model instance is applied to predicted surface sensor data (not shown) to generate the predicted downhole sensor data.

[0056] In comparing the predicted downhole sensor data generated with the first model instance from the measured surface sensor data, plot 300, and the predicted downhole sensor data generated with the second model instance from predicted surface sensor data, plot 350, the predicted downhole sensor data generated with the first model instance deviates more greatly from the measured downhole sensor data. Specifically, the downhole sensor data that is predicted based on measured surface data deviates more greatly from the measured downhole sensor data than the downhole sensor data that is predicted based on predicted surface data. More specifically, the deviation difference is much greater in plot 300 than in plot 350. This deviation occurs within timeframe 302 shown on both plots 300 and 350. This deviation difference can correspond to an occurrence of a deviation event at some point within this timeframe 302. Specifically, a deviation event can occur at the beginning of the timeframe 302, otherwise when the large differences in deviation between the predicted and measured downhole data first occurs between the plots 300 and 350. As discussed previously, this deviation event can include an applicable event that can cause the gathering of inaccurate measurements, e.g. inaccurate surface sensor measurements.

[0057] FIG. 4A and 4B illustrate a flowchart 400 of another example method of identifying events occurring in association with downhole operation through multi-variable validation with predicted and measured data, according to some examples of the present disclosure. The method shown in FIGS. 4A and 4B is provided by way of example, as there are a variety of ways to carry out the method. Additionally, while the example method is illustrated with a particular order of steps, those of ordinary skill in the art will appreciate that FIGS. 4A and 4B and the modules shown therein can be executed in any order and can include fewer or more modules than illustrated. Each module shown in FIGS. 4A and 4B represents one or more steps, processes, methods or routines in the method. The modules will be discussed with respect to the example environments described herein.

[0058] At module 402, a first model instance is validated by comparing data of a second variable that is predicted through the first model instance to measured data of the second variable. Specifically, as in module 202 a set of first measured data of a first variable can be accessed. Then, as in module 208, the first model instance can be applied based on the first measured data of the first variable to generate predicted data of the second variable for the first model instance. As follows and as in module 210, the first model instance can be validated by comparing the predicted data of the second variable for the first model instance to the measured data of the second variable. Applicable techniques for comparing different sets of data can be used in comparing data to validate the models described herein, such as the previously described comparison techniques.

[0059] The first variable can include an applicable variable associated with operation in a downhole environment. Specifically, the first variable can include data that is capable of being gathered by a surface sensor to the downhole environment. Alternatively, the first variable can include data that is capable of being gathered by a subsurface sensor in the downhole environment. For example, measurements of a surface pressure can be accessed. In another example, measurements of an intake pressure at a downhole pump can be accessed.

[0060] The second variable can include an applicable variable associated with operation in a downhole environment. Specifically, the second variable can include data that is capable of being gathered by a surface sensor for the downhole environment. Alternatively, the second variable can include data that is capable of being gathered by a subsurface sensor in the downhole environment.

[0061] The second variable and the first variable can be associated with different locations in the downhole environment. For example, the first variable can be a surface variable that is measured at the surface of the downhole environment. As follows, the second variable can be a subsurface variable that is measured in the downhole environment. Alternatively, the first variable can be a subsurface variable that is measured in the downhole environment. As follows, the second variable can be a surface variable that is measured at the surface of the downhole environment. In alternative embodiments, the first variable and second variable can each be measured in the downhole environment, such as the first variable being measured at the heel of a wellbore and the second variable being measured at the toe, or vice versa.

[0062] At module 404, a second model instance is validated by comparing data of the second variable that is predicted through the second model instance to the measured data of the second variable. Specifically, as in module 204, data of the first variable can be predicted. Then, as in module 212, the second model instance can be applied based on the first predicted data of the first variable to generate predicted data of the second variable for the second model instance. As follows and as in module 214, the second model instance can be validated by comparing the predicted data of the second variable for the second model instance to the measured data of the second variable.

[0063] At module 406, a set of second measured data of the first variable is accessed. The second measured data of the first variable can be different than the first measured data of the first variable that is used in applying the first model instance. Specifically, the second measured data can be measured data of the first variable that is gathered after the first measured data of the first variable is gathered. More specifically, the second measured data of the first variable can be gathered after the occurrence of some kind of event, e.g. a deviation event. For example, a deviation event can be detected, e.g. based on the first model instance and the second model instance. In turn, the deviation event can be remedied, e.g. sensor recalibration, and data of the first variable can be measured again, e.g. after the deviation event has been remedied. Further, the first measured data of the first variable and the second measured data of the first variable can correspond to movement of a sensor to different positions.

[0064] At module 408 second predicted data of the first variable is generated. The predicted data of the first variable that is generated at module 408 can be different from the predicted data of the first variable that is used in applying the second model instance. Specifically, the second predicted data can be predicted for the first variable after the first predicted data of the first variable is generated. Specifically, the second predicted data of the first variable can be generated after the occurrence of some kind of event, e.g. a deviation event. For example, a deviation event can be detected, e.g. based on the first model instance and the second model instance. In turn, the deviation event can be remedied, e.g. sensor recalibration, and data of the first variable can be predicted again, e.g. after the deviation event has been remedied. An applicable model can be used in generating the second predicted data of the first variable, e.g. a physics-based model, a machine learning or artificial intelligence model, or a hybrid model.

[0065] At module 410, a third model instance is applied to the second measured data of the first variable to generate predicted data of the second variable for the third model instance. An applicable model can be applied for the third model instance. At module 412, the third model instance is validated by comparing the predicted data of the second variable for the third model instance to second measured data of the second variable. Specifically, second measured data of the second variable can be gathered that is separate from the first measured data of the second variable that is used to validate the first and second model instances at modules 402 and 404. The second measured data of the second variable can be gathered after the occurrence of some kind of event, e.g. a deviation event. For example, a deviation event can be detected, e.g. based on the first model instance and the second model instance. In turn, the deviation event can be remedied, e.g. sensor recalibration, and data of the second variable can be measured again, e.g. after the deviation event has been remedied. As follows, this second measured data of the second variable can be compared to the predicted output of the second variable for the third model instance to validate the third model instance.

[0066] At module 414, a fourth model instance is applied to the second predicted data of the first variable to generate predicted data of the second variable for the fourth model instance. This is in contrast to the third model instance, where measured data of the first variable is applied to generate predicted data of the second variable. While the accuracy of the measured data will affect the first model instance, since predicted data is applied in the second model instance, its precision can be unaffected by the accuracy of the measured data. An applicable model can be applied for the fourth model instance. At module 416, the fourth model instance is validated by comparing the predicted data of the second variable for the fourth model instance to the second measured data of the second variable.

[0067] At module 418, it is identified whether a second deviation event, e.g. associated with gathering the second measurement data of the first variable, has occurred after the first deviation event. Specifically, whether a second deviation event has occurred can be identified by comparing the validation of the first model instance, the validation of the second model instance, the validation of the third model instance, and the validation of the fourth model instance. The second deviation event can correspond to recalibration of a sensor configured to gather the measured data of the first variable, the sensor malfunctioning in gathering the measured data of the first variable, the sensor being damaged during operation in the downhole environment, a down plugging event, or a combination thereof. For example, the model instances can be compared to determine if sensor measurement data continues to drift after the sensor has been recalibrated.

[0068] The second deviation event can be undetectable by a human in perceiving the second measured data of the first variable, the second predicted data of the first variable, the second measured data of the second variable, the predicted data of the second variable for the third model instance, the predicted data of the second variable for the fourth model instance, or a combination thereof. For example, the deviations between the generated and expected output of the third and fourth model instances can be too small for a human to perceive in a graphical representation. However, by comparing the deviations through a computer system, the occurrence of the deviation event can be recognized.

[0069] In comparing the model instances to determine whether the second deviation event has occurred, similarity of the validations of the model instances, e.g. the deviation between the generated and the expected output of the models, can be indicative of whether the second deviation event has occurred. For example, substantial similarity between the validations of each of the four model instances can be indicative that the second deviation event has not occurred. In another example, the first model instance and second model instance validations can be substantially similar while the third model instance and fourth model instance validations can be substantially similar. As follows, less deviation in the expected and predicted output of the third and fourth model instances in comparison to the first and second model instances can be indicative that the second deviation event has occurred. In another example, the first model instance and second model instance validations can be substantially similar while the third model instance validation can have less deviation between expected and predicted output than the first model instance and second model instance validations. Further in the example, the fourth model instance validation can have even less deviation between expected and predicted output than the third model instance validation. In this scenario, there is a possibility, e.g. within a specific probability range, that the second deviation has occurred.

[0070] The method represented by the flowchart 400 shown in FIGS. 4A and 4B can be applied to past data to detect previously occurring deviation events. Specifically, the method can be applied within certain timeframes in the past to detect the occurrence of deviation events during such timeframe. For example, the measured data of the first variable and the measured data of the second variable can be accessed from a past timeframe. As follows, the first model instance, the second mode instance, the third model instance, and the fourth model instance can be applied and validated to detect the occurrence of one or more deviation events during the past timeframe.

[0071] Deviation events and the variables, both measured and predicted, that are described herein can be presented to a user through a display on a graphical user interface. Specifically, indicia or a representation of an indication of whether a deviation event has occurred can be displayed to a user. In turn, the user can initiate remedial actions in response to the warning.

[0072] Deviation events can be displayed relative to measured and predicted data of variables. Specifically, a deviation event can be highlighted as a deviation between predicted values of variables. Deviation events that are displayed can comprise potential deviation events and true deviation events. A potential deviation event can be an event that is identified relative to a threshold for determining an occurrence of a deviation event. For example, a potential deviation event can comprise that predicted values deviate by a threshold amount. A true deviation event can be an event that is identified as an actual deviation event. The following FIGs. illustrate example displays of variables and example indicators of potential deviation events.

[0073] FIG. 5A illustrates a plot 500 of predicted downhole sensor data that is generated through application of a first model instance in comparison to measured downhole sensor data. The plot 500 includes predicted downhole sensor data of the first model instance, measured downhole sensor data, and measured surface sensor data. The first model instance is applied to measured surface sensor data to generate the predicted downhole sensor data of the first model instance.

[0074] FIG. 5B illustrates a plot 510 of predicted downhole sensor data that is generated through application of a second model instance in comparison to the measured downhole sensor data. The plot 510 includes predicted downhole sensor data of the second model instance, measured downhole sensor data, and measured surface sensor data. The second model instance is applied to predicted surface sensor data (not shown) to generate the predicted downhole sensor data of the second model instance.

[0075] FIG. 5C illustrates a plot 520 of predicted downhole sensor data that is generated through application of a third model instance in comparison to measured downhole sensor data. The plot 520 includes predicted downhole sensor data of the third model instance, measured downhole sensor data, and measured surface sensor data. The third model instance is applied to measured surface sensor data that is gathered after a first deviation event to generate the predicted downhole sensor data of the third model instance.

[0076] FIG. 5D illustrates a plot 530 of predicted downhole sensor data that is generated through application of a fourth model instance in comparison to the measured downhole sensor data. The plot 530 includes predicted downhole sensor data of the fourth model instance, measured downhole sensor data, and measured surface sensor data. The fourth model instance is applied to predicted surface sensor data (not shown) that is generated after the first deviation event to generate the predicted downhole sensor data of the fourth model instance.

[0077] In comparing the predicted downhole sensor data generated with the first model instance from the measured surface sensor data, plot 500, and the predicted downhole sensor data generated with the second model instance from predicted surface sensor data, plot 510, during corresponding timeframes 502 and 512, the predicted downhole sensor data and the measured downhole sensor data for both the first model instance and the second model instance have deviated greatly. The amount of deviation can be quantified as less than or equal to 10% deviation, between 10% and 25% deviation, or greater than or equal to 25% deviation. In various embodiments, this deviation is so great that an operator is capable of visualizing this while monitoring operations in the downhole environment. This large deviation between the expected output and the actual output of the first and second model instances corresponds to a first deviation event.

[0078] After this first deviation event, a remedial event can be performed, e.g. recalibration, and new surface sensor data can be gathered and applied in the third model instance. Similarly, after the first deviation event and the remedial event, new surface sensor data can be predicted and applied in the fourth model instance. Each model instance can continue to run after the deviation event, with the first and second model instances using corresponding measured surface sensor data and predicted surface sensor data from before the deviation event.

[0079] Each model instance has a corresponding timeframe 504, 514, 524, and 534 after the first deviation event. As shown in the timeframes 504 and 514, both the first and second model instances are still exhibiting large deviations between the predicted downhole sensor data, predicted output, and the measured downhole sensor data, expected output. As shown in timeframes 524 and 534, the third and fourth model instances are exhibiting less deviation between the predicted output and the expected output. Further, it appears that the fourth model instance is exhibiting even less deviation between predicted and expected output than the third model instance. This could indicate the possible occurrence of a second deviation event after the first deviation event.

[0080] The detection of this possible second deviation event can be used in controlling operations associated with the downhole environment. Specifically, a remedial event can be performed in response to the second deviation event. Further, other wellbore actions that are not intended to remedy the second deviation event can be controlled and / or initiated based on the second deviation event.

[0081] While four model instances are applied in the embodiments of the technology represented in FIGS. 4A-5D, more or fewer model instances can be applied to implement the technology described herein.

[0082] The technology described herein can be performed across a plurality of wellbores, e.g. as part of field operations. Specifically variables can be monitored across a plurality of wells. In turn, a deviation event can be detected in one or more of the wells. As follows, an operation, e.g. remedial action, can be performed in one or more of the wells in response to the deviation event. For example, electrical submersible pumps in a plurality of wells can be monitored. A deviation event in relation to operation of one or more of the electrical submersible pumps can be detected in one or more of the wells. As follows, an operation can be performed in the one or more wells in response to the deviation event.

[0083] In FIG. 6, the disclosure now turns to a further discussion of models that can be used to implement the technology described herein. FIG. 6 is an example of a deep learning neural network 600 that can be used to implement all or a portion of the systems and techniques described herein, according to some examples of the present disclosure. An input layer 620 can be configured to receive data. Neural network 600 includes multiple hidden layers 622a, 622b, through 622n. The hidden layers 622a, 622b, through 622n include “n” number of hidden layers, where “n” is an integer greater than or equal to one. The number of hidden layers can be made to include as many layers as needed for the given application. Neural network 600 further includes an output layer 621 that provides an output resulting from the processing performed by the hidden layers 622a, 622b, through 622n.

[0084] Neural network 600 is a multi-layer neural network of interconnected nodes. Each node can represent a piece of information. Information associated with the nodes is shared among the different layers and each layer retains information as information is processed. In some cases, the neural network 600 can include a feed-forward network, in which case there are no feedback connections where outputs of the network are fed back into itself. In some cases, the neural network 600 can include a recurrent neural network, which can have loops that allow information to be carried across nodes while reading in input.

[0085] Information can be exchanged between nodes through node-to-node interconnections between the various layers. Nodes of the input layer 620 can activate a set of nodes in the first hidden layer 622a. For example, as shown, each of the input nodes of the input layer 620 is connected to each of the nodes of the first hidden layer 622a. The nodes of the first hidden layer 622a can transform the information of each input node by applying activation functions to the input node information. The information derived from the transformation can then be passed to and can activate the nodes of the next hidden layer 622b, which can perform their own designated functions. Example functions include convolutional, up-sampling, data transformation, and / or any other suitable functions. The output of the hidden layer 622b can then activate nodes of the next hidden layer, and so on. The output of the last hidden layer 622n can activate one or more nodes of the output layer 621, at which an output is provided. In some cases, while nodes in the neural network 600 are shown as having multiple output lines, a node can have a single output and all lines shown as being output from a node represent the same output value.

[0086] In some cases, each node or interconnection between nodes can have a weight that is a set of parameters derived from the training of the neural network 600. Once the neural network 600 is trained, it can be referred to as a trained neural network, which can be used to classify one or more activities. For example, an interconnection between nodes can represent a piece of information learned about the interconnected nodes. The interconnection can have a tunable numeric weight that can be tuned (e.g., based on a training dataset), allowing the neural network 600 to be adaptive to inputs and able to learn as more and more data is processed.

[0087] The neural network 600 is pre-trained to process the features from the data in the input layer 620 using the different hidden layers 622a, 622b, through 622n in order to provide the output through the output layer 621.

[0088] In some cases, the neural network 600 can adjust the weights of the nodes using a training process called backpropagation. A backpropagation process can include a forward pass, a loss function, a backward pass, and a weight update. The forward pass, loss function, backward pass, and parameter / weight update is performed for one training iteration. The process can be repeated for a certain number of iterations for each set of training data until the neural network 600 is trained well enough so that the weights of the layers are accurately tuned.

[0089] To perform training, a loss function can be used to analyze error in the output. Any suitable loss function definition can be used, such as a Cross-Entropy loss. Another example of a loss function includes the mean squared error (MSE), defined as E_total = ∑(1 / 2 (target-output)^2). The loss can be set to be equal to the value of E_total.

[0090] The loss (or error) will be high for the initial training data since the actual values / training output will be much different than the predicted output. The goal of training is to minimize the amount of loss so that the predicted output is the same as the training output. The neural network 600 can perform a backward pass by determining which inputs (weights) most contributed to the loss of the network, and can adjust the weights so that the loss decreases and is eventually minimized.

[0091] The neural network 600 can include any suitable deep network. One example includes a Convolutional Neural Network (CNN), which includes an input layer and an output layer, with multiple hidden layers between the input and out layers. The hidden layers of a CNN include a series of convolutional, nonlinear, pooling (for downsampling), and fully connected layers. The neural network 600 can include any other deep network other than a CNN, such as an autoencoder, Deep Belief Nets (DBNs), Recurrent Neural Networks (RNNs), among others.

[0092] As understood by those of skill in the art, machine-learning based classification techniques can vary depending on the desired implementation. For example, machine-learning classification schemes can utilize one or more of the following, alone or in combination: hidden Markov models; RNNs; CNNs; deep learning; Bayesian symbolic methods; Generative Adversarial Networks (GANs); support vector machines; image registration methods; and applicable rule-based systems. Where regression algorithms are used, they may include but are not limited to: a Stochastic Gradient Descent Regressor, a Passive Aggressive Regressor, etc.

[0093] Machine learning classification models can also be based on clustering algorithms (e.g., a Mini-batch K-means clustering algorithm), a recommendation algorithm (e.g., a Minwise Hashing algorithm, or Euclidean Locality-Sensitive Hashing (LSH) algorithm), and / or an anomaly detection algorithm, such as a local outlier factor. Additionally, machine-learning models can employ a dimensionality reduction approach, such as, one or more of: a Mini-batch Dictionary Learning algorithm, an incremental Principal Component Analysis (PCA) algorithm, a Latent Dirichlet Allocation algorithm, and / or a Mini-batch K-means algorithm, etc.

[0094] FIG. 7 illustrates an example processor-based system with which some aspects of the subject technology can be implemented. For example, processor-based system 700 can be any computing device, or any component thereof, in which the components of the system are in communication with each other using connection 705. Connection 705 can be a physical connection via a bus, or a direct connection into processor 710, such as in a chipset architecture. Connection 705 can also be a virtual connection, networked connection, or logical connection.

[0095] In some embodiments, computing system 700 is a distributed system in which the functions described in this disclosure can be distributed within a datacenter, multiple data centers, a peer network, etc. In some embodiments, one or more of the described system components represents many such components each performing some or all of the function for which the component is described. In some embodiments, the components can be physical or virtual devices.

[0096] Example system 700 includes at least one processing unit (Central Processing Unit (CPU) or processor) 710 and connection 705 that couples various system components including system memory 715, such as Read-Only Memory (ROM) 720 and Random-Access Memory (RAM) 725 to processor 710. Computing system 700 can include a cache of high-speed memory 712 connected directly with, in close physical proximity to, or integrated as part of processor 710.

[0097] Processor 710 can include any general-purpose processor and a hardware service or software service, such as services 732, 734, and 736 stored in storage device 730, configured to control processor 710 as well as a special-purpose processor where software instructions are incorporated into the actual processor design. Processor 710 may be a completely self-contained computing system, containing multiple cores or processors, a bus, memory controller, cache, etc. A multi-core processor may be symmetric or asymmetric. Further, the processor 710 can be a distributed computing system.

[0098] To enable user interaction, computing system 700 includes an input device 745, which can represent any number of input mechanisms, such as a microphone for speech, a touch-sensitive screen for gesture or graphical input, keyboard, mouse, motion input, speech, etc. Computing system 700 can also include output device 735, which can be one or more of a number of output mechanisms known to those of skill in the art. In some instances, multimodal systems can enable a user to provide multiple types of input / output to communicate with computing system 700. Computing system 700 can include communications interface 740, which can generally govern and manage the user input and system output. The communication interface may perform or facilitate receipt and / or transmission wired or wireless communications via wired and / or wireless transceivers, including those making use of an audio jack / plug, a microphone jack / plug, a Universal Serial Bus (USB) port / plug, an Apple® Lightning® port / plug, an Ethernet port / plug, a fiber optic port / plug, a proprietary wired port / plug, a BLUETOOTH® wireless signal transfer, a BLUETOOTH® low energy (BLE) wireless signal transfer, an IBEACON® wireless signal transfer, a Radio-Frequency Identification (RFID) wireless signal transfer, Near-Field Communications (NFC) wireless signal transfer, Dedicated Short Range Communication (DSRC) wireless signal transfer, 802.11 Wi-Fi® wireless signal transfer, Wireless Local Area Network (WLAN) signal transfer, Visible Light Communication (VLC) signal transfer, Worldwide Interoperability for Microwave Access (WiMAX), Infrared (IR) communication wireless signal transfer, Public Switched Telephone Network (PSTN) signal transfer, Integrated Services Digital Network (ISDN) signal transfer, 3G / 4G / 5G / LTE cellular data network wireless signal transfer, ad-hoc network signal transfer, radio wave signal transfer, microwave signal transfer, infrared signal transfer, visible light signal transfer signal transfer, ultraviolet light signal transfer, wireless signal transfer along the electromagnetic spectrum, or some combination thereof.

[0099] Communication interface 740 may also include one or more Global Navigation Satellite System (GNSS) receivers or transceivers that are used to determine a location of the computing system 700 based on receipt of one or more signals from one or more satellites associated with one or more GNSS systems. GNSS systems include, but are not limited to, the US-based Global Positioning System (GPS), the Russia-based Global Navigation Satellite System (GLONASS), the China-based BeiDou Navigation Satellite System (BDS), and the Europe-based Galileo GNSS. There is no restriction on operating on any particular hardware arrangement, and therefore the basic features here may easily be substituted for improved hardware or firmware arrangements as they are developed.

[0100] Storage device 730 can be a non-volatile and / or non-transitory and / or computer-readable memory device and can be a hard disk or other types of computer readable media which can store data that are accessible by a computer, such as magnetic cassettes, flash memory cards, solid state memory devices, digital versatile disks, cartridges, a floppy disk, a flexible disk, a hard disk, magnetic tape, a magnetic strip / stripe, any other magnetic storage medium, flash memory, memristor memory, any other solid-state memory, a Compact Disc (CD) Read Only Memory (CD-ROM) optical disc, a rewritable CD optical disc, a Digital Video Disk (DVD) optical disc, a Blu-ray Disc (BD) optical disc, a holographic optical disk, another optical medium, a Secure Digital (SD) card, a micro SD (microSD) card, a Memory Stick® card, a smartcard chip, a EMV chip, a Subscriber Identity Module (SIM) card, a mini / micro / nano / pico SIM card, another Integrated Circuit (IC) chip / card, Random-Access Memory (RAM), Atatic RAM (SRAM), Dynamic RAM (DRAM), Read-Only Memory (ROM), Programmable ROM (PROM), Erasable PROM (EPROM), Electrically Erasable PROM (EEPROM), flash EPROM (FLASHEPROM), cache memory (L1 / L2 / L3 / L4 / L5 / L#), Resistive RAM (RRAM / ReRAM), Phase Change Memory (PCM), Spin Transfer Torque RAM (STT-RAM), another memory chip or cartridge, and / or a combination thereof.

[0101] Storage device 730 can include software services, servers, services, etc., that when the code that defines such software is executed by the processor 710, it causes the system 700 to perform a function. In some embodiments, a hardware service that performs a particular function can include the software component stored in a computer-readable medium in connection with the necessary hardware components, such as processor 710, connection 705, output device 735, etc., to carry out the function.

[0102] For clarity of explanation, in some instances the present technology may be presented as including individual functional blocks including functional blocks comprising devices, device components, steps or routines in a method embodied in software, or combinations of hardware and software.

[0103] In some embodiments the computer-readable storage devices, mediums, and memories can include a cable or wireless signal containing a bit stream and the like. However, when mentioned, non-transitory computer-readable storage media expressly exclude media such as energy, carrier signals, electromagnetic waves, and signals per se.

[0104] Methods according to the above-described examples can be implemented using computer-executable instructions that are stored or otherwise available from computer readable media. Such instructions can include, for example, instructions and data which cause or otherwise configure a general purpose computer, special purpose computer, or a processing device to perform a certain function or group of functions. Portions of computer resources used can be accessible over a network. The computer executable instructions may be, for example, binaries, intermediate format instructions such as assembly language, firmware, source code, etc. Examples of computer-readable media that may be used to store instructions, information used, and / or information created during methods according to described examples include magnetic or optical disks, flash memory, USB devices provided with non-volatile memory, networked storage devices, and so on.

[0105] Devices implementing methods according to these disclosures can include hardware, firmware and / or software, and can take any of a variety of form factors. Typical examples of such form factors include laptops, smart phones, small form factor personal computers, personal digital assistants, rackmount devices, standalone devices, and so on. Functionality described herein also can be embodied in peripherals or add-in cards. Such functionality can also be implemented on a circuit board among different chips or different processes executing in a single device, by way of further example.

[0106] The instructions, media for conveying such instructions, computing resources for executing them, and other structures for supporting such computing resources are example means for providing the functions described in the disclosure.

[0107] In the foregoing description, aspects of the application are described with reference to specific embodiments thereof, but those skilled in the art will recognize that the application is not limited thereto. Thus, while illustrative embodiments of the application have been described in detail herein, it is to be understood that the disclosed concepts may be otherwise variously embodied and employed, and that the appended claims are intended to be construed to include such variations, except as limited by the prior art. Various features and aspects of the above-described subject matter may be used individually or jointly. Further, embodiments can be utilized in any number of environments and applications beyond those described herein without departing from the broader spirit and scope of the specification. The specification and drawings are, accordingly, to be regarded as illustrative rather than restrictive. For the purposes of illustration, methods were described in a particular order. It should be appreciated that in alternate embodiments, the methods may be performed in a different order than that described.

[0108] Where components are described as being “configured to” perform certain operations, such configuration can be accomplished, for example, by designing electronic circuits or other hardware to perform the operation, by programming programmable electronic circuits (e.g., microprocessors, or other suitable electronic circuits) to perform the operation, or any combination thereof.

[0109] The various illustrative logical blocks, modules, circuits, and algorithm steps described in connection with the examples disclosed herein may be implemented as electronic hardware, computer software, firmware, or combinations thereof. To clearly illustrate this interchangeability of hardware and software, various illustrative components, blocks, modules, circuits, and steps have been described above generally in terms of their functionality. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the overall system. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.

[0110] The techniques described herein may also be implemented in electronic hardware, computer software, firmware, or any combination thereof. Such techniques may be implemented in any of a variety of devices such as general purposes computers, wireless communication device handsets, or integrated circuit devices having multiple uses including application in wireless communication device handsets and other devices. Any features described as modules or components may be implemented together in an integrated logic device or separately as discrete but interoperable logic devices. If implemented in software, the techniques may be realized at least in part by a computer-readable data storage medium comprising program code including instructions that, when executed, performs one or more of the method, algorithms, and / or operations described above. The computer-readable data storage medium may form part of a computer program product, which may include packaging materials.

[0111] The computer-readable medium may include memory or data storage media, such as random access memory (RAM) such as synchronous dynamic random access memory (SDRAM), read-only memory (ROM), non-volatile random access memory (NVRAM), electrically erasable programmable read-only memory (EEPROM), FLASH memory, magnetic or optical data storage media, and the like. The techniques additionally, or alternatively, may be realized at least in part by a computer-readable communication medium that carries or communicates program code in the form of instructions or data structures and that can be accessed, read, and / or executed by a computer, such as propagated signals or waves.

[0112] Other embodiments of the disclosure may be practiced in network computing environments with many types of computer system configurations, including personal computers, hand-held devices, multi-processor systems, microprocessor-based or programmable consumer electronics, network PCs, minicomputers, mainframe computers, and the like. Embodiments may also be practiced in distributed computing environments where tasks are performed by local and remote processing devices that are linked (either by hardwired links, wireless links, or by a combination thereof) through a communications network. In a distributed computing environment, program modules may be located in both local and remote memory storage devices.

[0113] In the above description, terms such as "upper," "upward," "lower," "downward," "above," "below," "downhole," "uphole," "longitudinal," "lateral," and the like, as used herein, shall mean in relation to the bottom or furthest extent of the surrounding wellbore even though the wellbore or portions of it may be deviated or horizontal. Correspondingly, the transverse, axial, lateral, longitudinal, radial, etc., orientations shall mean orientations relative to the orientation of the wellbore or tool. Additionally, the illustrate embodiments are illustrated such that the orientation is such that the right-hand side is downhole compared to the left-hand side.

[0114] The term "coupled" is defined as connected, whether directly or indirectly through intervening components, and is not necessarily limited to physical connections. The connection can be such that the objects are permanently connected or releasably connected. The term "outside" refers to a region that is beyond the outermost confines of a physical object. The term "inside" indicates that at least a portion of a region is partially contained within a boundary formed by the object. The term "substantially" is defined to be essentially conforming to the particular dimension, shape or another word that substantially modifies, such that the component need not be exact. For example, substantially cylindrical means that the object resembles a cylinder, but can have one or more deviations from a true cylinder.

[0115] The term "axially" means substantially along a direction of the axis of the object. If not specified, the term axially is such that it refers to the longer axis of the object.

[0116] Although a variety of information was used to explain aspects within the scope of the appended claims, no limitation of the claims should be implied based on particular features or arrangements, as one of ordinary skill would be able to derive a wide variety of implementations. Further and although some subject matter may have been described in language specific to structural features and / or method steps, it is to be understood that the subject matter defined in the appended claims is not necessarily limited to these described features or acts. Such functionality can be distributed differently or performed in components other than those identified herein. The described features and steps are disclosed as possible components of systems and methods within the scope of the appended claims.

[0117] Claim language or other language in the disclosure reciting “at least one of” a set and / or “one or more” of a set indicates that one member of the set or multiple members of the set (in any combination) satisfy the claim. For example, claim language reciting “at least one of A and B” or “at least one of A or B” means A, B, or A and B. In another example, claim language reciting “at least one of A, B, and C” or “at least one of A, B, or C” means A, B, C, or A and B, or A and C, or B and C, or A and B and C. The language “at least one of” a set and / or “one or more” of a set does not limit the set to the items listed in the set. For example, claim language reciting “at least one of A and B” or “at least one of A or B” can mean A, B, or A and B, and can additionally include items not listed in the set of A and B.

[0118] Illustrative examples of the disclosure include:

[0119] Aspect 1. A computer-implemented method implemented through one or more processors executing instructions stored on a non-transitory computer-readable storage medium, the method comprising: accessing a first set of measured data of a first variable associated with an operation in a downhole environment; generating a first set of predicted data of the first variable; accessing a first set of measured data of a second variable associated with the operation in the downhole environment; applying a first model instance to the first set of measured data of the first variable to generate a first set of predicted data of the second variable; validating the first model instance by comparing the first set of predicted data of the second variable to the first set of measured data of the second variable; applying a second model instance to the first set of predicted data of the first variable to generate a second set of predicted data of the second variable; validating the second model instance by comparing the second set of predicted data of the second variable to the first set of measured data of the second variable; and determining whether a deviation event associated with gathering the first set of measured data of the first variable has occurred by comparing the validation of the first model instance with the validation of the second model instance.

[0120] Aspect 2. The computer-implemented method of Aspect 1, wherein the first variable associated with the operation in the downhole environment is a surface variable and the second variable associated with the operation in the downhole environment is a subsurface variable.

[0121] Aspect 3. The computer-implemented method of either Aspect 1 or 2, wherein the first variable associated with the operation in the downhole environment is a subsurface variable and the second variable associated with the operation in the downhole environment is a surface variable.

[0122] Aspect 4. The computer-implemented method of any of Aspects 1 through 3, wherein the deviation event corresponds to malfunctioning of a sensor or recalibration of a sensor.

[0123] Aspect 5. The computer-implemented method of any of Aspects 1 through 4, wherein the deviation event is undetectable by a human in perceiving the first set of measured data of the first variable, the first set of predicted data of the first variable, the first set of measured data of the second variable, the first set of predicted data of the second variable for the first model instance, the second set of predicted data of the second variable for the second model instance, or a combination thereof.

[0124] Aspect 6. The computer-implemented method of any of Aspects 1 through 5, further comprising performing a remedy in operation in the downhole environment in response to detection of the deviation event, wherein the remedy in operation in the downhole environment comprises recalibrating a sensor configured to gather the measured data of the first variable, repairing the sensor, remedying a down plugging event, or a combination thereof.

[0125] Aspect 7. The computer-implemented method of any of Aspects 1 through 6, wherein the first set of measured data of the first variable and the first set of measured data of the second variable are accessed from a past timeframe, and the first model instance and the second model instance are applied and validated to detect an occurrence of the deviation event as a past event during the past timeframe.

[0126] Aspect 8. The computer-implemented method of any of Aspects 1 through 7, further comprising: accessing a second set of measured data of the first variable; generating a second set of predicted data of the first variable; accessing a second set of measured data of the second variable; applying a third model instance to the second set of measured data of the first variable to generate a third set of predicted data of the second variable; validating the third model instance by comparing the third set of predicted data of the second variable to the second set of measured data of the second variable; applying a fourth model instance to the second set of predicted data of the first variable to generate a fourth set of predicted data of the second variable; validating the fourth model instance by comparing the fourth set of predicted data of the second variable to the second set of measured data of the second variable; and determining whether a second deviation event associated with gathering the second set of measured data of the first variable has occurred after the deviation event by comparing the validation of the third model instance with the validation of the fourth model instance.

[0127] Aspect 9. The computer-implemented method of Aspect 8, further comprising: wherein the second set of measured data of the first variable and the second set of measured data of the second variable are measured after a remedial action is performed in response to detection of the deviation event and the second set of predicted data of the first variable corresponds to after performance of the remedial action, the method further comprising: applying the third model instance to the second set of measured data of the first variable to generate the third set of predicted data of the second variable after performance of the remedial action; validating the third model instance by comparing the third set of the predicted data of the second variable to the second set of measured data of the second variable; applying the fourth model instance to the second set of predicted data of the first variable to generate the fourth set of predicted data of the second variable after performance of the remedial action; validating the fourth model instance by comparing the fourth set of predicted data of the second variable to the second set of measured data of the second variable; and identifying whether the second deviation event associated with gathering the second measured data of the first variable has occurred after performance of the remedial action by comparing the validation of the third model instance with the validation of the fourth model instance.

[0128] Aspect 10. The computer-implemented method of either Aspect 8 or 9, wherein the second deviation event corresponds to recalibration of a sensor configured to gather the measured data of the first variable, the sensor malfunctioning in gathering the measured data of the first variable, the sensor being damaged during operation in the downhole environment, a down plugging event, or a combination thereof.

[0129] Aspect 11. The computer-implemented method of any of Aspects 8 through 10, wherein the second deviation event is undetectable by a human in perceiving the second set of measured data of the first variable, the second set of predicted data of the first variable, the second set of measured data of the second variable, the third set of predicted data of the second variable, the fourth set of predicted data of the second variable for the fourth model instance, or a combination thereof.

[0130] Aspect 12. The computer-implemented method of any of Aspects 8 through 11, further comprising determining whether the second deviation event has occurred by comparing the validation of the first model instance, the validation of the second model instance, the validation of the third model instance, and the validation of the fourth model instance.

[0131] Aspect 13. The computer-implemented method of Aspect 12, further comprising: identifying that the validation of the first model instance, the validation of the second model instance, the validation of the third model instance, and the validation of the fourth model instance are substantially similar; and identifying that the second deviation event has not occurred based on the substantial similarity between the validation of the first model instance, the validation of the second model instance, the validation of the third model instance, and the validation of the fourth model instance.

[0132] Aspect 14. The computer-implemented method of Aspect 12, further comprising: identifying that the validation of the first model instance and the validation of the second model instance are substantially similar and that the validation of the third model instance and the validation of the fourth model instance are substantially similar; comparing the validation of the first model instance and the validation of the second model instance with the validation of the third model instance and the validation of the fourth model instance; and identifying an occurrence of the second deviation event based on amounts that the third set of predicted data of the second variable and the fourth set of predicted data of the second variable deviate from the second set of measured data of the second variable in comparison to amounts that the first set of predicted data of the second variable and the second set of predicted data of the second variable deviate from the first set of measured data of the second variable.

[0133] Aspect 15. The computer-implemented method of Aspect 14, further comprising identifying the occurrence of the second deviation event if the amounts that the third set of predicted data of the second variable and the fourth set of predicted data of the second variable deviate from the second set of measured data of the second variable are less than the amounts that the first set of predicted data of the second variable and the second set of predicted data of the second variable deviate from the first set of measured data of the second variable.

[0134] Aspect 16. The computer-implemented method of any of Aspects 12 through 15, further comprising: identifying that the validation of the first model instance and the validation of the second model instance are substantially similar; identifying that an amount that the third set of predicted data of the second variable deviates from the second set of measured data of the second variable is less than amounts that the first set of predicted data of the second variable and the second set of predicted data of the second variable deviate from the first set of measured data of the second variable; identifying that an amount that fourth set of predicted data of the second variable deviates from the second set of measured data of the second variable is less than the amount that the third set of predicted data of the second variable deviates from the second set of measured data of the second variable; and identifying a possible occurrence of the second deviation event based on that the amount that the fourth set of predicted data of the second variable deviates from the second set of measured data of the second variable deviate is less than the amount that the third set of predicted data of the second variable deviates from the second set of measured data of the second variable.

[0135] Aspect 17. The computer-implemented method of any of Aspects 8 through 16, wherein the first set of measured data of the first variable, the first set of measured data of the second variable, the second set of measured data of the first variable, and the second set of measured data of the second variable are accessed from a past timeframe, and the first model instance, the second model instance, the third model instance, and the fourth model instance are applied and validated to detect an occurrence of the deviation event and the second deviation event as past events during the past timeframe.

[0136] Aspect 18. A system comprising: one or more processors; and at least one computer-readable storage medium having stored therein instructions which, when executed by the one or more processors, cause the one or more processors to: access a first set of measured data of a first variable associated with an operation in a downhole environment; generate a first set of predicted data of the first variable; access a first set of measured data of a second variable associated with the operation in the downhole environment; apply a first model instance to the first set of measured data of the first variable to generate a first set of predicted data of the second variable; validate the first model instance by comparing the first set of predicted data of the second variable to the first set of measured data of the second variable; apply a second model instance to the first set of predicted data of the first variable to generate a second set of predicted data of the second variable; validate the second model instance by comparing the second set of predicted data of the second variable to the first set of measured data of the second variable; and determine whether a deviation event associated with gathering the first set of measured data of the first variable has occurred by comparing the validation of the first model instance with the validation of the second model instance.

[0137] Aspect 19. The system of Aspect 18, wherein the instructions are further configured to cause the one or more processors to: access a second set of measured data of the first variable; generate a second set of predicted data of the first variable; access a second set of measured data of the second variable; apply a third model instance to the second set of measured data of the first variable to generate a third set of predicted data of the second variable; validate the third model instance by comparing the third set of predicted data of the second variable to the second set of measured data of the second variable; apply a fourth model instance to the second set of predicted data of the first variable to generate a fourth set of predicted data of the second variable; validate the fourth model instance by comparing the fourth set of predicted data of the second variable to the second set of measured data of the second variable; and determine whether a second deviation event associated with gathering the second set of measured data of the first variable has occurred after the deviation event by comparing the validation of the third model instance with the validation of the fourth model instance.

[0138] Aspect 20. A non-transitory computer-readable storage medium storing instructions for causing one or more processors to: access a first set of measured data of a first variable associated with an operation in a downhole environment; generate a first set of predicted data of the first variable; access a first set of measured data of a second variable associated with the operation in the downhole environment; apply a first model instance to the first set of measured data of the first variable to generate a first set of predicted data of the second variable; validate the first model instance by comparing the first set of predicted data of the second variable to the first set of measured data of the second variable; apply a second model instance to the first set of predicted data of the first variable to generate a second set of predicted data of the second variable; validate the second model instance by comparing the second set of predicted data of the second variable to the first set of measured data of the second variable; and determine whether a deviation event associated with gathering the first set of measured data of the first variable has occurred by comparing the validation of the first model instance with the validation of the second model instance.

[0139] Aspect 21. A method comprising: disposing an electrical submersible pump in a downhole environment; configuring the electrical submersible pump to operate in the downhole environment during a wellbore operation; accessing a first set of measured data of a first variable associated with the wellbore operation in the downhole environment; generating a first set of predicted data of the first variable; accessing a first set of measured data of a second variable associated with the wellbore operation in the downhole environment; applying a first model instance to the first set of measured data of the first variable to generate a first set of predicted data of the second variable; validating the first model instance by comparing the first set of predicted data of the second variable to the first set of measured data of the second variable; applying a second model instance to the first set of predicted data of the first variable to generate a second set of predicted data of the second variable; validating the second model instance by comparing the second set of predicted data of the second variable to the first set of measured data of the second variable; and determining whether a deviation event associated with gathering the first set of measured data of the first variable has occurred by comparing the validation of the first model instance with the validation of the second model instance.

[0140] Aspect 22. The method of Aspect 21, wherein the deviation event is associated with operation of the electrical submersible pump during the wellbore operation.

[0141] Aspect 23. The method of Aspect 22, wherein either or both the first variable and the second variable are associated with operation of the electrical submersible pump and comprise a pump intake pressure, a pump intake temperature, a motor temperature, a pump discharge pressure, vibration in a y-direction in association with operation of the electrical submersible pump, vibration in a x-direction in association with operation of the electrical submersible pump, vibration in a z-direction in association with operation of the electrical submersible pump or a combination thereof.

[0142] Aspect 24. A method comprising: performing a wellbore operation in a downhole environment; accessing a first set of measured data of a first variable associated with the wellbore operation in the downhole environment; generating a first set of predicted data of the first variable; accessing a first set of measured data of a second variable associated with the wellbore operation in the downhole environment; applying a first model instance to the first set of measured data of the first variable to generate a first set of predicted data of the second variable; validating the first model instance by comparing the first set of predicted data of the second variable to the first set of measured data of the second variable; applying a second model instance to the first set of predicted data of the first variable to generate a second set of predicted data of the second variable; validating the second model instance by comparing the second set of predicted data of the second variable to the first set of measured data of the second variable; and determining whether a deviation event associated with gathering the first set of measured data of the first variable has occurred by comparing the validation of the first model instance with the validation of the second model instance.

[0143] Aspect 25. The method of Aspect 24, wherein the deviation event is associated with performance of the wellbore operation.

[0144] Aspect 26. The method of either of Aspects 24 or 25, wherein either or both the first variable and the second variable comprise a three-phase flow rate and the wellbore operation is a well production operation.

[0145] Aspect 27. The method of any of Aspects 24 through 26, wherein either or both the first variable and the second variable comprise a solid, sand, or gas migration rate.

[0146] Aspect 28. The method of any of Aspects 24 through 27, wherein either or both the first variable and the second variable comprise a frequency component of a control signal for controlling either or both the speed and the torque of a motor in the wellbore operation.

[0147] Aspect 29. The method of any of Aspects 24 through 28, wherein either or both the first variable and the second variable comprise an output current of a variable speed drive used in the wellbore operation.

[0148] Aspect 30. The method of any of Aspects 24 through 29, wherein either or both the first variable and the second variable comprise an output voltage of a variable speed drive used in the wellbore operation.

[0149] Aspect 31. The method of any of Aspects 24 through 30, wherein either or both the first variable and the second variable comprise a tubing pressure of tubing used in the downhole environment during the wellbore operation.

[0150] Aspect 32. The method of any of Aspects 24 through 31, wherein either or both the first variable and the second variable comprise a casing pressure between a wellbore and casing in the downhole environment or an annulus pressure between production tubing and the casing.

[0151] Aspect 33. The method of any of Aspects 24 through 32, wherein either or both the first variable and the second variable comprise a flowline pressure within a flowline used during the wellbore operation.

[0152] Aspect 34. The method of any of Aspects 24 through 33, wherein either or both the first set of measured data of the first variable and the first set of measured data of the second variable are received from the downhole environment through a wireline or fiber telemetry system.

[0153] Aspect 35. The method of any of Aspects 24 through 34, further comprising recalibrating one or more downhole sensors in response to the deviation event through a wireline or fiber telemetry system.

[0154] Aspect 36. A method comprising: accessing a first set of measured data of a first variable associated with an operation in a downhole environment; generating a first set of predicted data of the first variable; accessing a first set of measured data of a second variable associated with the operation in the downhole environment; applying a first model instance to the first set of measured data of the first variable to generate a first set of predicted data of the second variable; validating the first model instance by comparing the first set of predicted data of the second variable to the first set of measured data of the second variable; applying a second model instance to the first set of predicted data of the first variable to generate a second set of predicted data of the second variable; validating the second model instance by comparing the second set of predicted data of the second variable to the first set of measured data of the second variable; determining whether a deviation event associated with gathering the first set of measured data of the first variable has occurred by comparing the validation of the first model instance with the validation of the second model instance; and outputting through a graphical user interface a display indicating whether the deviation event has occurred.

[0155] Aspect 37. The method of Aspect 36, further comprising outputting the display comprising the first set of measured data of the first variable, the first set of predicted data of the first variable, the first set of measured data of the second variable, the first set of predicted data of the second variable, the second set of predicted data of the second variable, or a combination thereof.

[0156] Aspect 38. The method of either of Aspects 36 or 37, further comprising indicating on the display an occurrence of the deviation event relative to the first set of measured data of the first variable, the first set of predicted data of the first variable, the first set of measured data of the second variable, the first set of predicted data of the second variable, the second set of predicted data of the second variable, or a combination thereof.

[0157] Aspect 39. The method of any of Aspects 36 through 38, further comprising indicating on the display a plurality of potential deviation events and an indication of whether the deviation event is a true deviation event or a potential deviation event of the plurality of deviation events.

[0158] Aspect 40. A method comprising: accessing a first set of measured data of a first variable associated with performing an operation in a downhole environment; generating a first set of predicted data of the first variable; accessing a first set of measured data of a second variable associated with performing the operation in the downhole environment; applying a first model instance to the first set of measured data of the first variable to generate a first set of predicted data of the second variable; validating the first model instance by comparing the first set of predicted data of the second variable to the first set of measured data of the second variable; applying a second model instance to the first set of predicted data of the first variable to generate a second set of predicted data of the second variable; validating the second model instance by comparing the second set of predicted data of the second variable to the first set of measured data of the second variable; determining whether a deviation event associated with gathering the first set of measured data of the first variable has occurred by comparing the validation of the first model instance with the validation of the second model instance; and performing a well intervention operation in response to determining that the deviation event has occurred.

[0159] Aspect 41. The method of Aspect 40, wherein the operation comprises functioning of an electrical submersible pump disposed within a wellbore of the downhole environment.

[0160] Aspect 42. The method of either of Aspects 40 or 41, wherein the well intervention operation comprises replacing the electrical submersible pump, replacing a sensor associated with functioning of the electrical submersible pump, adjusting an operational parameter of the electrical submersible pump, recalibrating the sensor associated with functioning of the electrical submersible pump, or a combination thereof.

[0161] Aspect 43. A method comprising: monitoring wellbore operations of a plurality of electrical submersible pumps disposed in a plurality of downhole environment; accessing a first set of measured data of a first variable associated with a wellbore operation of the wellbore operations in a downhole environment of the plurality of downhole environments; generating a first set of predicted data of the first variable; accessing a first set of measured data of a second variable associated with the wellbore operation in the downhole environment; applying a first model instance to the first set of measured data of the first variable to generate a first set of predicted data of the second variable; validating the first model instance by comparing the first set of predicted data of the second variable to the first set of measured data of the second variable; applying a second model instance to the first set of predicted data of the first variable to generate a second set of predicted data of the second variable; validating the second model instance by comparing the second set of predicted data of the second variable to the first set of measured data of the second variable; determining whether a deviation event associated with gathering the first set of measured data of the first variable has occurred by comparing the validation of the first model instance with the validation of the second model instance; and performing a well intervention operation on one or more electrical submersible pumps of the plurality of electrical submersible pumps in response to determining that the deviation event has occurred.

[0162] Aspect 44. The method of Aspect 43, wherein the well intervention operation comprises replacing the one or more electrical submersible pump, replacing one or more sensors associated with functioning of the one or more electrical submersible pumps, adjusting one or more operational parameters of the one or more electrical submersible pumps, recalibrating the one or more sensors associated with functioning of the one or more electrical submersible pump, or a combination thereof.

[0163] Aspect 45. A system comprising means for performing a method according to any of Aspects 1 through 17.

[0164] Aspect 46. A system comprising means for performing a method according to any of Aspects 21 through 23.

[0165] Aspect 47. A system comprising means for performing a method according to any of Aspects 24 through 35.

[0166] Aspect 48. A system comprising means for performing a method according to any of Aspects 36 through 39.

[0167] Aspect 49. A system comprising means for performing a method according to any of Aspects 40 through 42.

[0168] A system comprising means for performing a method according to either Aspects 43 or 44.

Claims

1. A computer-implemented method implemented through one or more processors executing instructions stored on a non-transitory computer-readable storage medium, the method comprising:accessing a first set of measured data of a first variable associated with an operation in a downhole environment;generating a first set of predicted data of the first variable;accessing a first set of measured data of a second variable associated with the operation in the downhole environment;applying a first model instance to the first set of measured data of the first variable to generate a first set of predicted data of the second variable;validating the first model instance by comparing the first set of predicted data of the second variable to the first set of measured data of the second variable;applying a second model instance to the first set of predicted data of the first variable to generate a second set of predicted data of the second variable;validating the second model instance by comparing the second set of predicted data of the second variable to the first set of measured data of the second variable; anddetermining whether a deviation event associated with gathering the first set of measured data of the first variable has occurred by comparing the validation of the first model instance with the validation of the second model instance.

2. The computer-implemented method of claim 1, wherein the first variable associated with the operation in the downhole environment is a surface variable and the second variable associated with the operation in the downhole environment is a subsurface variable.

3. The computer-implemented method of claim 1, wherein the first variable associated with the operation in the downhole environment is a subsurface variable and the second variable associated with the operation in the downhole environment is a surface variable.

4. The computer-implemented method of claim 1, wherein the deviation event corresponds to malfunctioning of a sensor or recalibration of a sensor.

5. The computer-implemented method of claim 1, wherein the deviation event is undetectable by a human in perceiving the first set of measured data of the first variable, the first set of predicted data of the first variable, the first set of measured data of the second variable, the first set of predicted data of the second variable for the first model instance, the second set of predicted data of the second variable for the second model instance, or a combination thereof.

6. The computer-implemented method of claim 1, further comprising performing a remedy in operation in the downhole environment in response to detection of the deviation event, wherein the remedy in operation in the downhole environment comprises recalibrating a sensor configured to gather the measured data of the first variable, repairing the sensor, remedying a down plugging event, or a combination thereof.

7. The computer-implemented method of claim 1, wherein the first set of measured data of the first variable and the first set of measured data of the second variable are accessed from a past timeframe, and the first model instance and the second model instance are applied and validated to detect an occurrence of the deviation event as a past event during the past timeframe.

8. The computer-implemented method of claim 1, further comprising:accessing a second set of measured data of the first variable;generating a second set of predicted data of the first variable;accessing a second set of measured data of the second variable;applying a third model instance to the second set of measured data of the first variable to generate a third set of predicted data of the second variable;validating the third model instance by comparing the third set of predicted data of the second variable to the second set of measured data of the second variable;applying a fourth model instance to the second set of predicted data of the first variable to generate a fourth set of predicted data of the second variable;validating the fourth model instance by comparing the fourth set of predicted data of the second variable to the second set of measured data of the second variable; anddetermining whether a second deviation event associated with gathering the second set of measured data of the first variable has occurred after the deviation event by comparing the validation of the third model instance with the validation of the fourth model instance.

9. The computer-implemented method of claim 8, wherein the second set of measured data of the first variable and the second set of measured data of the second variable are measured after a remedial action is performed in response to detection of the deviation event and the second set of predicted data of the first variable corresponds to after performance of the remedial action, the method further comprising:applying the third model instance to the second set of measured data of the first variable to generate the third set of predicted data of the second variable after performance of the remedial action;validating the third model instance by comparing the third set of the predicted data of the second variable to the second set of measured data of the second variable;applying the fourth model instance to the second set of predicted data of the first variable to generate the fourth set of predicted data of the second variable after performance of the remedial action;validating the fourth model instance by comparing the fourth set of predicted data of the second variable to the second set of measured data of the second variable; andidentifying whether the second deviation event associated with gathering the second measured data of the first variable has occurred after performance of the remedial action by comparing the validation of the third model instance with the validation of the fourth model instance.

10. The computer-implemented method of claim 8, wherein the second deviation event corresponds to recalibration of a sensor configured to gather the measured data of the first variable, the sensor malfunctioning in gathering the measured data of the first variable, the sensor being damaged during operation in the downhole environment, a down plugging event, or a combination thereof.

11. The computer-implemented method of claim 8, wherein the second deviation event is undetectable by a human in perceiving the second set of measured data of the first variable, the second set of predicted data of the first variable, the second set of measured data of the second variable, the third set of predicted data of the second variable, the fourth set of predicted data of the second variable for the fourth model instance, or a combination thereof.

12. The computer-implemented method of claim 8, further comprising determining whether the second deviation event has occurred by comparing the validation of the first model instance, the validation of the second model instance, the validation of the third model instance, and the validation of the fourth model instance.

13. The computer-implemented method of claim 12, further comprising:identifying that the validation of the first model instance, the validation of the second model instance, the validation of the third model instance, and the validation of the fourth model instance are substantially similar; andidentifying that the second deviation event has not occurred based on the substantial similarity between the validation of the first model instance, the validation of the second model instance, the validation of the third model instance, and the validation of the fourth model instance.

14. The computer-implemented method of claim 12, further comprising:identifying that the validation of the first model instance and the validation of the second model instance are substantially similar and that the validation of the third model instance and the validation of the fourth model instance are substantially similar;comparing the validation of the first model instance and the validation of the second model instance with the validation of the third model instance and the validation of the fourth model instance; andidentifying an occurrence of the second deviation event based on amounts that the third set of predicted data of the second variable and the fourth set of predicted data of the second variable deviate from the second set of measured data of the second variable in comparison to amounts that the first set of predicted data of the second variable and the second set of predicted data of the second variable deviate from the first set of measured data of the second variable.

15. The computer-implemented method of claim 14, further comprising identifying the occurrence of the second deviation event if the amounts that the third set of predicted data of the second variable and the fourth set of predicted data of the second variable deviate from the second set of measured data of the second variable are less than the amounts that the first set of predicted data of the second variable and the second set of predicted data of the second variable deviate from the first set of measured data of the second variable.

16. The computer-implemented method of claim 12, further comprising:identifying that the validation of the first model instance and the validation of the second model instance are substantially similar;identifying that an amount that the third set of predicted data of the second variable deviates from the second set of measured data of the second variable is less than amounts that the first set of predicted data of the second variable and the second set of predicted data of the second variable deviate from the first set of measured data of the second variable;identifying that an amount that fourth set of predicted data of the second variable deviates from the second set of measured data of the second variable is less than the amount that the third set of predicted data of the second variable deviates from the second set of measured data of the second variable; andidentifying a possible occurrence of the second deviation event based on that the amount that the fourth set of predicted data of the second variable deviates from the second set of measured data of the second variable deviate is less than the amount that the third set of predicted data of the second variable deviates from the second set of measured data of the second variable.

17. The computer-implemented method of claim 8, wherein the first set of measured data of the first variable, the first set of measured data of the second variable, the second set of measured data of the first variable, and the second set of measured data of the second variable are accessed from a past timeframe, and the first model instance, the second model instance, the third model instance, and the fourth model instance are applied and validated to detect an occurrence of the deviation event and the second deviation event as past events during the past timeframe.

18. A system comprising:one or more processors; andat least one computer-readable storage medium having stored therein instructions which, when executed by the one or more processors, cause the one or more processors to:access a first set of measured data of a first variable associated with an operation in a downhole environment;generate a first set of predicted data of the first variable;access a first set of measured data of a second variable associated with the operation in the downhole environment;apply a first model instance to the first set of measured data of the first variable to generate a first set of predicted data of the second variable;validate the first model instance by comparing the first set of predicted data of the second variable to the first set of measured data of the second variable;apply a second model instance to the first set of predicted data of the first variable to generate a second set of predicted data of the second variable;validate the second model instance by comparing the second set of predicted data of the second variable to the first set of measured data of the second variable; anddetermine whether a deviation event associated with gathering the first set of measured data of the first variable has occurred by comparing the validation of the first model instance with the validation of the second model instance.

19. The system of claim 18, wherein the instructions are further configured to cause the one or more processors to:access a second set of measured data of the first variable;generate a second set of predicted data of the first variable;access a second set of measured data of the second variable;apply a third model instance to the second set of measured data of the first variable to generate a third set of predicted data of the second variable;validate the third model instance by comparing the third set of predicted data of the second variable to the second set of measured data of the second variable;apply a fourth model instance to the second set of predicted data of the first variable to generate a fourth set of predicted data of the second variable;validate the fourth model instance by comparing the fourth set of predicted data of the second variable to the second set of measured data of the second variable; anddetermine whether a second deviation event associated with gathering the second set of measured data of the first variable has occurred after the deviation event by comparing the validation of the third model instance with the validation of the fourth model instance.

20. A non-transitory computer-readable storage medium storing instructions for causing one or more processors to:access a first set of measured data of a first variable associated with an operation in a downhole environment;generate a first set of predicted data of the first variable;access a first set of measured data of a second variable associated with the operation in the downhole environment;apply a first model instance to the first set of measured data of the first variable to generate a first set of predicted data of the second variable;validate the first model instance by comparing the first set of predicted data of the second variable to the first set of measured data of the second variable;apply a second model instance to the first set of predicted data of the first variable to generate a second set of predicted data of the second variable;validate the second model instance by comparing the second set of predicted data of the second variable to the first set of measured data of the second variable; anddetermine whether a deviation event associated with gathering the first set of measured data of the first variable has occurred by comparing the validation of the first model instance with the validation of the second model instance.

21. A method comprising:disposing an electrical submersible pump in a downhole environment;configuring the electrical submersible pump to operate in the downhole environment during a wellbore operation;accessing a first set of measured data of a first variable associated with the wellbore operation in the downhole environment;generating a first set of predicted data of the first variable;accessing a first set of measured data of a second variable associated with the wellbore operation in the downhole environment;applying a first model instance to the first set of measured data of the first variable to generate a first set of predicted data of the second variable;validating the first model instance by comparing the first set of predicted data of the second variable to the first set of measured data of the second variable;applying a second model instance to the first set of predicted data of the first variable to generate a second set of predicted data of the second variable;validating the second model instance by comparing the second set of predicted data of the second variable to the first set of measured data of the second variable; anddetermining whether a deviation event associated with gathering the first set of measured data of the first variable has occurred by comparing the validation of the first model instance with the validation of the second model instance.

22. The method of claim 21, wherein the deviation event is associated with operation of the electrical submersible pump during the wellbore operation.

23. The method of claim 22, wherein either or both the first variable and the second variable are associated with operation of the electrical submersible pump and comprise a pump intake pressure, a pump intake temperature, a motor temperature, a pump discharge pressure, vibration in a y-direction in association with operation of the electrical submersible pump, vibration in a x-direction in association with operation of the electrical submersible pump, vibration in a z-direction in association with operation of the electrical submersible pump or a combination thereof.

24. A method comprising:performing a wellbore operation in a downhole environment;accessing a first set of measured data of a first variable associated with the wellbore operation in the downhole environment;generating a first set of predicted data of the first variable;accessing a first set of measured data of a second variable associated with the wellbore operation in the downhole environment;applying a first model instance to the first set of measured data of the first variable to generate a first set of predicted data of the second variable;validating the first model instance by comparing the first set of predicted data of the second variable to the first set of measured data of the second variable;applying a second model instance to the first set of predicted data of the first variable to generate a second set of predicted data of the second variable;validating the second model instance by comparing the second set of predicted data of the second variable to the first set of measured data of the second variable; anddetermining whether a deviation event associated with gathering the first set of measured data of the first variable has occurred by comparing the validation of the first model instance with the validation of the second model instance.

25. The method of claim 24, wherein the deviation event is associated with performance of the wellbore operation.

26. The method of claim 24, wherein either or both the first variable and the second variable comprise a three-phase flow rate and the wellbore operation is a well production operation.

27. The method of claim 24, wherein either or both the first variable and the second variable comprise a solid, sand, or gas migration rate.

28. The method of claim 24, wherein either or both the first variable and the second variable comprise a frequency component of a control signal for controlling either or both a speed and a torque of a motor in the wellbore operation.

29. The method of claim 24, wherein either or both the first variable and the second variable comprise an output current of a variable speed drive used in the wellbore operation.

30. The method of claim 24, wherein either or both the first variable and the second variable comprise an output voltage of a variable speed drive used in the wellbore operation.

31. The method of claim 24, wherein either or both the first variable and the second variable comprise a tubing pressure of tubing used in the downhole environment during the wellbore operation.

32. The method of claim 24, wherein either or both the first variable and the second variable comprise a casing pressure between a wellbore and casing in the downhole environment or an annulus pressure between production tubing and the casing.

33. The method of claim 24, wherein either or both the first variable and the second variable comprise a flowline pressure within a flowline used during the wellbore operation.

34. The method of claim 24, wherein either or both the first set of measured data of the first variable and the first set of measured data of the second variable are received from the downhole environment through a wireline or fiber telemetry system.

35. The method of claim 24, further comprising recalibrating one or more downhole sensors in response to the deviation event through a wireline or fiber telemetry system.

36. A method comprising:accessing a first set of measured data of a first variable associated with an operation in a downhole environment;generating a first set of predicted data of the first variable;accessing a first set of measured data of a second variable associated with the operation in the downhole environment;applying a first model instance to the first set of measured data of the first variable to generate a first set of predicted data of the second variable;validating the first model instance by comparing the first set of predicted data of the second variable to the first set of measured data of the second variable;applying a second model instance to the first set of predicted data of the first variable to generate a second set of predicted data of the second variable;validating the second model instance by comparing the second set of predicted data of the second variable to the first set of measured data of the second variable;determining whether a deviation event associated with gathering the first set of measured data of the first variable has occurred by comparing the validation of the first model instance with the validation of the second model instance; andoutputting through a graphical user interface a display indicating whether the deviation event has occurred.

37. The method of claim 36, further comprising outputting the display comprising the first set of measured data of the first variable, the first set of predicted data of the first variable, the first set of measured data of the second variable, the first set of predicted data of the second variable, the second set of predicted data of the second variable, or a combination thereof.

38. The method of claim 36, further comprising indicating on the display an occurrence of the deviation event relative to the first set of measured data of the first variable, the first set of predicted data of the first variable, the first set of measured data of the second variable, the first set of predicted data of the second variable, the second set of predicted data of the second variable, or a combination thereof.

39. The method of claim 36, further comprising indicating on the display a plurality of potential deviation events and an indication of whether the deviation event is a true deviation event or a potential deviation event of the plurality of deviation events.

40. A method comprising:accessing a first set of measured data of a first variable associated with performing an operation in a downhole environment;generating a first set of predicted data of the first variable;accessing a first set of measured data of a second variable associated with performing the operation in the downhole environment;applying a first model instance to the first set of measured data of the first variable to generate a first set of predicted data of the second variable;validating the first model instance by comparing the first set of predicted data of the second variable to the first set of measured data of the second variable;applying a second model instance to the first set of predicted data of the first variable to generate a second set of predicted data of the second variable;validating the second model instance by comparing the second set of predicted data of the second variable to the first set of measured data of the second variable;determining whether a deviation event associated with gathering the first set of measured data of the first variable has occurred by comparing the validation of the first model instance with the validation of the second model instance;performing a well intervention operation in response to determining that the deviation event has occurred.

41. The method of claim 40, wherein the operation comprises functioning of an electrical submersible pump disposed within a wellbore of the downhole environment.

42. The method of claim 41, wherein the well intervention operation comprises replacing the electrical submersible pump, replacing a sensor associated with functioning of the electrical submersible pump, adjusting an operational parameter of the electrical submersible pump, recalibrating the sensor associated with functioning of the electrical submersible pump, or a combination thereof.

43. A method comprising:monitoring wellbore operations of a plurality of electrical submersible pumps disposed in a plurality of downhole environment;accessing a first set of measured data of a first variable associated with a wellbore operation of the wellbore operations in a downhole environment of the plurality of downhole environments;generating a first set of predicted data of the first variable;accessing a first set of measured data of a second variable associated with the wellbore operation in the downhole environment;applying a first model instance to the first set of measured data of the first variable to generate a first set of predicted data of the second variable;validating the first model instance by comparing the first set of predicted data of the second variable to the first set of measured data of the second variable;applying a second model instance to the first set of predicted data of the first variable to generate a second set of predicted data of the second variable;validating the second model instance by comparing the second set of predicted data of the second variable to the first set of measured data of the second variable;determining whether a deviation event associated with gathering the first set of measured data of the first variable has occurred by comparing the validation of the first model instance with the validation of the second model instance; andperforming a well intervention operation on one or more electrical submersible pumps of the plurality of electrical submersible pumps in response to determining that the deviation event has occurred.

44. The method of claim 43, wherein the well intervention operation comprises replacing the one or more electrical submersible pump, replacing one or more sensors associated with functioning of the one or more electrical submersible pumps, adjusting one or more operational parameters of the one or more electrical submersible pumps, recalibrating the one or more sensors associated with functioning of the one or more electrical submersible pump, or a combination thereof.