Method for predicting mechanical parameters of stratum rock in real time based on machine learning

Abstract

The invention belongs to the technical field of intelligent oil and gas field development, particularly relates to a method and system for predicting stratum rock mechanical parameters in real time based on machine learning and electronic equipment, and aims at solving the problem that in the prior art, real-time, continuous and accurate prediction of the stratum rock mechanical parameters in the drilling process cannot be achieved. The method comprises the steps that drilling basic data corresponding to a first drilling well are obtained and serve as input data, and the first drilling well is a drilling well with stratum rock mechanical parameters to be predicted; preprocessing the input data to obtain preprocessed data; calculating the correlation of each parameter in the preprocessed data, and taking the parameter of which the correlation is located in a set correlation threshold interval as an input parameter; and inputting the input parameters into the trained rock mechanics prediction model to obtain a prediction result of the stratum rock mechanics parameters of the first drilling well. According to the invention, real-time, continuous and accurate prediction of the mechanical parameters of the stratum rock in the drilling process is realized.

Description

Technical Field

[0001] The present invention belongs to the technical field of intelligent oil and gas field development, and specifically relates to a method, system and electronic equipment for real-time prediction of formation rock mechanical parameters based on machine learning. Background Art

[0002] The most important part of oil exploration and development is drilling, the purpose of which is to break the formation rock to reach the oil and gas reservoir formation, and at the same time establish a reliable channel for long-term exploitation of oil and gas resources. Drilling involves a series of tasks, such as breaking rocks, drilling, obtaining ground stress, calculating formation pressure profiles, analyzing wellbore stability, wellbore structure design, casing strength design, hydraulic fracturing design, etc., all of which require the support of rock mechanics parameters. Rock mechanics parameters include rock elastic parameters (Young's modulus, Poisson's ratio) and strength parameters (tensile strength, tensile strength, cohesion, internal friction angle, etc.). If these parameters cannot be accurately identified and a reasonable drilling method cannot be formulated, it may lead to downhole accidents and increase costs. Correctly understanding formation rock parameters is an important basis for safe and efficient construction of drilling projects.

[0003] At present, there are two main methods for obtaining rock mechanical parameters: indoor testing of core sampling. Indoor testing of core sampling is the most basic and direct method, but the test data is limited and discrete, and cannot continuously reflect the trend of rock mechanical parameters and heterogeneity, and can only be used for post-drilling analysis; geophysical logging prediction. The geophysical logging prediction method uses empirical formulas to calculate and obtain continuous data, but professional logging instruments must be lowered, which is costly and can only be used for post-drilling analysis. In summary, in recent years, there has been great progress in the prediction methods of formation pore pressure and rock mechanical parameters. However, most of the existing prediction methods have problems such as limited scope of application, low accuracy, cumbersome calculation process, poor economic benefits, complex procedures, and can only be used for post-drilling analysis, and cannot achieve real-time, continuous and accurate prediction during the drilling process.

[0004] During the drilling process, a lot of data is involved, such as drilling pressure, drilling speed, rotation speed, torque and other drilling information, which can reflect the physical properties of underground rocks. However, a single data cannot fully characterize the characteristics of formation rocks. Machine learning methods can process large-scale, high-dimensional data, automatically extract features and learn models, realize real-time prediction and improve prediction accuracy, and adapt to different geological environments and engineering needs. Therefore, they are gradually widely used in the field of oil and gas development. Many experts and scholars have successfully applied machine learning methods in the identification of formation lithology, prediction of rock mechanics and rock anti-drilling characteristic parameters, and drill bit selection. Therefore, it has obvious advantages to use the drilling data information obtained during the drilling process to carry out the prediction of formation rock mechanics parameters. Based on this, the present invention proposes a method for real-time prediction of formation rock mechanics parameters based on machine learning. Summary of the invention

[0005] In order to solve the above problems in the prior art, that is, to solve the problem that the prior art cannot achieve real-time, continuous and accurate prediction of formation rock mechanical parameters during drilling, the first aspect of the present invention proposes a method for real-time prediction of formation rock mechanical parameters based on machine learning, the method comprising:

[0006] S100, obtaining basic drilling data corresponding to a first well as input data; the first well is a well for which formation rock mechanical parameters are to be predicted; the basic drilling data includes drilling data, mud logging data and well logging data;

[0007] S200, preprocessing the input data to obtain preprocessed data; the preprocessing includes outlier detection, maximum and minimum normalization processing, and smoothing processing;

[0008] S300, calculating the correlation of each parameter in the preprocessed data, and taking the parameters whose correlation is within a set correlation threshold range as input parameters;

[0009] S400, inputting the input parameters into a trained rock mechanics prediction model to obtain a prediction result of the formation rock mechanics parameters of the first well;

[0010] Among them, the rock mechanics prediction model is built based on LightGBM.

[0011] In some preferred embodiments, the input parameters include hook load, mechanical drilling speed, pump pressure, drilling pressure, displacement, density, and viscosity; and the predicted results include uniaxial compressive strength and internal friction angle.

[0012] In some preferred embodiments, the rock mechanics prediction model is trained by:

[0013] A100, obtaining basic drilling data corresponding to a set number of wells as input data; the basic drilling data includes drilling data, logging data and well logging data;

[0014] A200, preprocessing the input data to obtain preprocessed data; the preprocessing includes outlier detection, maximum and minimum normalization processing, and smoothing processing;

[0015] A300, calculating the correlation of each parameter in the preprocessed data, and taking the parameter whose correlation is within the set correlation threshold range as the input parameter;

[0016] A400, inputting the input parameters into a pre-built rock mechanics prediction model to obtain a prediction result of the formation rock mechanics parameters of the first well;

[0017] A500, calculating the loss value based on the prediction result and the true value label corresponding to the input parameter, and then updating the model parameters of the rock mechanics prediction model;

[0018] A600, cycle A400-A500 until a trained rock mechanics prediction model is obtained.

[0019] In some preferred embodiments, the correlation of each parameter in the preprocessed data is calculated by: calculating the correlation of each parameter in the preprocessed data by Pearson coefficient:

[0020]

[0021] Among them, ρ X,Y It represents the correlation coefficient between the set parameters, that is, the correlation. X and Y represent two feature data, that is, two parameters. cov(X,Y) is the covariance of X and Y. σ X is the standard deviation of X, σ Y is the standard deviation of Y.

[0022] In some preferred embodiments, the true value label corresponding to the uniaxial compressive strength is calculated as follows:

[0023] σ c =(0.0045+0.003V sh )E

[0024] E d =ρV s 2 (3V p 2 -4V s 2 ) / (V p 2 -2V s 2 )

[0025] V s =(0.61~0.53)V

[0026]

[0027] Among them, σ c is the uniaxial compressive strength, E d is the dynamic Young's modulus, V p is the longitudinal wave velocity, V s is the shear wave velocity, V sh is the volume content of mud, ρ is the rock density, GCUR is the Hillch index, I GR is the mud content index, GR, GR min , GRmax They represent the natural gamma values ​​of the target layer, pure mudstone layer and pure sandstone layer respectively.

[0028] In some preferred embodiments, the internal friction angle corresponds to a true value label, and the calculation method is:

[0029]

[0030]

[0031] in, represents the internal friction angle, C represents the cohesion, A is a constant related to the rock properties, μ d represents the dynamic Poisson's ratio.

[0032] In a second aspect of the present invention, a system for real-time prediction of formation rock mechanical parameters based on machine learning is proposed, the system comprising:

[0033] A data acquisition module is configured to acquire basic drilling data corresponding to a first wellbore as input data; the first wellbore is a wellbore for which formation rock mechanical parameters are to be predicted; the basic drilling data includes drilling data, logging data and well logging data;

[0034] A preprocessing module is configured to preprocess the input data to obtain preprocessed data; the preprocessing includes outlier detection, maximum and minimum normalization processing, and smoothing processing;

[0035] A parameter calculation module is configured to calculate the correlation of each parameter in the preprocessed data, and take the parameters whose correlation is within a set correlation threshold range as input parameters;

[0036] A prediction result acquisition module, configured to input the input parameters into a trained rock mechanics prediction model to obtain a prediction result of the formation rock mechanics parameters of the first wellbore;

[0037] Among them, the rock mechanics prediction model is built based on LightGBM.

[0038] The third aspect of the present invention proposes an electronic device, comprising: at least one processor; and a memory communicatively connected to at least one of the processors; wherein the memory stores instructions executable by the processor, and the instructions are used to be executed by the processor to implement the above-mentioned method for real-time prediction of formation rock mechanical parameters based on machine learning.

[0039] In a fourth aspect of the present invention, a computer-readable storage medium is proposed, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to be executed by the computer to implement the above-mentioned method for real-time prediction of formation rock mechanical parameters based on machine learning.

[0040] Beneficial effects of the present invention:

[0041] The present invention realizes the real-time, continuous and accurate prediction of the formation rock mechanical parameters during the drilling process, which is convenient for on-site personnel to timely understand the geological conditions, take early prevention measures for possible accidents, and timely adjust the construction parameters. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Other features, objects and advantages of the present application will become more apparent by reading the detailed description of non-limiting embodiments made with reference to the following drawings.

[0043] Figure 1 It is a flow chart of a method for real-time prediction of formation rock mechanical parameters based on machine learning according to an embodiment of the present invention;

[0044] Figure 2 It is a schematic diagram of a framework of a system for real-time prediction of formation rock mechanical parameters based on machine learning according to an embodiment of the present invention;

[0045] Figure 3 It is a simplified flow chart of a method for real-time prediction of formation rock mechanical parameters based on machine learning according to an embodiment of the present invention;

[0046] Figure 4 is a smoothed graph of acoustic wave time difference data according to an embodiment of the present invention;

[0047] Figure 5 is a feature correlation diagram of an embodiment of the present invention;

[0048] Figure 6 is a Young's modulus prediction result diagram of a test well (i.e., a first drilling well) according to an embodiment of the present invention;

[0049] Figure 7 It is a structural diagram of a computer system of an embodiment of the present invention that is suitable for implementing an electronic device of an embodiment of the present application. DETAILED DESCRIPTION

[0050] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution in the embodiment of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiment is a part of the embodiment of the present invention, not all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0051] The present application will be further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are only used to explain the relevant invention, rather than to limit the invention. It is also necessary to explain that, for ease of description, only the parts related to the relevant invention are shown in the accompanying drawings.

[0052] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application may be combined with each other.

[0053] The method of the present invention for real-time prediction of formation rock mechanical parameters based on machine learning is as follows: Figure 1 As shown, the following steps are included:

[0054] S100, obtaining basic drilling data corresponding to a first well as input data; the first well is a well for which formation rock mechanical parameters are to be predicted; the basic drilling data includes drilling data, mud logging data and well logging data;

[0055] S200, preprocessing the input data to obtain preprocessed data; the preprocessing includes outlier detection, maximum and minimum normalization processing, and smoothing processing;

[0056] S300, calculating the correlation of each parameter in the preprocessed data, and taking the parameters whose correlation is within a set correlation threshold range as input parameters;

[0057] S400, inputting the input parameters into a trained rock mechanics prediction model to obtain a prediction result of the formation rock mechanics parameters of the first well;

[0058] Among them, the rock mechanics prediction model is built based on LightGBM.

[0059] In order to more clearly illustrate the method for real-time prediction of formation rock mechanical parameters based on machine learning of the present invention, each step in an embodiment of the method of the present invention is described in detail below with reference to the accompanying drawings.

[0060] In the following embodiments, the training process of the rock mechanics prediction model is first described in detail, and then the process of obtaining the prediction results of the formation rock mechanics parameters of drilling by a method for real-time prediction of formation rock mechanics parameters based on machine learning is described in detail.

[0061] 1. The training process of rock mechanics prediction model, such as Figure 3 Shown

[0062] A100, obtaining basic drilling data corresponding to a set number of wells as input data; the basic drilling data includes drilling data, logging data and well logging data;

[0063] In this embodiment, 48,062 pieces of data, including 30 characteristic parameters, are preferably collected from the drilling data, logging data and daily drilling reports of 17 wells, referred to as basic drilling data, i.e., basic drilling data includes drilling data, logging data and logging data;

[0064] Among them, drilling data, that is, real-time drilling data including rotation speed, drilling pressure, torque, flow rate, temperature, etc.; logging data, that is, path data including drill bit diameter, drill tool combination, drilling fluid parameters, etc.; logging data, that is, conventional logging data including sonic logging, density logging, gamma logging, etc., automatically acquire and parse real-time drilling basic data through a fixed data format.

[0065] A200, preprocessing the input data to obtain preprocessed data; the preprocessing includes outlier detection, maximum and minimum normalization processing, and smoothing processing;

[0066] In this embodiment, a series of data preprocessing methods and processes are used to preprocess the drilling parameters, involving noise reduction, missing value processing, and smoothing processing, as follows:

[0067] The 3σ principle is used to process and detect outliers (i.e. outlier detection), and the calculation formula is:

[0068] P(|xu|>3σ)<=0.003 (1)

[0069] Among them, P is the probability of interval numerical distribution, x is the sample value, u is the mean, and σ is the standard deviation.

[0070] The calculation formula for the maximum and minimum normalization processing is:

[0071]

[0072] Among them, x * is the standard value, x min is the minimum value of the sample data, x max is the maximum value of the sample data.

[0073] The LOWESS technology is used for smoothing calculation (i.e., smoothing processing). When the smoothing parameter (alph) is 0.05 and the number of iterations (iterations) is 1, the smoothing effect is optimal and the sequence regularity is retained to the maximum extent. Figure 4 Shown is a smoothed graph of acoustic wave time difference data.

[0074] A300, calculating the correlation of each parameter in the preprocessed data, and taking the parameter whose correlation is within the set correlation threshold range as the input parameter.

[0075] In this embodiment, the correlation coefficient ρ between the main parameters is preferably calculated by the Pearson coefficient X,Y, that is, correlation, including block, well depth, mechanical speed, hook load, drilling pressure, speed, pump pressure, displacement, and Young's modulus. The calculation formula is:

[0076]

[0077] Among them, X and Y represent two feature data respectively, cov(X,Y) is the covariance of X and Y, σ X is the standard deviation of X, σ Y is the standard deviation of Y.

[0078] The feature correlation diagram obtained by analysis is as follows: Figure 5 As shown in the figure. The absolute values ​​of the correlation coefficients between mechanical drilling rate, hook load, drilling pressure, rotation speed, displacement and target parameters (Ed, pore pressure) are mostly distributed in the range of 0.2 to 0.5, and the correlation is good. This shows that engineering parameters can be effectively used for prediction. Based on the feature correlation coefficient and the knowledge in the petroleum field, 8 features were selected: hook load, mechanical drilling rate, pump pressure, drilling pressure, displacement, density, and viscosity, and 14 training sets and 3 test wells were selected.

[0079] A400, inputting the input parameters into a pre-built rock mechanics prediction model to obtain a prediction result of the formation rock mechanics parameters of the first well;

[0080] In this embodiment, it is preferred to construct a model for real-time prediction of rock mechanics parameters (i.e., a rock mechanics prediction model) based on the LightGBM (Light Gradient Boosting Machine) algorithm; the prediction results include uniaxial compressive strength and internal friction angle. In other embodiments, the prediction results can be set and selected according to actual conditions.

[0081] A500, based on the prediction result and the true value label corresponding to the input parameter, calculate the loss value, and then update the model parameters of the rock mechanics prediction model.

[0082] In this embodiment, the longitudinal wave data is converted into the shear wave velocity, and the calculation formula is:

[0083] V s =(0.61~0.53)V p (4)

[0084] Among them, V s is the shear wave velocity, V p is the longitudinal wave velocity.

[0085] The natural gamma logging data is used to estimate the mud content, and the calculation formula is:

[0086]

[0087] Among them, V sh is the volume content of mud, GCUR is the Hillich index, I GR is the mud content index, GR,GRmax min They represent the natural gamma values ​​of the target layer, pure mudstone layer and pure sandstone layer respectively.

[0088] The dynamic Poisson's ratio and dynamic Young's modulus are calculated according to the longitudinal wave velocity and the transverse wave velocity. The calculation formula is:

[0089] E d =ρV s 2 (3V p 2 -4V s 2 ) / (V p 2 -2V s 2 ) (6)

[0090] μ d =(V p 2 -2V s 2 ) / 2(V p 2 -V s 2 ) (7)

[0091] Among them, μ d is the dynamic Poisson's ratio, E d is the dynamic Young's modulus, V p is the longitudinal wave velocity, V s is the shear wave velocity.

[0092] The uniaxial compressive strength is calculated based on Young's modulus and clay content, and the calculation formula is:

[0093] σ c =(0.0045+0.003V sh )E d (8)

[0094] Among them, σ c is the uniaxial compressive strength, that is, the true value label corresponding to the uniaxial compressive strength.

[0095] According to the propagation characteristics of longitudinal and transverse waves in rocks, the cohesive force is calculated by the dynamic Young's elastic modulus, Poisson's ratio, longitudinal wave velocity, and transverse wave velocity. The calculation formula is:

[0096]

[0097] Among them, C is the cohesion, ρ is the rock density, and A is a constant related to rock properties, which is obtained by regressing the measured cohesion with the logging interpretation results.

[0098] Calculation of internal friction angle from cohesion That is, the true value label corresponding to the internal friction angle, and its calculation formula is:

[0099]

[0100] Based on the prediction results and the true value labels corresponding to the input parameters, the loss value is calculated, and then the model parameters of the rock mechanics prediction model are updated. Through different adjustment optimizations, the model with the best performance is obtained; the mean square error, root mean square error and mean absolute error of the model with the best performance are calculated to evaluate the performance of the model.

[0101] In the present invention, learning_rate (learning rate) = 0.3, max_depth (maximum depth of the tree) = 6, n_estimators (number of trees) = 100 are set, and the model with the best performance is obtained.

[0102] A600, cycle A400-A500 until a trained rock mechanics prediction model is obtained.

[0103] In this embodiment, the rock mechanics prediction model is trained cyclically until a trained rock mechanics prediction model is obtained.

[0104] In addition, after training, the present invention evaluates the performance of the model by calculating the mean square error, mean absolute error, mean absolute percentage error and R-square index of the model with the best performance.

[0105] Mean square error (MSE), the difference between the predicted value and the true value (such as Figure 6 The calculation formula is:

[0106]

[0107] Where n represents the number of samples, y i represents the true value of the i-th sample, Represents the predicted value of the i-th sample.

[0108] The mean absolute error (MAE) is obtained by taking the absolute value of the difference between the predicted value and the true value and averaging it. The calculation formula is:

[0109]

[0110] The mean absolute percentage error (MAPE), the relative error between the actual value and the predicted value, is summed and averaged, expressed as a percentage, and the calculation formula is:

[0111]

[0112] R-Squared is an indicator used to evaluate the goodness of fit of a statistical model. It can be used to indicate the degree to which the independent variable explains the dependent variable. The value of R-Squared ranges from 0 to 1, where 0 means that the model cannot explain the changes in the dependent variable, and 1 means that the model perfectly explains the changes in the dependent variable.

[0113] The calculation formula of R-squared is:

[0114]

[0115] Where n represents the number of samples, Y i is the true value of the i-th sample, is the predicted value of the ith sample, is the sample mean.

[0116] The evaluation results are shown in Table 1. The values ​​of MSE, MAE, MAPE and R-square in Table 1 show that, as shown in Table 1, although the error of the test well is larger than that of the training set, the R-square value shows that the goodness of fit is above 70%. Therefore, the LightGBM model has good prediction ability and can well use drilling parameters to reflect the changing trends and laws of rock mechanical parameters.

[0117] Table 1 Dataset MSE R-square MAE MAPE Training set 3.96 0.96 1.37 0.09 Test well 1 23.96 0.77 4 0.25 Test well 2 25.99 0.73 3.95 0.33

[0118] 2. A method for real-time prediction of formation rock mechanical parameters based on machine learning.

[0119] S100, obtaining basic drilling data corresponding to a first well as input data; the first well is a well for which formation rock mechanical parameters are to be predicted; the basic drilling data includes drilling data, mud logging data and well logging data;

[0120] In this embodiment, basic drilling data of the well for which formation rock mechanical parameters are to be predicted (ie, the first well) is first obtained as input data.

[0121] S200, preprocessing the input data to obtain preprocessed data; the preprocessing includes outlier detection, maximum and minimum normalization processing, and smoothing processing;

[0122] In this embodiment, the acquired basic drilling data is preprocessed, and the preprocessing process is specifically described in A200 above.

[0123] S300, calculating the correlation of each parameter in the preprocessed data, and taking the parameters whose correlation is within the set correlation threshold range as input parameters;

[0124] In this embodiment, the correlation of each parameter in the preprocessed data is preferably calculated by the Pearson coefficient. The input parameters include hook load, mechanical drilling speed, pump pressure, drilling pressure, displacement, density, and viscosity;

[0125] S400, inputting the input parameters into a trained rock mechanics prediction model to obtain prediction results of the formation rock mechanics parameters of the first drilling; the rock mechanics prediction model is constructed based on LightGBM.

[0126] In this embodiment, the above-mentioned trained rock mechanics prediction model is used to obtain the prediction results of the formation rock mechanics parameters corresponding to the drilling of the formation rock mechanics parameters to be predicted. The prediction results include uniaxial compressive strength and internal friction angle, and are transmitted through the system to the device used by the user or other devices for display.

[0127] The present invention realizes the real-time prediction of rock mechanical parameters during the drilling process, which is convenient for on-site personnel to timely understand the geological conditions, take early prevention measures for possible accidents, and timely adjust the construction parameters.

[0128] A system for real-time prediction of formation rock mechanical parameters based on machine learning in a second embodiment of the present invention is provided. Figure 2 As shown, the system includes:

[0129] The data acquisition module 100 is configured to acquire basic drilling data corresponding to a first wellbore as input data; the first wellbore is a wellbore for which formation rock mechanical parameters are to be predicted; the basic drilling data includes drilling data, logging data and well logging data;

[0130] A preprocessing module 200 is configured to preprocess the input data to obtain preprocessed data; the preprocessing includes outlier detection, maximum and minimum normalization processing, and smoothing processing;

[0131] A parameter calculation module 300 is configured to calculate the correlation of each parameter in the preprocessed data, and take the parameter whose correlation is within a set correlation threshold range as an input parameter;

[0132] A prediction result acquisition module 400 is configured to input the input parameters into a trained rock mechanics prediction model to obtain a prediction result of the formation rock mechanics parameters of the first wellbore;

[0133] Among them, the rock mechanics prediction model is built based on LightGBM.

[0134] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process and related instructions of the system described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0135] It should be noted that the above embodiment provides a system for real-time prediction of formation rock mechanical parameters based on machine learning, which is only illustrated by the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the modules or steps in the embodiments of the present invention can be decomposed or combined. For example, the modules in the above embodiments can be combined into one module, or further divided into multiple sub-modules to complete all or part of the functions described above. The names of the modules and steps involved in the embodiments of the present invention are only for distinguishing the modules or steps, and are not regarded as improper limitations of the present invention.

[0136] An electronic device according to the third embodiment of the present invention comprises at least one processor; and a memory communicatively connected to at least one of the processors; wherein the memory stores instructions executable by the processor, and the instructions are used to be executed by the processor to implement the above-mentioned method for real-time prediction of formation rock mechanical parameters based on machine learning.

[0137] A computer-readable storage medium according to a fourth embodiment of the present invention stores computer instructions, and the computer instructions are used to be executed by the computer to implement the above-mentioned method for real-time prediction of formation rock mechanical parameters based on machine learning.

[0138] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process and related instructions of the electronic device and readable storage medium described above can refer to the corresponding process in the aforementioned method example and will not be repeated here.

[0139] Reference below Figure 7 , which shows a schematic diagram of the structure of a computer system of a server suitable for implementing the method, system, electronic device, and readable storage medium embodiments of the present application. Figure 7 The server shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0140] like Figure 7As shown, the computer system includes a central processing unit (CPU) 701, which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) 702 or the program loaded from the storage part 708 to the random access memory (RAM) 703. Various programs and data required for system operation are also stored in the RAM 703. The CPU 701, ROM 702 and RAM 703 are connected to each other through a bus 704. An input / output (I / O) interface 705 is also connected to the bus 704.

[0141] The following components are connected to the I / O interface 705: an input section 706 including a keyboard, a mouse, etc.; an output section 707 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 708 including a hard disk, etc.; and a communication section 709 including a network interface card such as a LAN (Local Area Network) card, a modem, etc. The communication section 709 performs communication processing via a network such as the Internet. A drive 710 is also connected to the I / O interface 705 as needed. A removable medium 711, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 710 as needed so that a computer program read therefrom is installed into the storage section 708 as needed.

[0142] In particular, according to an embodiment of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a computer readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through the communication part 709, and / or installed from a removable medium 711. When the computer program is executed by the central processing unit (CPU701), the above functions defined in the method of the present application are executed. It should be noted that the computer-readable medium mentioned in the present application may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application In the present application, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present application, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in combination with an instruction execution system, device or device. The program code contained on the computer-readable medium may be transmitted using any suitable medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0143] Computer program code for performing the operations of the present application may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0144] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present application. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0145] The terms "first", "second", etc. are used to distinguish similar objects rather than to describe or indicate a particular order or sequence.

[0146] The term "comprise" or any other similar term is intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus / device that includes a list of elements includes not only those elements but also other elements not expressly listed, or also includes elements inherent to such process, method, article, or apparatus / device.

[0147] So far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present invention.

Claims

1. A method for real-time prediction of formation rock mechanical parameters based on machine learning, characterized in that: The method includes: S100, obtaining basic drilling data corresponding to a first well as input data; the first well is a well for which formation rock mechanical parameters are to be predicted; the basic drilling data includes drilling data, mud logging data and well logging data; S200, preprocessing the input data to obtain preprocessed data; the preprocessing includes outlier detection, maximum and minimum normalization processing, and smoothing processing; S300, calculating the correlation of each parameter in the preprocessed data, and taking the parameters whose correlation is within a set correlation threshold range as input parameters; S400, inputting the input parameters into a trained rock mechanics prediction model to obtain a prediction result of the formation rock mechanics parameters of the first well; Among them, the rock mechanics prediction model is built based on LightGBM.

2. The method for real-time prediction of formation rock mechanical parameters based on machine learning according to claim 1 is characterized in that: The input parameters include hook load, mechanical drilling speed, pump pressure, drilling pressure, displacement, density, and viscosity; the prediction results include uniaxial compressive strength and internal friction angle.

3. The method for real-time prediction of formation rock mechanical parameters based on machine learning according to claim 2 is characterized in that: The rock mechanics prediction model is trained by: A100, obtaining basic drilling data corresponding to a set number of wells as input data; the basic drilling data includes drilling data, logging data and well logging data; A200, preprocessing the input data to obtain preprocessed data; The preprocessing includes outlier detection, maximum and minimum normalization processing, and smoothing processing; A300, calculating the correlation of each parameter in the preprocessed data, and taking the parameter whose correlation is within the set correlation threshold range as the input parameter; A400, inputting the input parameters into a pre-built rock mechanics prediction model to obtain a prediction result of the formation rock mechanics parameters of the first well; A500, calculating the loss value based on the prediction result and the true value label corresponding to the input parameter, and then updating the model parameters of the rock mechanics prediction model; A600, cycle A400-A500 until a trained rock mechanics prediction model is obtained.

4. The method for real-time prediction of formation rock mechanical parameters based on machine learning according to claim 3 is characterized in that: The correlation of each parameter in the preprocessed data is calculated by: calculating the correlation of each parameter in the preprocessed data by Pearson coefficient: Among them, ρ X,Y It represents the correlation coefficient between the set parameters, that is, the correlation. X and Y represent two feature data, that is, two parameters. cov(X,Y) is the covariance of X and Y. σ X is the standard deviation of X, σ Y is the standard deviation of Y.

5. The method for real-time prediction of formation rock mechanical parameters based on machine learning according to claim 3 is characterized in that: The true value label corresponding to the uniaxial compressive strength is calculated as follows: s c =(0.0045+0.003V sh )E d E d =ρV s 2 (3V p 2 -4V s 2 ) / (V p 2 -2V s 2 ) V s =(0.61~0.53)V p Among them, σ c is the uniaxial compressive strength, E d is the dynamic Young's modulus, V p is the longitudinal wave velocity, V s is the shear wave velocity, V sh is the volume content of mud, ρ is the rock density, GCUR is the Hillch index, I GR is the mud content index, GR, GR min , GR max They represent the natural gamma values ​​of the target layer, pure mudstone layer and pure sandstone layer respectively.

6. The method for real-time prediction of formation rock mechanical parameters based on machine learning according to claim 5 is characterized in that: The internal friction angle and its corresponding true value label are calculated as follows: in, represents the internal friction angle, C represents the cohesion, A is a constant related to the rock properties, μ d represents the dynamic Poisson's ratio.

7. A system for real-time prediction of formation rock mechanical parameters based on machine learning, characterized in that: The system includes: A data acquisition module is configured to acquire basic drilling data corresponding to a first wellbore as input data; the first wellbore is a wellbore for which formation rock mechanical parameters are to be predicted; the basic drilling data includes drilling data, logging data and well logging data; A preprocessing module is configured to preprocess the input data to obtain preprocessed data; the preprocessing includes outlier detection, maximum and minimum normalization processing, and smoothing processing; A parameter calculation module is configured to calculate the correlation of each parameter in the preprocessed data, and take the parameters whose correlation is within a set correlation threshold range as input parameters; A prediction result acquisition module, configured to input the input parameters into a trained rock mechanics prediction model to obtain a prediction result of the formation rock mechanics parameters of the first wellbore; Among them, the rock mechanics prediction model is built based on LightGBM.

8. An electronic device, characterized in that: include: at least one processor; and a memory communicatively coupled to at least one of the processors; In which, the memory stores instructions that can be executed by the processor, and the instructions are used to be executed by the processor to implement the method for real-time prediction of formation rock mechanical parameters based on machine learning as described in any one of claims 1-6.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to be executed by the computer to implement the method for real-time prediction of formation rock mechanical parameters based on machine learning as described in any one of claims 1-6.

Images