A correction method and system for intelligent drilling computing model

Abstract

The application discloses a correction method and system for an intelligent drilling calculation model, and comprises the following steps: obtaining drilling static data and real drilling measurement data of a current drilling well, and based on the data, calculating drilling state parameters by using an original drilling calculation model used in a current drilling process; monitoring real-time deviation between downhole states calculated by the model and actual measurement states according to the drilling state parameters and the real drilling measurement data, and analyzing reasons for current deviation according to the real-time deviation monitoring result; and when the current deviation is caused by the model calculation, correcting the original drilling calculation model by evaluating a current deviation amount. The application can continuously improve calculation accuracy of the drilling calculation model in the drilling process, and meanwhile, realizes automatic correction of the model.

Description

Technical Field

[0001] This invention relates to the field of petroleum engineering technology, and in particular to a correction method and system for intelligent drilling calculation models. Background Technology

[0002] As exploration and development continue to deepen, the difficulty of oil and gas exploration and development is increasing, the geological conditions are becoming more complex, the reservoir depth is increasing, and the drilling engineering faces more and more complex situations. A large number of heterogeneous, uncertain, unstructured, and non-numerical problems constitute the "black box" of drilling engineering, which leads to increasingly higher costs for dealing with drilling risks and accidents. Achieving safe, efficient, and low-cost drilling has become the primary goal of the drilling industry.

[0003] However, the performance of current drilling industry sensor components is constrained by factors such as high-frequency vibration, high temperature and pressure, and drilling fluid flow in the well. It is impossible to rely entirely on sensors to obtain comprehensive downhole parameters to identify downhole conditions and predict downhole risks. On the other hand, using more precision sensors in drilling operations would significantly increase drilling costs, which contradicts the goal of accelerating drilling, improving efficiency, and controlling costs. Therefore, large-scale implementation is not possible, and only a small number of sensing and measuring instruments are added in key exploration wells to monitor downhole parameters.

[0004] Meanwhile, the development of intelligent drilling in the current drilling industry not only requires automated drilling rigs, but also relies heavily on various drilling calculation models to calculate, analyze, diagnose, predict, optimize, and make decisions regarding drilling conditions, thereby achieving intelligent and precise control of automated drilling rigs. While theoretically, calculation models based on key parameters such as formation pressure, wellbore pressure, and friction torque can analyze downhole conditions, trends, and risks, these models are based on numerous assumptions from their inception. The calculation results deviate from the actual downhole environment, and as drilling operations extend the wellbore, changes in temperature, pressure, wellbore trajectory, and the rheology of the wellbore fluid and the drill string condition continuously increase the calculation deviation, ultimately affecting the accuracy of wellbore condition identification and risk warning results. Advanced and complex drilling calculation models rely on a large number of input parameters, such as tubing and wellbore geometry, various tuning parameters, and fluid properties. Therefore, these calculation models are difficult to configure on-site and require significant time. Furthermore, each well may require experts to correctly initialize the calculation model and adjust it as needed during drilling.

[0005] Existing model calibration methods mainly involve adding a deviation value to the calculation results to adjust them to match the measured values. This is a forced fitting method, which will quickly lead to deviations in subsequent drilling calculations. At the same time, it requires personnel to monitor the model usage so that adjustments can be made at any time. In some cases, manual adjustments to the model are even required, which is time-consuming, labor-intensive, and ineffective.

[0006] Therefore, existing technologies need to develop a scheme that can automatically correct the calculation model of the drilling process in real time, so as to maintain the calculation accuracy of the model by realizing automatic correction of the calculation model in the automated and intelligent drilling process. Summary of the Invention

[0007] To address the aforementioned technical problems, this invention provides a correction method for an intelligent drilling calculation model, comprising: acquiring current drilling static data and actual drilling measurement data; based on this, calculating drilling state parameters using the original drilling calculation model used in the current drilling process; monitoring the real-time deviation between the drilling state calculated by the model and the actual measurement state according to the drilling state parameters and the actual drilling measurement data, and analyzing the cause of the current deviation based on the real-time deviation monitoring results; and correcting the original drilling calculation model by evaluating the current deviation amount when the current deviation is caused by the model calculation.

[0008] Preferably, the step of analyzing the cause of the current deviation based on the real-time deviation monitoring results includes: when the current deviation is greater than a preset first deviation threshold, continuing to analyze whether the current deviation is caused by the fluctuation of the current drilling operation control parameters; wherein, if the current deviation is not caused by the fluctuation of the current drilling operation control parameters, the original drilling calculation model needs to be corrected and a model correction instruction is generated immediately.

[0009] Preferably, the current drilling control parameters are diagnosed to determine whether fluctuations occur within a specified time period, in order to determine the cause of the current deviation. If fluctuations occur, the drilling state parameters are recalculated; if no fluctuations occur, it is determined that the current deviation is caused by model calculation.

[0010] Preferably, if the current deviation is less than the first deviation threshold, the drilling status parameters are calculated again.

[0011] Preferably, the step of correcting the original drilling calculation model using the current deviation includes: identifying the current drilling rig type; based on the drilling rig type identification result, according to the drilling operation control parameters under stable conditions or by adding drilling operation control parameters with preset characteristics and fluctuations, using multiple preset parameter adjustment models, calculating the deviation between the drilling state parameters corresponding to each parameter adjustment model and the actual drilling measurement data, so as to correct the original drilling calculation model, wherein the multiple parameter adjustment models are formed by adjusting one or more model parameters of the original drilling calculation model using different strategies.

[0012] Preferably, when the current drilling rig is a non-automatic drilling rig, drilling operation control parameters under stable conditions are obtained; based on the drilling operation control parameters under stable conditions, drilling state parameters under different simulation conditions are calculated using multiple preset parameter adjustment models; further, combined with actual drilling measurement data under stable conditions, the calculation deviation of the parameter adjustment model corresponding to different simulation conditions is obtained; based on the calculation deviation of the parameter adjustment model, the optimal parameter adjustment model is selected, and the optimal parameter adjustment model is used to replace and update the original drilling calculation model; or based on the calculation deviation of the parameter adjustment model, multiple first backup parameter adjustment models are selected, and the model set formed by the multiple first backup parameter adjustment models is used to replace and update the original drilling calculation model.

[0013] Preferably, when the current drilling rig is an automated drilling rig, a small periodic fluctuation is added to the current drilling operation control parameters through the drilling rig control system; based on the drilling operation control parameters with the added small periodic fluctuation, the drilling state parameters under different simulation conditions are calculated using multiple preset parameter adjustment models; further, combined with the actual drilling measurement data obtained after adding the small periodic fluctuation to the input of the original drilling calculation model, the calculation deviation of the parameter adjustment model corresponding to different simulation conditions is obtained; based on the calculation deviation of the parameter adjustment model, multiple second backup parameter adjustment models are selected, and the original drilling calculation model is replaced and updated using the model set formed by the multiple second backup parameter adjustment models.

[0014] Preferably, the real-time deviation is monitored or the deviation is calculated by comparing the various first-type wellhead parameters in the drilling status parameters with the various second-type wellhead parameters in the real-time logging data.

[0015] Preferably, the process of monitoring real-time deviation or calculating deviation includes: forming multiple first-type parameter curves based on the first-type wellhead parameters, and forming multiple second-type parameter curves based on the second-type wellhead parameters; and completing the monitoring of real-time deviation or the calculation of deviation by calculating the difference in slope, the degree of change in curve spatial distance, or the degree of change in curve similarity between the multiple first-type parameter curves and the multiple second-type parameter curves.

[0016] Preferably, the drilling static data includes at least wellbore structure, wellbore trajectory, drill string assembly and formation stratification information; the actual drilling measurement data includes at least real-time logging data and logging-while-drilling data; and the drilling state parameters include at least the wellbore pressure distribution characteristics, cuttings distribution characteristics, drill string friction distribution characteristics, drill string torque distribution characteristics, wellhead stand-up pressure, wellhead casing pressure and wellhead torque.

[0017] Preferably, the wellhead parameters include at least stand pressure, casing pressure, and torque.

[0018] Preferably, multiple parameter tuning models are run in parallel using multiple cores and multiple threads, and the drilling state parameters corresponding to each parameter tuning model are calculated respectively.

[0019] On the other hand, the present invention also provides a calibration system for an intelligent drilling calculation model. The calibration system is executed according to the calibration method described above. The calibration system includes: a target well database for storing static drilling data and actual drilling measurement data of the currently drilling well; a real-time calculation module for acquiring the static drilling data and actual drilling measurement data of the currently drilling well, and based on this, calculating drilling state parameters using the original drilling calculation model used in the current drilling process; a model deviation monitoring module for real-time deviation monitoring of the drilling state calculated by the model from the actual measured state based on the drilling state parameters and the actual drilling measurement data, and analyzing the cause of the current deviation based on the real-time deviation monitoring results; and an automatic calibration module for correcting the original drilling calculation model by evaluating the current deviation amount when the current deviation is caused by model calculation.

[0020] Compared with the prior art, one or more embodiments of the above solutions may have the following advantages or beneficial effects:

[0021] This invention discloses a calibration method and system for intelligent drilling calculation models. Applied to the petroleum engineering field, this method and system monitors calculation deviations in real time, utilizes multi-core, multi-threaded parallel multi-group parameter-tuning models, and selects the optimal model or set of models with the smallest calculation error for application. This ensures consistently accurate calculation results during drilling, solving the problem of existing models requiring professional parameter tuning and having poor sustainability of calculation accuracy. This promotes the application of automated and intelligent drilling systems, ultimately helping drilling engineers achieve efficient and high-quality drilling. Therefore, this invention not only continuously improves the calculation accuracy of the calculation model during drilling but also achieves automatic model calibration, eliminating the need for time-consuming and labor-intensive configuration, continuous tracking, monitoring, and adjustment by experts. Furthermore, this invention not only helps improve calculation accuracy, assisting drilling personnel in better monitoring downhole conditions and predicting drilling risks, but also eliminates the obstacle of requiring continuous maintenance by professional experts for the deployment and application of automated drilling systems, making automated / intelligent systems a more cost-effective choice for drilling operations.

[0022] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the description, claims and drawings. Attached Figure Description

[0023] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with the embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0024] Figure 1 This is a step diagram of a correction method for an intelligent drilling calculation model according to an embodiment of this application.

[0025] Figure 2 This is a flowchart illustrating a correction method for an intelligent drilling calculation model according to an embodiment of this application.

[0026] Figure 3 This is a schematic diagram illustrating the model correction principle when the drilling operation control parameters are subjected to minor disturbances in the correction method for the intelligent drilling calculation model according to an embodiment of this application.

[0027] Figure 4 This is a schematic diagram of the structure of a correction system for an intelligent drilling calculation model according to an embodiment of this application. Detailed Implementation

[0028] The embodiments of the present invention will be described in detail below with reference to the accompanying drawings and examples, so that the process of how the present invention uses technical means to solve technical problems and achieve technical effects can be fully understood and implemented accordingly. It should be noted that, as long as there is no conflict, the various embodiments and features in the various embodiments of the present invention can be combined with each other, and the resulting technical solutions are all within the protection scope of the present invention.

[0029] Furthermore, the steps illustrated in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Also, although a logical order is shown in the flowcharts, in some cases the steps shown or described may be performed in a different order than that shown here.

[0030] As exploration and development continue to deepen, the difficulty of oil and gas exploration and development is increasing, the geological conditions are becoming more complex, the reservoir depth is increasing, and the drilling engineering faces more and more complex situations. A large number of heterogeneous, uncertain, unstructured, and non-numerical problems constitute the "black box" of drilling engineering, which leads to increasingly higher costs for dealing with drilling risks and accidents. Achieving safe, efficient, and low-cost drilling has become the primary goal of the drilling industry.

[0031] However, the performance of current drilling industry sensor components is constrained by factors such as high-frequency vibration, high temperature and pressure, and drilling fluid flow in the well. It is impossible to rely entirely on sensors to obtain comprehensive downhole parameters to identify downhole conditions and predict downhole risks. On the other hand, using more precision sensors in drilling operations would significantly increase drilling costs, which contradicts the goal of accelerating drilling, improving efficiency, and controlling costs. Therefore, large-scale implementation is not possible, and only a small number of sensing and measuring instruments are added in key exploration wells to monitor downhole parameters.

[0032] Meanwhile, the development of intelligent drilling in the current drilling industry not only requires automated drilling rigs, but also relies heavily on various drilling calculation models to calculate, analyze, diagnose, predict, optimize, and make decisions regarding drilling conditions, thereby achieving intelligent and precise control of automated drilling rigs. While theoretically, calculation models based on key parameters such as formation pressure, wellbore pressure, and friction torque can analyze downhole conditions, trends, and risks, these models are based on numerous assumptions from their inception. The calculation results deviate from the actual downhole environment, and as drilling operations extend the wellbore, changes in temperature, pressure, wellbore trajectory, and the rheology of the wellbore fluid and the drill string condition continuously increase the calculation deviation, ultimately affecting the accuracy of wellbore condition identification and risk warning results. Advanced and complex drilling calculation models rely on a large number of input parameters, such as tubing and wellbore geometry, various tuning parameters, and fluid properties. Therefore, these calculation models are difficult to configure on-site and require significant time. Furthermore, each well may require experts to correctly initialize the calculation model and adjust it as needed during drilling.

[0033] Existing model calibration methods mainly involve adding a deviation value to the calculation results to adjust them to match the measured values. This is a forced fitting method, which will quickly lead to deviations in subsequent drilling calculations. At the same time, it requires personnel to monitor the model usage so that adjustments can be made at any time. In some cases, manual adjustments to the model are even required, which is time-consuming, labor-intensive, and ineffective.

[0034] Therefore, to address one or more of the aforementioned technical problems, this application proposes a correction method and system for intelligent drilling calculation models. The method and system first monitor the real-time deviation between actual drilling measurement data and the drilling state parameters calculated by the intelligent drilling calculation model, continuously analyzing the causes of the current deviation. Then, when the current deviation is caused by model calculation, the original intelligent drilling calculation model is corrected in real-time based on the evaluation results of the current deviation. In this way, the present invention can automatically correct the drilling process calculation model in real-time, maintaining the model's calculation accuracy during automated and intelligent drilling processes by achieving automatic correction of the calculation model.

[0035] Figure 1This is a step diagram of a correction method for an intelligent drilling calculation model according to an embodiment of this application. Figure 2 This is a schematic flowchart illustrating the correction method for an intelligent drilling calculation model according to an embodiment of this application. The following is in conjunction with... Figure 1 and Figure 2 The present invention provides a detailed description of the calibration method for intelligent drilling calculation models (hereinafter referred to as the "automatic model calibration method") described in the embodiments of the present invention.

[0036] refer to Figure 1 Step S110 obtains the current drilling static data and actual drilling measurement data. Based on this, the drilling state parameters are calculated using the original drilling calculation model used in the current drilling process.

[0037] In this example, a database about the target well (i.e., the current well) will be built before implementing the automatic model correction method. The target well database stores: drilling (related) static data and various real-time measurement data obtained from the drilling site (i.e., actual drilling measurement data). The drilling static data includes at least: wellbore structure, wellbore trajectory, drill string assembly, and formation stratification information. The actual drilling measurement data includes at least: real-time logging data and logging-while-drilling data.

[0038] In step S110, refer to Figure 2 First, an automatic model correction system for implementing intelligent drilling calculations needs to be deployed at the drilling site. Then, drilling-related static data about the target wells needs to be imported into the target well database, and various real-time measurement data from the well site needs to be connected to the target well database in the automatic model correction system.

[0039] Next, in step S110, after the drilling operation begins, the automatic calibration system of the calculation model will be run based on the data stored in the target well database. Based on drilling-related static data and actual drilling measurement data, the system first determines multiple drilling operation control parameters required to control the drilling rig during the current drilling operation. Then, based on these multiple drilling operation control parameters, the system uses models such as wellbore hydraulics and friction torque within the original (intelligent) drilling calculation model to calculate drilling state parameters in real time. In this embodiment of the invention, the drilling state parameters include at least: wellbore pressure distribution characteristics, cuttings distribution characteristics, drill string friction distribution characteristics, drill string torque distribution characteristics, wellhead stand-up pressure, wellhead casing pressure, and wellhead torque.

[0040] Continue to refer to Figure 1 Step S120 uses the drilling status parameters calculated in real time in step S110 and the actual drilling measurement data collected in real time to monitor the real-time deviation between the drilling status calculated by the model and the actual measurement status, and analyzes the cause of the current deviation based on the real-time deviation monitoring results.

[0041] In step S120, firstly, the drilling status parameters and the actual drilling measurement data collected in real time at the drilling site need to be compared and analyzed to monitor the actual deviation between the drilling status parameters and the actual drilling measurement data in real time. Specifically, since the drilling status parameters calculated based on the model include multiple wellhead parameters, and the actual drilling measurement data collected in real time from the drilling site also includes multiple wellhead parameters, this embodiment of the invention will, during the real-time deviation monitoring process, compare each type of first-class wellhead parameter in the drilling status parameters with each type of second-class wellhead parameter in the real-time logging data to monitor the real-time deviation or to calculate the deviation as described below. The first-class wellhead parameters include, but are not limited to, wellhead stand-up pressure, wellhead casing pressure, and wellhead torque calculated based on the model, while the second-class wellhead parameters include, but are not limited to, wellhead stand-up pressure, wellhead casing pressure, and wellhead torque measured in real time.

[0042] Furthermore, in the process of real-time deviation monitoring (or calculating the deviation corresponding to different parameter adjustment models as described below), firstly, multiple first-type parameter curves are formed based on various first-type wellhead parameters, and multiple second-type parameter curves are formed based on various second-type wellhead parameters; then, by calculating the difference in slope of the multiple first-type parameter curves and the multiple second-type parameter curves, or the degree of change in the spatial distance between the curves, or the degree of change in the similarity of the curves, the task of monitoring the real-time deviation or the task of calculating the deviation as described below is completed.

[0043] In other words, after obtaining multiple first-type parameter curves and multiple second-type parameter curves, this embodiment of the invention determines the deviation between the first-type parameter curve (e.g., the wellhead standpressure curve formed based on wellhead standpressure data calculated from the model) and the second-type parameter curve (e.g., the wellhead standpressure curve formed based on wellhead standpressure data measured at the actual well site) corresponding to the same type of parameter. In this example, deviation monitoring can be calculated by taking the difference between the data points of the "slope of the model-calculated parameter curve" and the "slope of the wellhead measured parameter curve," or by calculating the change in spatial distance between the two curves, or by calculating the change in similarity between the two curves. Furthermore, the calculation of spatial distance change can be selected from one of the algorithms of Euclidean distance, cosine distance, and Manhattan distance; the calculation of curve similarity change can be selected from one of the methods based on points (e.g., EDR, LCSS, DTW), shapes (e.g., Frechet, Hausdorff), and segments (e.g., One Way Distance, LIP distance).

[0044] Next, continue to refer to Figure 2Furthermore, based on the real-time deviation monitoring results obtained above, the causes of the current deviation are analyzed. In practical applications, the real-time deviations monitored by this invention may be caused by fluctuations in drilling parameters, or by deviations between the model calculation results and actual measurement data generated during the implementation of intelligent drilling calculations by the original drilling calculation model. This type of deviation is the type of deviation that needs to be corrected in the subsequent step S130.

[0045] Specifically, in the deviation cause analysis, the first step is to determine, based on the current deviation monitoring (calculation) results, whether the current deviation has reached a level requiring model correction (whether it exceeds the critical deviation threshold allowed for safe drilling operations). If it has reached a level requiring model correction, further analysis is conducted to determine whether the current deviation is caused by fluctuations in the current drilling operation control parameters. Conversely, if the current deviation has not reached a level requiring model correction, there is no need to continue diagnosing the cause of the deviation, and the original drilling calculation model can continue to be used to calculate the drilling state parameters.

[0046] Furthermore, when diagnosing whether the current deviation has reached the level requiring model correction, if the deviation data in the current deviation monitoring results is greater than a preset first deviation threshold, it is determined that the current deviation has reached the level requiring model correction, and then further analysis is conducted to determine whether the current deviation is caused by fluctuations in the current drilling operation control parameters. Conversely, if the deviation data in the current deviation monitoring results is not less than the aforementioned first deviation threshold, it is determined that the current deviation has not reached the level requiring model correction. In this case, it is necessary to return to step S110 to continue calculating the drilling state parameters based on the actual drilling measurement data and using the original drilling calculation model.

[0047] Furthermore, when diagnosing whether the current deviation is caused by fluctuations in the current drilling operation control parameters, it is necessary to diagnose whether fluctuations occur in the current drilling control parameters within a specified time period to determine the cause of the current deviation. If fluctuations occur in the current drilling control parameters within the specified time period, it is determined that the current deviation is caused by fluctuations in the drilling operation control parameters. In this case, it is necessary to return to step S110 to continue calculating the drilling state parameters based on actual drilling measurement data and the original drilling calculation model. The deviation exceeding the limit judgment will only be performed when the drilling operation control parameters stabilize. Conversely, if the current drilling control parameters do not fluctuate within the specified time period (state stable), it is determined that the current deviation is not caused by fluctuations in the drilling operation control parameters, but by the calculations of the original drilling calculation model itself.

[0048] Furthermore, if the current deviation is not caused by fluctuations in the drilling operation control parameters, but by the original drilling calculation model itself, then the original drilling calculation model needs to be corrected. When the drilling operation control parameters are stable (without fluctuations), a model correction command should be generated immediately to initiate the following step S130.

[0049] like Figure 1 As shown, in step S130, when the current deviation is caused by the model calculation, the original drilling calculation model is corrected by evaluating the current deviation amount.

[0050] In step S130, after the drilling operation control parameters stabilize (without fluctuations), the original model calibration process begins under the instruction of the aforementioned model calibration command. Specifically, in the original model calibration process, firstly, the current drilling rig type is identified. Then, based on the identification result of the current drilling rig type, according to the drilling operation control parameters under stable conditions or drilling operation control parameters with pre-defined fluctuations, multiple preset parameter adjustment models are used to calculate the deviation (calculated deviation) between the drilling state parameters corresponding to each parameter adjustment model and the actual drilling measurement data, so as to calibrate the original drilling calculation model.

[0051] In this embodiment of the invention, multiple parameter tuning models are pre-constructed, each model being formed based on the original drilling calculation model after executing different parameter tuning strategies. Furthermore, these multiple parameter tuning models are formed by adjusting one or more model parameters of the original drilling calculation model by different adjustment ranges.

[0052] In addition, when running multiple parameter tuning models and calculating the deviation corresponding to different parameter tuning models, multiple parameter tuning models are run in parallel using multiple cores and multiple threads, and the drilling state parameters corresponding to the respective parameter tuning models are calculated separately.

[0053] In practical applications, the original drilling calculation model is constructed based on model parameters that characterize different aspects of the drilling environment (e.g., drilling fluid rheological parameters, downhole temperature distribution characteristics, downhole pressure distribution characteristics, etc.). Therefore, different parameter tuning strategies are employed when tuning the original drilling calculation model, resulting in corresponding tuning models for each strategy. In this embodiment of the invention, the parameter tuning strategy includes: first, randomly selecting one or more model parameters that need adjustment from all model parameters required to construct the original drilling calculation model; then, adjusting the magnitudes of these parameters to varying degrees to form corresponding tuning models. Thus, this invention enables the pre-construction of multiple tuning models.

[0054] Specifically, refer to Figure 2When identifying the current drilling rig type, it is necessary to determine whether the drilling rig used for the current drilling operation is an automated drilling rig. Automated drilling rigs have an integrated control system that can control and adjust drilling equipment such as mud pumps, winches, and top drives through data and instructions (drilling rig operation control parameters) output by the centralized control system. This changes the actual drilling parameters applied to the drilling operation, allowing each drilling rig to complete the drilling operation using the current actual drilling parameters.

[0055] In the first embodiment, when the current drilling rig is a non-automatic drilling rig, and the drilling operation control parameters are stable, (step 1) the drilling operation control parameters under stable conditions are obtained; then, (step 2) based on the drilling operation control parameters under stable conditions, multiple preset parameter adjustment models are used to calculate the drilling state parameters under different simulation conditions, and further combined with the actual drilling measurement data under stable conditions, the calculation deviation of the parameter adjustment model corresponding to different simulation conditions is obtained; finally, (step 3) based on the calculation deviation obtained for different parameter adjustment models, a new parameter adjustment model is formed to replace and update the original drilling calculation model.

[0056] In the third step, a specific embodiment of the present invention can select the optimal parameter tuning model based on the calculation deviation obtained for different parameter tuning models, and use the optimal parameter tuning model to replace and update the original drilling calculation model.

[0057] Furthermore, when the drilling rig used in the current drilling operation is not an automated drilling rig, real-time data of the drilling operation control parameters within a stable time period is extracted. The real-time drilling operation control parameter data under stable conditions is sent to multiple sets of parameter tuning models based on computers or servers. At this time, multi-core multi-threaded parallel operation of each set of parameter tuning models is adopted, so that each set of parameter tuning models calculates the corresponding drilling state parameters. Combined with the actual drilling measurement data collected in real time at the drilling site, the wellhead parameters in each set of drilling state parameters and the wellhead parameters in the real-time measurement data are further monitored for deviation. A corresponding set of (model) calculation deviations is calculated for each set of parameter tuning models to evaluate the calculation deviations of each set of parameter tuning models. Finally, the parameter tuning model with the smallest calculation deviation below the preset calculation deviation threshold is selected as the optimal parameter tuning model, so as to update and apply the original drilling calculation model using the selected optimal parameter tuning model. Therefore, by verifying and determining the optimal parameter adjustment model that best matches the drilling operation control parameters under stable conditions, the current original drilling calculation model is automatically corrected. Then, the process returns to step S110, using the current optimal parameter adjustment model as the original drilling calculation model for subsequent automatic correction processes. This achieves the goal of continuously and automatically correcting the intelligent drilling calculation model during the drilling process.

[0058] Since the drilling rig is currently an automated drilling rig, there may be multiple models whose deviations are less than the calculated deviation threshold after parameter tuning. Either the model with the smallest deviation can be selected, or multiple models with the required deviations can be constructed into a model set and then weighted for calculation. Constructing a model set can avoid the faster deviation of a single calibrated model. The deviations of multiple models may be high or low, and the weighted calculation of the model set can reduce the problem of the rapid increase of the deviation of a single model, and can make the accuracy of the calibrated calculation results last longer.

[0059] In the third step described above, another specific embodiment of the present invention can further select multiple first backup parameter tuning models based on the calculation deviations obtained for different parameter tuning models, and use the model set formed by the multiple first backup parameter tuning models to replace and update the original drilling calculation model. The process of using the first backup model and generating the model set using the first backup model in this embodiment of the present invention is similar to the process of using the second backup model and generating the model set using the second backup model described below, and the application process of the model set generated based on the first backup model is similar to the application process of the model set generated based on the second backup model described below, therefore, it will not be described in detail here.

[0060] In the second embodiment, when the current drilling rig is an automated drilling rig, (Step 1) a small periodic fluctuation is added to all current drilling operation control parameters through the drilling rig control system; then, (Step 2) based on the drilling operation control parameters with added small periodic fluctuations, the drilling state parameters under different simulation conditions are calculated using multiple preset parameter adjustment models. Furthermore, combined with the actual drilling measurement data obtained after adding small periodic fluctuations to the input of the original drilling calculation model, the calculation deviation of the parameter adjustment model corresponding to different simulation conditions is obtained; (Step 3) based on the calculation deviations obtained for different parameter adjustment models, multiple backup parameter adjustment models are selected, and the original drilling calculation model is replaced and updated using the model set formed by the multiple backup parameter adjustment models.

[0061] Figure 3 This is a schematic diagram illustrating the model correction principle when minor disturbances are added to the drilling operation control parameters in the correction method for the intelligent drilling calculation model according to an embodiment of this application. When the drilling rig used in the current drilling operation is an automated drilling rig, such as... Figure 3 As shown, a small periodic fluctuation is first added to the current drilling operation control parameters (x1, x2, ... represent different drilling operation control parameters) through the drilling rig control system. (For example, in the first set of input disturbances, a corresponding small periodic fluctuation is added to each drilling operation control parameter: x1 + Δ...) 11 x2+Δ 21 ..., where Δ 11 Δ 21……These represent the minute periodic fluctuations corresponding to various drilling operation control parameters. The parameter before the subscript Δ indicates the parameter number, and the parameter after the subscript Δ indicates the group number of the disturbance input. Then, multiple drilling operation control parameters with added minute fluctuations are extracted. The data for each drilling operation control parameter with added minute periodic fluctuations (in the first group of input disturbances, parameters such as the friction coefficient C in the parameter tuning model can be adjusted) are then processed. F Drag coefficient parameter C D Incorporating fluctuations, where C F and C D The subscripts (indicating the group number of the disturbance input) are fed into multiple (N) parameter tuning models based on a computer or server. At this point, a multi-core, multi-threaded parallel execution of each parameter tuning model ensures that each model calculates the corresponding drilling state parameters. Combined with real-time drilling measurement data collected from the drilling site, the wellhead parameters in each group of drilling state parameters are further monitored for deviation from the wellhead parameters in the real-time measurement data. A corresponding (model) calculation deviation is calculated for each parameter tuning model to evaluate its calculation deviation. Finally, parameter tuning models with calculation deviations below a preset calculation deviation threshold are selected as candidate parameter tuning models, forming a set of calibration models. This set of calibration models, composed of multiple candidate parameter tuning models, is then used to update and apply the original drilling calculation model. Therefore, by selecting a parameter adjustment model less affected by the fluctuations of the input drilling operation control parameters, the current original drilling calculation model is automatically corrected. Then, the process returns to step S110, using the current corrected model set as the original drilling calculation model for subsequent automatic correction processes. This achieves the goal of continuously and automatically correcting the intelligent drilling calculation model during the drilling process.

[0062] It should be noted that in practical applications, the calibration model set is applied through parallel computing. When each candidate parameter tuning model in the set calculates multiple drilling state parameters, the calculation results of the same type of drilling state parameters are weighted and averaged to obtain the corrected calculation result of that parameter. Thus, after obtaining the corrected calculation results of drilling state parameters for all categories, the intelligent drilling calculation task of using the calibration model set to calculate drilling state parameters is realized.

[0063] On the other hand, based on the above-mentioned automatic model correction method, the present invention also provides a correction system for intelligent drilling calculation models (hereinafter referred to as "automatic model correction system"), which executes according to the above-described automatic model correction method. Figure 4 This is a schematic diagram of the structure of a correction system for an intelligent drilling calculation model according to an embodiment of this application. Figure 4As shown, the automatic model correction system of the present invention includes: a target well database 401, a real-time calculation module 402, a model deviation monitoring module 403, and an automatic correction module 404.

[0064] Specifically, the target well database 401 is used to store the current drilling static data and actual drilling measurement data of the well. The target well database 401 mainly stores the data required by the calculation model, such as real-time drilling data and static data, and provides it to the real-time calculation module 402.

[0065] The real-time calculation module 402 is implemented according to the method described in step S110 above. It is used to acquire the current drilling static data and actual drilling measurement data of the well. Based on this, it calculates the drilling state parameters using the original drilling calculation model used in the current drilling process. The real-time calculation module 402 is responsible for running the calculation models of wellbore hydraulics, friction torque, etc., and calculating various drilling state parameters such as pressure distribution in the wellbore, cuttings distribution, drill string friction distribution, drill string torque distribution, wellhead stand-up pressure / casing pressure, and wellhead torque. The calculation results can be displayed to engineers for reference or output to the corresponding monitoring / prediction system for application, so as to further assess / predict the risk status of drilling downhole operations.

[0066] The model deviation monitoring module 403 is implemented according to the method described in step S120 above. It is used to monitor the real-time deviation between the drilling state calculated by the model and the actual measured state based on the drilling state parameters output by the real-time calculation module 402 and the actual drilling measurement data obtained from well site measurements. Based on the real-time deviation monitoring results, it analyzes the cause of the current deviation. The model deviation monitoring module 403 is responsible for real-time deviation monitoring between the wellhead parameters (including but not limited to standpipe pressure, casing pressure, torque, etc.) calculated by the model and the wellhead parameters (including but not limited to standpipe pressure, casing pressure, torque, etc.) measured by logging, and determining whether automatic model correction is needed.

[0067] The automatic correction module 404 is implemented according to the method described in step S130 above. When the current deviation is caused by model calculation, it corrects the original drilling calculation model by evaluating the current deviation amount. The automatic correction module 404 runs multiple sets of parameter-tuned models in parallel using a multi-core, multi-threaded computer or server. It evaluates the parameter-tuning correction effect using operational data from stable time periods or operational data with added minor perturbations, ultimately completing the automatic correction and application of the model and improving the model's calculation accuracy.

[0068] This invention discloses a calibration method and system for intelligent drilling calculation models. Applied to the petroleum engineering field, this method and system monitors calculation deviations in real time, utilizes multi-core, multi-threaded parallel multi-group parameter-tuning models, and selects the optimal model or set of models with the smallest calculation error for application. This ensures consistently accurate calculation results during drilling, solving the problem of existing models requiring professional parameter tuning and having poor sustainability of calculation accuracy. This promotes the application of automated and intelligent drilling systems, ultimately helping drilling engineers achieve efficient and high-quality drilling. Therefore, this invention not only continuously improves the calculation accuracy of the calculation model during drilling but also achieves automatic model calibration, eliminating the need for time-consuming and labor-intensive configuration, continuous tracking, monitoring, and adjustment by experts. Furthermore, this invention not only helps improve calculation accuracy, assisting drilling personnel in better monitoring downhole conditions and predicting drilling risks, but also eliminates the obstacle of requiring continuous maintenance by professional experts for the deployment and application of automated drilling systems, making automated / intelligent systems a more cost-effective choice for drilling operations.

[0069] The above description is merely a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

[0070] It should be understood that the embodiments disclosed herein are not limited to the specific structures, processing steps, or materials disclosed herein, but should be extended to equivalent substitutions of these features as understood by those skilled in the art. It should also be understood that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting.

[0071] The phrase "an embodiment" or "an embodiment" used in this specification means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the invention. Therefore, the phrase "an embodiment" or "an embodiment" appearing in various places throughout the specification does not necessarily refer to the same embodiment.

[0072] While the embodiments disclosed in this invention are as described above, the content is merely for the purpose of facilitating understanding of the invention and is not intended to limit the invention. Any person skilled in the art to which this invention pertains may make any modifications and changes in form and detail of the implementation without departing from the spirit and scope disclosed herein; however, the scope of patent protection of this invention shall still be determined by the scope defined in the appended claims.

Claims

1. A calibration method for an intelligent drilling calculation model, characterized in that, include: Obtain the current drilling static data and actual drilling measurement data. Based on this, use the original drilling calculation model used in the current drilling process to calculate the drilling state parameters. Based on the drilling status parameters and the actual drilling measurement data, the deviation between the drilling status calculated by the model and the actual measurement status is monitored in real time, and the cause of the current deviation is analyzed based on the real-time deviation monitoring results. When the current deviation is caused by model calculation, the original drilling calculation model is corrected by evaluating the current deviation amount, including: Identify the current drilling rig type; Based on the drilling rig type identification results, and according to the drilling operation control parameters under stable conditions or the addition of drilling operation control parameters with preset characteristics and fluctuations, multiple preset parameter adjustment models are used to calculate the deviation between the drilling state parameters corresponding to each parameter adjustment model and the actual drilling measurement data, so as to correct the original drilling calculation model. The multiple parameter adjustment models are formed by adjusting one or more model parameters of the original drilling calculation model with different strategies.

2. The correction method according to claim 1, characterized in that, The steps for analyzing the causes of the current deviation based on real-time deviation monitoring results include: When the current deviation exceeds a preset first deviation threshold, further analysis is conducted to determine if the current deviation is caused by fluctuations in the current drilling operation control parameters. If the current deviation is not caused by fluctuations in the current drilling operation control parameters, the original drilling calculation model needs to be corrected, and a model correction command should be generated immediately.

3. The correction method according to claim 2, characterized in that, The drilling control parameters are diagnosed to determine whether fluctuations occur within a specified time period in order to identify the cause of the current deviation. If fluctuations occur, the drilling status parameters are recalculated. If no fluctuations occur, then the current deviation is determined to be caused by model calculations.

4. The correction method according to claim 2, characterized in that, If the current deviation is less than the first deviation threshold, the drilling status parameters continue to be calculated.

5. The correction method according to claim 1, characterized in that, When the current drilling rig is a non-automatic drilling rig, obtain the drilling operation control parameters under stable conditions; Based on the drilling operation control parameters under the steady state, drilling state parameters under different simulation conditions are calculated using multiple preset parameter adjustment models. Furthermore, combined with actual drilling measurement data under the steady state, the calculation deviation of the parameter adjustment model corresponding to different simulation conditions is obtained. Based on the deviation calculated by the parameter tuning model, the optimal parameter tuning model is selected and used to replace and update the original drilling calculation model. Alternatively, based on the deviation calculated by the parameter tuning model, multiple first backup parameter tuning models are selected and the model set formed by the multiple first backup parameter tuning models is used to replace and update the original drilling calculation model.

6. The correction method according to claim 1, characterized in that, When the drilling rig is currently an automated drilling rig, small periodic fluctuations are added to the control parameters of various drilling operations through the drilling rig control system; Based on the drilling operation control parameters with added micro-periodic fluctuations, the drilling state parameters under different simulation conditions are calculated using a variety of preset parameter adjustment models. Furthermore, by combining the actual drilling measurement data obtained after adding micro-periodic fluctuations to the input of the original drilling calculation model, the calculation deviation of the parameter adjustment model corresponding to different simulation conditions is obtained. Based on the deviation calculated by the parameter tuning model, multiple second backup parameter tuning models are selected, and the original drilling calculation model is replaced and updated using the model set formed by the multiple second backup parameter tuning models.

7. The correction method according to any one of claims 1 to 6, characterized in that, The real-time deviation can be monitored or the deviation can be calculated by comparing the various first-type wellhead parameters in the drilling status parameters with the various second-type wellhead parameters in the real-time logging data.

8. The correction method according to claim 7, characterized in that, The process of monitoring real-time deviations or calculating deviations includes: Multiple first-type parameter curves are generated based on the aforementioned first-type wellhead parameters, and multiple second-type parameter curves are generated based on the aforementioned second-type wellhead parameters; By calculating the difference in slope between the multiple first-type parameter curves and the multiple second-type parameter curves, or the degree of change in curve spatial distance, or the degree of change in curve similarity, the real-time deviation can be monitored or the deviation can be calculated.

9. The correction method according to any one of claims 1 to 6, characterized in that, The drilling static data includes at least the wellbore structure, wellbore trajectory, drill string assembly, and formation stratification information. The actual drilling measurement data includes at least real-time logging data and logging-while-drilling data. The drilling state parameters include at least the pressure distribution characteristics in the wellbore, cuttings distribution characteristics, drill string friction distribution characteristics, drill string torque distribution characteristics, wellhead stand-up pressure, wellhead casing pressure, and wellhead torque.

10. The correction method according to claim 7, characterized in that, The wellhead parameters include at least stand pressure, casing pressure, and torque.

11. The correction method according to claim 1, characterized in that, Multiple parameter tuning models are run in parallel using a multi-core, multi-threaded approach, and the drilling state parameters corresponding to each parameter tuning model are calculated.

12. A calibration system for an intelligent drilling calculation model, characterized in that, The calibration system is performed according to the calibration method as described in any one of claims 1 to 11, wherein the calibration system comprises: The target well database stores the static drilling data and actual drilling measurement data of the currently drilling wells; The real-time calculation module is used to acquire the current drilling static data and actual drilling measurement data. Based on this, it calculates the drilling state parameters using the original drilling calculation model used in the current drilling process. The model deviation monitoring module is used to monitor the deviation between the drilling state calculated by the model and the actual measurement state in real time based on the drilling state parameters and the actual drilling measurement data, and to analyze the cause of the current deviation based on the real-time deviation monitoring results. An automatic correction module is used to correct the original drilling calculation model by evaluating the current deviation amount when the current deviation is caused by model calculation.

Images