Parameter prediction model training method, anti-drilling characteristic parameter prediction method and equipment
By constructing a parameter prediction model on the Bayesian maximum entropy algorithm and fusing well measurement and recording data for iterative training, the problem of predicting anti-drilling characteristic parameters in complex lithologic formations of deep wells and ultra-deep wells is solved, achieving higher prediction accuracy and real-time performance, guiding the effective speed up of drilling.
Patent Information
- Application Number
- CN202311412665.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-10-27
- Publication Date
- 2025-04-29
AI Technical Summary
During the drilling process of oil and gas fields, the prediction of drilling resistance characteristics parameters is difficult to accurately predict, especially in complex lithologic formations of deep and ultra-deep wells. It is difficult for the existing technology to effectively utilize logging and seismic data, resulting in inaccurate prediction results.
The parameter prediction model is constructed based on the Bayesian maximum entropy algorithm, and the well measurement and recording data during the drilling process are fused, and iterative training is carried out to predict the drilling resistance characteristic parameters of the target well in real time.
It improves the accuracy of prediction of formation resistance characteristic parameters during drilling in deep and extra-deep reservoirs, and is suitable for complex geological environments and complex wellbore conditions. It can provide accurate drill resistance characteristic parameters in real time and guide the speed up of drilling.
Smart Images
Figure CN119903335A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of oil exploration technology, and in particular to a training method for a parameter prediction model, and a method and device for predicting anti-drilling characteristic parameters. Background Art
[0002] During the drilling process of oil and gas fields, the drilling speed is affected by many geological factors. The anti-drilling characteristic parameters are an important basis for optimizing the drilling speed, including indicators such as drillability grade, rock abrasiveness, and rock compressive strength. The anti-drilling characteristic parameter values are usually obtained based on the interpolation and fusion of indoor experiments, logging data, and seismic data. However, during the drilling process in the mine, due to cost issues, it is impossible to obtain logging and seismic data from all adjacent well areas.
[0003] The common process of predicting formation anti-drilling characteristic parameters based on logging data can only consider the parameters in the empirical formula, and it is difficult to coordinate with other measurement methods to predict the formation anti-drilling characteristic parameters. For complex lithology formations in deep and ultra-deep wells, the temperature, pressure, and ground stress parameters in the same well section vary complexly, and it is also difficult to predict the formation anti-drilling characteristic parameters of the sections to be drilled and being drilled by empirical formulas and spatial interpolation. Summary of the invention
[0004] The present application provides a parameter prediction model training method, an anti-drilling characteristic parameter prediction method and device, which can improve the accuracy of predicting formation anti-drilling characteristic parameters. The technical solution is as follows:
[0005] On the one hand, an embodiment of the present application provides a method for training a parameter prediction model, comprising:
[0006] Obtain sample well logging data and sample mud logging data based on the drilling data of the sample well;
[0007] Fusion of the sample well logging data and the sample mud logging data to construct a sample data set;
[0008] The sample data set is used to iteratively train a parameter prediction model, which is constructed based on a Bayesian maximum entropy algorithm and is used to predict parameters based on real-time logging data of a target well, and output anti-drilling characteristic parameters of the target well, wherein the anti-drilling characteristic parameters include at least one of a drillability extreme value, a grinding coefficient, rock tensile strength, and rock compressive strength.
[0009] On the other hand, an embodiment of the present application provides a method for predicting anti-drilling characteristic parameters, comprising:
[0010] Obtain real-time logging data and real-time well logging data of the target well;
[0011] The real-time logging data and the real-time well logging data are input into a parameter prediction model to obtain the anti-drilling characteristic parameters of the target well. The parameter prediction model is constructed based on a Bayesian maximum entropy algorithm and is obtained by iterative training using a sample data set. The anti-drilling characteristic parameters include at least one of drillability extreme value, grinding coefficient, rock tensile strength and rock compressive strength.
[0012] On the other hand, an embodiment of the present application provides a training device for a parameter prediction model, comprising:
[0013] A first acquisition module is used to acquire sample well logging data and sample mud logging data based on the drilling data of the sample well;
[0014] A data fusion module, used for fusing the sample well logging data and the sample mud logging data to construct a sample data set;
[0015] A model training module is used to iteratively train a parameter prediction model using the sample data set. The parameter prediction model is constructed based on a Bayesian maximum entropy algorithm and is used to perform parameter prediction based on real-time logging data of a target well, and output the anti-drilling characteristic parameters of the target well. The anti-drilling characteristic parameters include at least one of a drillability extreme value, a grinding coefficient, rock tensile strength, and rock compressive strength.
[0016] On the other hand, an embodiment of the present application provides an anti-drilling characteristic parameter prediction device, comprising:
[0017] The second acquisition module is used to acquire real-time logging data and real-time well logging data of the target well;
[0018] A parameter prediction module is used to input the real-time logging data and the real-time well logging data into a parameter prediction model to obtain the anti-drilling characteristic parameters of the target well. The parameter prediction model is constructed based on a Bayesian maximum entropy algorithm and is obtained by iterative training using a sample data set. The anti-drilling characteristic parameters include at least one of a drillability extreme value, a grinding coefficient, a rock tensile strength, and a rock compressive strength.
[0019] On the other hand, an embodiment of the present application provides an electronic device, which includes a memory and a processor; a computer program is stored in the memory, and when the computer program is executed by the processor, the method described in the above aspects is implemented.
[0020] On the other hand, an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and the computer program is loaded and executed by a processor to implement the method as described in the above aspects.
[0021] The technical solution provided by this application includes at least the following beneficial effects:
[0022] The parameter prediction model training method, anti-drilling characteristic parameter prediction method and equipment provided in the present application, by integrating the logging data during the drilling process, construct a parameter prediction model based on the Bayesian maximum entropy algorithm, thereby improving the accuracy of the prediction of the formation anti-drilling characteristic parameters during the drilling process of deep and ultra-deep oil reservoirs. It is suitable for complex geological environments such as deep, ultra-deep and shale oil and gas, as well as complex wellbore conditions such as directional wells and horizontal wells. It can predict the formation anti-drilling characteristic parameters in real time, and provide effective guidance for speeding up drilling. BRIEF DESCRIPTION OF THE DRAWINGS
[0023] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for describing the embodiments.
[0024] Figure 1 is a flow chart of a method for training a parameter prediction model provided by an exemplary embodiment of the present application;
[0025] Figure 2 is a schematic diagram of sample well logging data and sample mud logging data provided by an exemplary embodiment of the present application;
[0026] Figure 3 A schematic diagram of a parameter prediction model provided by an exemplary embodiment of the present application;
[0027] Figure 4 is a flowchart of a method for training a parameter prediction model provided by another exemplary embodiment of the present application;
[0028] Figure 5 is a flow chart of a method for predicting anti-drilling characteristic parameters provided by an exemplary embodiment of the present application;
[0029] Figure 6 It is a schematic diagram of the identification result of drilling status of real-time curve data provided by an exemplary embodiment of the present application;
[0030] Figure 7 is a schematic diagram of actual rock compressive strength based on well logging data interpretation provided by an exemplary embodiment of the present application;
[0031] Figure 8 is a schematic diagram of compressive strength prediction results provided by an exemplary embodiment of the present application;
[0032] Fig. 9 is a schematic diagram of actual drillability based on well logging data interpretation provided by an exemplary embodiment of the present application;
[0033] Fig.10 is a schematic diagram of drillability prediction results provided by an exemplary embodiment of the present application;
[0034] Fig.11 is a structural block diagram of a training device for a parameter prediction model provided by an exemplary embodiment of the present application;
[0035] Fig.12 is a structural block diagram of an anti-drilling characteristic parameter prediction device provided by an exemplary embodiment of the present application;
[0036] Fig.13 It is a structural block diagram of an electronic device provided by an exemplary embodiment of the present application. DETAILED DESCRIPTION
[0037] In order to make the objectives, technical solutions and advantages of the present application clearer, the implementation methods of the present application will be further described in detail below with reference to the accompanying drawings.
[0038] Embodiment 1
[0039] Please refer to Figure 1 , which shows a flow chart of a method for training a parameter prediction model provided by an exemplary embodiment of the present application. The method comprises the following steps:
[0040] Step 101, acquiring sample well logging data and sample mud logging data based on the drilling data of the sample well.
[0041] In a possible implementation, the electronic device acquires drilling data, collects and manages the logging data and mud logging data, and constructs a database. The logging data includes but is not limited to gamma, acoustic time difference, density, etc., and the mud logging data includes but is not limited to well trajectory measurement data, drilling tool assembly data, and mud logging information, such as the rotation speed of the well depth rotary table, the rotation speed of the top drive, torque, mud density, well deviation, drill bit model, etc.
[0042] Indicative, Figure 2 It shows part of sample well logging data and sample logging data acquired by an electronic device.
[0043] Step 102: Fusing sample well logging data and sample mud logging data to construct a sample data set.
[0044] The electronic device performs data fusion based on the acquired data and establishes a sample data set so that the parameter prediction model can perform correlation analysis on the logging data, mud logging data and corresponding geological parameters.
[0045] The embodiment of the present application collaboratively predicts formation anti-drilling characteristic parameters based on multi-angle information such as logging data and path data. By integrating logging data during drilling, the accuracy of formation anti-drilling characteristic parameter prediction during drilling of deep and ultra-deep oil reservoirs is improved.
[0046] Step 103, iteratively train the parameter prediction model using the sample data set. The parameter prediction model is constructed based on the Bayesian maximum entropy algorithm and is used to perform parameter prediction based on the real-time logging data of the target well, and output the anti-drilling characteristic parameters of the target well.
[0047] The anti-drilling characteristic parameters include at least one of the drillability extreme value, the grinding coefficient, the rock tensile strength and the rock compressive strength.
[0048] Bayesian maximum entropy is a modern geostatistical method that can integrate data of different precision and quality from multiple methods when conducting spatial distribution research, and does not require the original data to follow Gaussian distribution. In the process of data processing, the data is mainly divided into hard data and soft data related to the study area, as well as expert knowledge and statistical laws.
[0049] The electronic equipment builds a parameter prediction model based on the Bayesian maximum entropy algorithm, analyzes the sample data, and establishes the response relationship between the logging data (gamma, density, sonic logging), the logging data (drill bit type, drill bit assembly, torque, hook load, mud density) and the formation's anti-drilling characteristics. Figure 3 A schematic diagram of a parameter prediction model is shown.
[0050] Compared with the traditional methods based on linear extrapolation, inverse distance interpolation and Kriging interpolation, the rock anti-drilling characteristic calculation model based on the Bayesian maximum entropy algorithm has the advantages of strong computational robustness and high computational accuracy, which reduces the cost of logging while drilling and reduces the uncertainty of result prediction caused by incomplete logging data.
[0051] In summary, the training method of the parameter prediction model provided in this application, by integrating the logging data during the drilling process, constructs a parameter prediction model based on the Bayesian maximum entropy algorithm, thereby improving the accuracy of the prediction of the formation anti-drilling characteristic parameters during the drilling process of deep and ultra-deep oil reservoirs. It is suitable for complex geological environments such as deep, ultra-deep and shale oil and gas, as well as complex wellbore conditions such as directional wells and horizontal wells. It can predict the formation anti-drilling characteristic parameters in real time, and provide effective guidance for speeding up drilling.
[0052] Embodiment 2
[0053] Please refer to Figure 4 , which shows a flow chart of a method for training a parameter prediction model provided by another exemplary embodiment of the present application. The method comprises the following steps:
[0054] Step 401, acquiring sample well logging data and sample mud logging data based on the drilling data of the sample well.
[0055] The specific implementation of step 401 may refer to the above step 101, and the embodiment of the present application will not be repeated here.
[0056] Step 402 , calibrate the sample well logging data and the sample mud logging data based on the same depth axis to construct an attribute set distributed in a three-dimensional space.
[0057] Electronic devices are based on Figure 2 The sample logging data and sample logging data shown are used to construct an attribute set for predicting the distribution of anti-drilling characteristic parameters in three-dimensional space. All attributes are calibrated together through the same depth axis. The specific attribute values include but are not limited to: depth, gamma logging curve, acoustic time difference curve, density logging curve, torque, drilling pressure, screw type and drill bit type. For data related to models such as screw type and drill bit type, the electronic device uniformly numbers them. For example, the electronic device sets the screw type and drill bit type numbers to Tl and Tb respectively, and then uses encoding technology to convert them into a binary array.
[0058] Step 403 , randomly distribute the attribute sets arranged according to the depth sequence to obtain a sample data set.
[0059] After the electronic device completes data collection and data management, it randomly distributes the data sets previously arranged in depth sequence to improve the accuracy of model prediction.
[0060] Step 404, based on the Bayesian maximum entropy algorithm, a parameter prediction model is constructed with preset parameters as initial model hyperparameters.
[0061] In one possible implementation, the electronic device uses the range, coefficient of variation, and statistical probability density similar to the Kriging algorithm as the initial model hyperparameters of the parameter prediction model, updates the hyperparameters of the Bayesian maximum entropy model during the model training phase, and establishes an anti-drilling characteristic parameter prediction model based on the logging data.
[0062] Step 405: Calculate sample anti-drilling characteristic parameters based on the sample well logging data and the sample mud logging data.
[0063] The electronic device calculates the sample anti-drilling characteristic parameters based on the sample logging data and the sample mud logging data, and optimizes the model parameters using the sample anti-drilling characteristic parameters actually calculated and the prediction results of the model training stage.
[0064] Specifically, the sample anti-drilling characteristic parameters include at least one of drillability extreme value, grinding coefficient, rock tensile strength and rock compressive strength.
[0065] The calculation formula for the drillability extreme value includes
[0066] K d =log2t
[0067] Among them, K d is the extreme value of drillability, and t is the average drilling time;
[0068] The calculation formula of grinding coefficient includes
[0069]
[0070] Among them, A f is the grinding coefficient of the formation, x1 is the wear slowing coefficient of the drill bit teeth, h f is the wear of drill teeth, n r is the rotation speed, D1 is the drilling pressure influence coefficient, d b is the drill bit diameter, p is the drill bit coefficient, t d is the pure drilling time of the drill bit, W cr is the critical specific drilling pressure per unit drill bit diameter, W s The drilling pressure per unit drill bit diameter;
[0071] The calculation formula for the tensile strength of rock includes
[0072]
[0073] Among them, S t is the tensile strength, V p is the longitudinal wave velocity, V s is the shear wave velocity, V cl is the mud content in the rock;
[0074] The calculation formula for rock compressive strength includes
[0075] σ c =0.0045E(1-V cl )+0.008V cl E
[0076] Among them, E is Young's elastic model.
[0077] Step 406, using the sample data set as the model input, the sample anti-drilling characteristic parameters as the target output, using the grid optimization method to initialize the model hyperparameters, and iteratively training the parameter prediction model.
[0078] The electronic equipment establishes a Bayesian maximum entropy prediction model, takes well depth, logging curves (gamma logging, sonic time difference logging, density logging, etc.), and logging data (drill bit size, drill bit type, screw type, torque, etc.) as input parameters, and formation anti-drilling characteristic parameters (rock drillability grade, rock abrasiveness, rock strength) as output parameters, and takes the range, coefficient of variation, and statistical probability density similar to the Kriging algorithm as initial hyperparameters, and uses the grid optimization method to determine the hyperparameters of the Bayesian maximum entropy model.
[0079] In the embodiment of the present application, by establishing the response relationship between logging data, mud logging data and formation anti-drilling characteristics, integrating logging data in the drilling process, and considering the statistical correspondence between different parameters, the accuracy of the prediction of formation anti-drilling characteristic parameters in the drilling process of deep and ultra-deep oil reservoirs is improved. It is helpful to provide scientific drilling tool optimization and construction parameter adjustment in the process of drilling parameter optimization, and can reduce drilling risks and improve drilling economic benefits based on data models while ensuring the interpretability of predictions.
[0080] Embodiment 3
[0081] Please refer to Figure 5 , which shows a flow chart of a method for predicting anti-drilling characteristic parameters provided by another exemplary embodiment of the present application. The method comprises the following steps:
[0082] Step 501, obtaining real-time logging data and real-time well logging data of a target well.
[0083] In the parameter prediction stage, the electronic device obtains the real-time logging data and real-time logging data corresponding to the target well for the predicted parameters. Among them, the logging data includes but is not limited to gamma, acoustic time difference, density, etc., and the logging data includes but is not limited to well trajectory measurement data, drilling tool assembly data, logging information, such as the rotation speed of the well depth rotary table, the rotation speed of the top drive, torque, mud density, well deviation, drill bit model, etc.
[0084] Step 502: input the real-time logging data and the real-time well logging data into a parameter prediction model to obtain the anti-drilling characteristic parameters of the target well.
[0085] The parameter prediction model is constructed based on the Bayesian maximum entropy algorithm and is obtained by iterative training using sample data sets. The anti-drilling characteristic parameters include at least one of drillability extreme value, grinding coefficient, rock tensile strength and rock compressive strength.
[0086] The model training process of the parameter prediction model can refer to the above Figure 1 and Figure 4 The corresponding embodiments are not described in detail in the embodiments of the present application.
[0087] To summarize, the method for predicting anti-drilling characteristic parameters provided in the present application, by integrating the logging data during the drilling process, constructs a parameter prediction model based on the Bayesian maximum entropy algorithm, thereby improving the accuracy of prediction of formation anti-drilling characteristic parameters during the drilling process of deep and ultra-deep oil reservoirs. It is suitable for complex geological environments such as deep, ultra-deep and shale oil and gas, as well as complex wellbore conditions such as directional wells and horizontal wells. It can predict formation anti-drilling characteristic parameters in real time, and provide effective guidance for speeding up drilling.
[0088] Optionally, when the electronic device obtains the relevant data of the target well, it first pre-processes the data to filter out the valid data. The above step 501 specifically includes the following steps:
[0089] Step 1: Obtain initial logging data and initial well logging data of the target well.
[0090] Step 2: Screen and process the initial logging data and initial well logging data based on the abnormal point screening algorithm to obtain real-time logging data and real-time well logging data corresponding to the actual drilling process.
[0091] In one possible implementation, the electronic device inputs real-time logging data and real-time well logging data from the remote transmission system, uses an abnormal point screening algorithm to obtain data of the actual drilling process, and uses the real-time input well depth, wellbore trajectory, drill bit and screw model and other drilling tool combinations, and drilling fluid density data as input parameters to perform real-time prediction of the formation's anti-drilling characteristic parameters. Figure 6 A schematic diagram of the result of identifying the drilling status based on real-time curve data is shown.
[0092] Optionally, the electronic device may also use a sliding window correction method to ensure that the acquired data is real-time data during the drilling process.
[0093] Embodiment 4
[0094] In combination with the above embodiments, the present application embodiment provides a parameter prediction model training and parameter prediction process as follows:
[0095] (1) Data Governance
[0096] A database of wellbore structure of neighboring wells in the block, drilling tool combination, rock anti-drilling characteristic parameters (rock abrasiveness, drillability grade, rock strength), and logging data (geological logging and engineering logging data) is established.
[0097] (2) Based on the existing data, a correlation analysis matrix is established among drilling parameters, logging curve data, logging parameters, and geological parameters.
[0098] (3) Create a training dataset
[0099] Based on steps (1) and (2), a data set based on the prediction of formation anti-drilling characteristics is established. All features are
[0100] (4) Establishing a prediction model for anti-drilling characteristic parameters using the Bayesian maximum entropy method
[0101] A Bayesian maximum entropy method was established, and the drilling parameters (bit pressure, rotary table speed, pump displacement), geological parameters (lithology, geological stratification), logging data (gamma, sonic time difference, density), and logging parameters (drill bit type, drill bit assembly, torque, hook load, mud density) were respectively input into the Bayesian maximum entropy model for the prediction of formation anti-drilling characteristic parameters (drillability grade, abrasiveness, rock strength).
[0102] (5) Real-time calculation of formation anti-drilling characteristic parameters
[0103] The real-time data of the drilling process is screened out, and the aforementioned drilling parameters, logging curve data, logging parameters, geological parameters, etc. are imported into the Bayesian maximum entropy model using a remote data acquisition system to realize the real-time calculation of the formation anti-drilling characteristic parameters.
[0104] Figure 7 and Figure 8 The actual rock compressive strength based on the interpretation of the well logging data and the rock compressive strength predicted by the method provided in the embodiment of the present application are respectively shown. Fig. 9 and Fig.10 The actual drillability based on the interpretation of the logging data and the drillability predicted by the method provided in the embodiment of the present application are shown respectively. It can be seen from the above experimental comparison results that the anti-drilling characteristic parameters predicted by the method provided in the present application are more accurate.
[0105] Embodiment 5
[0106] Please refer to Fig.11 , which shows a structural block diagram of a training device for a parameter prediction model provided by an exemplary embodiment of the present application, the device comprising:
[0107] The first acquisition module 1101 is used to acquire sample well logging data and sample mud logging data based on the drilling data of the sample well;
[0108] A data fusion module 1102 is used to fuse the sample well logging data and the sample mud logging data to construct a sample data set;
[0109] The model training module 1103 is used to iteratively train the parameter prediction model using the sample data set. The parameter prediction model is constructed based on the Bayesian maximum entropy algorithm, and is used to predict parameters based on the real-time logging data of the target well, and output the anti-drilling characteristic parameters of the target well. The anti-drilling characteristic parameters include at least one of the extreme value of drillability, grinding coefficient, rock tensile strength and rock compressive strength.
[0110] Optionally, the data fusion module 1102 is further used for:
[0111] Calibrate the sample well logging data and the sample mud logging data based on the same depth axis to construct an attribute set distributed in a three-dimensional space;
[0112] The attribute sets arranged according to the depth sequence are randomly distributed to obtain the sample data set.
[0113] Optionally, the model training module 1103 is further used to:
[0114] Based on the Bayesian maximum entropy algorithm, the parameter prediction model is constructed with preset parameters as initial model hyperparameters;
[0115] Calculating sample anti-drilling characteristic parameters based on the sample well logging data and the sample logging data;
[0116] The sample data set is used as the model input, the sample anti-drilling characteristic parameters are used as the target output, and the initial model hyperparameters of the grid optimization method are used to iteratively train the parameter prediction model.
[0117] Optionally, the sample anti-drilling characteristic parameter includes at least one of a drillability extreme value, a grinding coefficient, a rock tensile strength, and a rock compressive strength;
[0118] The calculation formula for the drillability extreme value includes
[0119] K d =log2t
[0120] Among them, K d is the extreme value of drillability, and t is the average drilling time;
[0121] The calculation formula of grinding coefficient includes
[0122]
[0123] Among them, A f is the grinding coefficient of the formation, c1 is the wear slowing coefficient of the drill bit teeth, h f is the wear of drill teeth, n r is the rotation speed, D1 is the drilling pressure influence coefficient, d b is the drill bit diameter, p is the drill bit coefficient, t d is the pure drilling time of the drill bit, W cr is the critical specific drilling pressure per unit drill bit diameter, W s The drilling pressure per unit drill bit diameter;
[0124] The calculation formula for the tensile strength of rock includes
[0125]
[0126] Among them, S t is the tensile strength, Vp is the longitudinal wave velocity, V s is the shear wave velocity, V cl is the mud content in the rock;
[0127] The calculation formula for rock compressive strength includes
[0128] σ c =0.0045E(1-V cl )+0.008V cl E
[0129] Among them, E is Young's elastic model.
[0130] Please refer to Fig.12 , which shows a structural block diagram of an anti-drilling characteristic parameter prediction device provided by an exemplary embodiment of the present application, the device comprising:
[0131] The second acquisition module 1201 is used to acquire real-time logging data and real-time well logging data of the target well;
[0132] The parameter prediction module 1202 is used to input the real-time logging data and the real-time well logging data into a parameter prediction model to obtain the anti-drilling characteristic parameters of the target well. The parameter prediction model is constructed based on a Bayesian maximum entropy algorithm and is obtained by iterative training using a sample data set. The anti-drilling characteristic parameters include at least one of drillability extreme value, grinding coefficient, rock tensile strength and rock compressive strength.
[0133] Optionally, the second acquisition module 1201 is further used to:
[0134] Acquiring initial logging data and initial well logging data of the target well;
[0135] The initial logging data and the initial well logging data are screened based on an abnormal point screening algorithm to obtain the real-time logging data and the real-time well logging data corresponding to the actual drilling process.
[0136] Embodiment 5
[0137] An embodiment of the present application provides an electronic device; Fig.13 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application is shown in FIG. Fig.13As shown, the electronic device 1300 includes: a processor 1301, at least one communication bus 1302, a user interface 1303, at least one external communication interface 1304, and a memory 1305. The communication bus 1302 is configured to realize connection and communication between these components. The user interface 1303 may include a display screen, and the external communication interface 1304 may include a standard wired interface and a wireless interface. The processor 1301 is configured to execute the program of the parameter prediction model training method or the anti-drilling characteristic parameter prediction method stored in the memory to implement the steps in the method provided in the above embodiment.
[0138] An embodiment of the present application further provides a computer-readable storage medium, in which a computer program is stored. The computer program is loaded and executed by a processor to implement the method described in the above embodiment.
[0139] The embodiment of the present application further provides a computer program product, which runs on a processor of a computer device, so that the computer device executes the method described in the above embodiment.
[0140] It should be noted here that the description of the above storage medium, electronic device, and remote control embodiments is similar to the description of the above method embodiments, and has similar beneficial effects as the method embodiments. For technical details not disclosed in the storage medium and device embodiments of this application, please refer to the description of the method embodiments of this application for understanding.
[0141] It should be understood that "one embodiment" or "an embodiment" mentioned throughout the specification means that specific features, structures or characteristics related to the embodiment are included in at least one embodiment of the present application. Therefore, "in one embodiment" or "in an embodiment" appearing throughout the specification does not necessarily refer to the same embodiment. In addition, these specific features, structures or characteristics can be combined in one or more embodiments in any suitable manner. It should be understood that in various embodiments of the present application, the size of the sequence number of the above-mentioned processes does not mean the order of execution, and the execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application. The above-mentioned sequence numbers of the embodiments of the present application are only for description and do not represent the advantages and disadvantages of the embodiments.
[0142] It should be noted that, in this article, the terms "include", "comprises" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, object or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, object or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, object or device including the element.
[0143] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.
[0144] The units described above as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units; they may be located in one place or distributed on multiple network units; some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.
[0145] In addition, all functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may be a separate unit, or two or more units may be integrated into one unit; the above-mentioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional units.
[0146] A person of ordinary skill in the art can understand that: all or part of the steps of implementing the above-mentioned method embodiment can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps of the above-mentioned method embodiment; and the aforementioned storage medium includes: mobile storage devices, read-only memories (ROM, Read Only Memory), disks or optical disks, etc. Various media that can store program codes.
[0147] Alternatively, if the above-mentioned integrated unit of the present application is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present application can essentially or in other words, the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a storage medium and includes a number of instructions for a controller to execute all or part of the methods described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, ROMs, magnetic disks, or optical disks.
[0148] The above is only an implementation method of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.
Claims
1. A training method for a parameter prediction model, characterized in that: include: Obtain sample well logging data and sample mud logging data based on the drilling data of the sample well; Fusion of the sample well logging data and the sample mud logging data to construct a sample data set; The sample data set is used to iteratively train a parameter prediction model, which is constructed based on a Bayesian maximum entropy algorithm and is used to predict parameters based on real-time logging data of a target well, and output anti-drilling characteristic parameters of the target well, wherein the anti-drilling characteristic parameters include at least one of a drillability extreme value, a grinding coefficient, rock tensile strength, and rock compressive strength.
2. The method according to claim 1, characterized in that The fusing the sample well logging data and the sample mud logging data to construct a sample data set includes: Calibrate the sample well logging data and the sample mud logging data based on the same depth axis to construct an attribute set distributed in a three-dimensional space; The attribute sets arranged according to the depth sequence are randomly distributed to obtain the sample data set.
3. The method according to claim 1, characterized in that The iterative training of the parameter prediction model using the sample data set includes: Based on the Bayesian maximum entropy algorithm, the parameter prediction model is constructed with preset parameters as initial model hyperparameters; Calculating sample anti-drilling characteristic parameters based on the sample well logging data and the sample logging data; The sample data set is used as the model input, the sample anti-drilling characteristic parameters are used as the target output, and the initial model hyperparameters of the grid optimization method are used to iteratively train the parameter prediction model.
4. The method according to claim 3, characterized in that The sample anti-drilling characteristic parameter includes at least one of a drillability extreme value, a grinding coefficient, a rock tensile strength, and a rock compressive strength; The calculation formula for the drillability extreme value includes K d =log2t Among them, K d is the extreme value of drillability, and t is the average drilling time; The calculation formula of grinding coefficient includes Among them, A f is the grinding coefficient of the formation, c1 is the wear slowing coefficient of the drill bit teeth, h f is the wear of drill teeth, n r is the rotation speed, D1 is the drilling pressure influence coefficient, d b is the drill bit diameter, p is the drill bit coefficient, t d is the pure drilling time of the drill bit, W cr is the critical specific drilling pressure per unit drill bit diameter, W s The drilling pressure per unit drill bit diameter; The calculation formula for the tensile strength of rock includes Among them, S t is the tensile strength, V p is the longitudinal wave velocity, V s is the shear wave velocity, V cl is the mud content in the rock; The calculation formula for rock compressive strength includes s c =0.0045E(1-V cl )+0.008V cl E Among them, E is Young's elastic model.
5. A method for predicting anti-drilling characteristic parameters, characterized in that: include: Obtain real-time logging data and real-time well logging data of the target well; The real-time logging data and the real-time well logging data are input into a parameter prediction model to obtain the anti-drilling characteristic parameters of the target well. The parameter prediction model is constructed based on a Bayesian maximum entropy algorithm and is obtained by iterative training using a sample data set. The anti-drilling characteristic parameters include at least one of drillability extreme value, grinding coefficient, rock tensile strength and rock compressive strength.
6. The method according to claim 5, characterized in that The step of obtaining the real-time logging data and real-time well logging data of the target well includes: Acquiring initial logging data and initial well logging data of the target well; The initial logging data and the initial well logging data are screened based on an abnormal point screening algorithm to obtain the real-time logging data and the real-time well logging data corresponding to the actual drilling process.
7. A training device for a parameter prediction model, characterized in that: include: A first acquisition module is used to acquire sample well logging data and sample mud logging data based on the drilling data of the sample well; A data fusion module, used for fusing the sample well logging data and the sample mud logging data to construct a sample data set; A model training module is used to iteratively train a parameter prediction model using the sample data set. The parameter prediction model is constructed based on a Bayesian maximum entropy algorithm and is used to perform parameter prediction based on real-time logging data of a target well, and output the anti-drilling characteristic parameters of the target well. The anti-drilling characteristic parameters include at least one of a drillability extreme value, a grinding coefficient, rock tensile strength, and rock compressive strength.
8. A device for predicting anti-drilling characteristic parameters, characterized in that: include: The second acquisition module is used to acquire real-time logging data and real-time well logging data of the target well; A parameter prediction module is used to input the real-time logging data and the real-time well logging data into a parameter prediction model to obtain the anti-drilling characteristic parameters of the target well. The parameter prediction model is constructed based on a Bayesian maximum entropy algorithm and is obtained by iterative training using a sample data set. The anti-drilling characteristic parameters include at least one of a drillability extreme value, a grinding coefficient, a rock tensile strength, and a rock compressive strength.
9. An electronic device, characterized in that: The method comprises a memory and a processor; the memory stores a computer program, and when the computer program is executed by the processor, the method according to any one of claims 1 to 4, or the method according to claim 5 or 6 is implemented.
10. A computer-readable storage medium, characterized in that: A computer program is stored, and the computer program is loaded and executed by a processor to implement the method according to any one of claims 1 to 4, or the method according to claim 5 or 6.