Drill bit type selection method, system and equipment based on neural network and medium

Abstract

The invention discloses a drill bit type selection method, system and equipment based on a neural network and a medium. Stratigraphic data are collected, rock samples are obtained through underground coring, and drilling resistance characteristic parameters of the rock samples are obtained; obtaining a common anti-drilling characteristic parameter single well profile of the rock sample and the stratum according to the anti-drilling characteristic parameters of the rock sample and the stratum data; extracting anti-drilling characteristic parameters corresponding to well sections drilled by different types of drill bits in the stratum in combination with the anti-drilling characteristic parameter single well profile; establishing a drill bit type selection neural network model, inputting the anti-drilling characteristic parameters of the stratum and the anti-drilling characteristic parameters of the rock sample into the drill bit type selection neural network model as self-learning samples, and training the drill bit type selection neural network model; and inputting stratum anti-drilling characteristic parameters of a to-be-selected drill bit type well into the trained drill bit type selection neural network model to obtain drill bit types adopted in different depth sections. And high-precision and high-efficiency type selection of full-well-section drill bit types is realized.

Description

Technical Field

[0001] The invention belongs to the field of drilling, and relates to a drill bit selection method, system, equipment and medium based on neural network. Background Art

[0002] Rock drillability evaluation is of great significance for drill bit optimization and improving drilling speed. Conventional rock drillability evaluation is measured under indoor normal pressure, without considering the influence of confining pressure on drillability. In the actual formation drilling process, rock drillability is affected by formation conditions. Reasonable selection of drill bits is a key factor affecting the drilling speed of drill bits. By optimizing and applying drill bits, the mechanical drilling speed can be improved and the cost of drilling construction can be reduced. At the same time, the occurrence of underground accidents can be reduced, ensuring safe drilling and meeting the needs of oil and gas field drilling construction.

[0003] First, conventional rock drillability evaluation is measured at normal indoor pressure, without considering the effect of confining pressure on drillability. In the actual formation drilling process, rock drillability is affected by formation conditions, so it cannot represent the true drillability of rock, which brings difficulties to drill bit selection. In addition, conventional drill bit selection methods generally do not take into account the rock strength parameters of the formation, but only consider the characteristics of formation drillability and abrasiveness and practical drill bits in adjacent wells to select the type of drill bit suitable for the formation. This drill bit selection method is highly subjective and has strong limitations. Therefore, it is difficult to have high efficiency and accuracy in drill bit selection, resulting in waste of manpower and material resources. Summary of the invention

[0004] The purpose of the present invention is to overcome the shortcomings of the above-mentioned prior art and provide a drill bit selection method, system, equipment and medium based on neural network, which realizes high-precision and high-efficiency selection of drill bit types for the entire well section, reduces manpower and material costs, and provides a reference for safe and efficient drilling in the work area.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions:

[0006] A drill bit selection method based on neural network includes the following processes:

[0007] S1, collect formation data, take coring samples downhole to obtain rock samples, and obtain the anti-drilling characteristic parameters of the rock samples;

[0008] S2, obtaining a single well profile of the common anti-drilling characteristic parameters of the rock sample and the formation according to the anti-drilling characteristic parameters of the rock sample and the formation data;

[0009] S3, combining the single well profile of the anti-drilling characteristic parameters, extracting the anti-drilling characteristic parameters corresponding to the well sections drilled by different types of drill bits in the formation;

[0010] S4, establishing a drill bit selection neural network model, taking the anti-drilling characteristic parameters corresponding to the well sections drilled by different types of drill bits in the formation and the anti-drilling characteristic parameters of the rock samples as self-learning samples, inputting them into the drill bit selection neural network model, and training the drill bit selection neural network model;

[0011] S5, inputting the formation anti-drilling characteristic parameters of the well of the drill bit type to be selected into the trained drill bit selection neural network model to obtain the drill bit types used in different depth sections.

[0012] Preferably, the anti-drilling characteristic parameters include natural gamma, sonic transit time, density, shale content, compressive strength, cohesion, internal friction angle, hardness, abrasiveness and drillability grade.

[0013] Preferably, before S3, the drilling and completion reports, drilling logs, well histories and other information of the wells drilled in the work area are used to sort out the drill bit types and models used in the study area and their corresponding information such as footage and drilling speed. Based on the information such as the drill bit type sorted out in the study area, a statistical analysis is performed on the information such as the drilling speed and footage. The drill bit type for the study formation is optimized based on the average drilling speed and average footage of the actual drill bits in the study area.

[0014] Preferably, after S3 is completed, drill bits with the same or similar performance indicators are classified into one category according to the drill bit IADC code, and the classified drill bits are numbered with different training codes to establish a database of the correspondence between formation anti-drilling characteristic parameters and drill bit types.

[0015] Preferably, in S5, the number of network layers and the number of nodes of the drill bit selection neural network model are set as follows: the neural network has a total of 5 layers, including 3 hidden layers, the number of nodes in the network input layer is set to 10, the number of nodes in the first hidden layer is set to 20, the number of nodes in the second hidden layer is set to 11, the number of nodes in the third hidden layer is set to 4, and the number of nodes in the output layer is set to 1; the number of network iterations is set to 100,000 times, the display interval is 50, the learning rate is 0.001, the target accuracy is 0.0001, the activation function 1 is logsig, the activation function 2 is logsig, the activation function 3 is purelin, and the error calculation function is traingdx.

[0016] Preferably, in S5, after obtaining the drill bit type result, it is compared with the drill bit type actually used to determine whether the selection accuracy meets the requirements.

[0017] Preferably, in S5, after obtaining the drill bit type result, if the predicted drill bit type in a certain depth section of the well is inconsistent with the actual use, combined with the drill bit type and its mechanical drilling speed and drilling size of the same layer section that has been drilled, the actual drilling effect is good, indicating that the drill bit type of the original well has good adaptability to the formation; if the actual drilling effect is poor, and the predicted drill bit type has a good drilling effect in the same layer section of the adjacent well, indicating that the drill bit type of the original well has poor adaptability to the formation, then the predicted drill bit type is recommended for this depth section, and on this basis, the drill bit type that is adapted to the formation characteristics of the entire well section is selected, which serves as the basis for selecting drill bits for formations with similar lithology in the same formation group in new drilling design.

[0018] A drill bit selection system based on neural network, comprising:

[0019] The rock sample parameter module is used to collect formation data, obtain rock samples by coring downhole, and obtain the anti-drilling characteristic parameters of rock samples;

[0020] The single well profile module is used to obtain the single well profile of the common anti-drilling characteristic parameters of the rock sample and the formation according to the anti-drilling characteristic parameters of the rock sample and the formation data;

[0021] The formation parameter module is used to extract the anti-drilling characteristic parameters corresponding to the well sections drilled by different types of drill bits in the formation in combination with the single well profile of the anti-drilling characteristic parameters;

[0022] A model module is used to establish a drill bit selection neural network model, and the anti-drilling characteristic parameters corresponding to the well sections drilled by different types of drill bits in the formation and the anti-drilling characteristic parameters of the rock samples are used as self-learning samples, which are input into the drill bit selection neural network model to train the drill bit selection neural network model;

[0023] The selection module is used to input the formation anti-drilling characteristic parameters of the well with the drill bit type to be selected into the trained drill bit selection neural network model to obtain the drill bit type used in different depth sections.

[0024] A computer device comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the drill bit selection method based on a neural network when executing the computer program.

[0025] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the drill bit selection method based on a neural network are implemented.

[0026] Compared with the prior art, the present invention has the following beneficial effects:

[0027] The present invention combines the anti-drilling characteristic parameters of the rock sample with the anti-drilling characteristic parameters corresponding to the well sections drilled by different types of drill bits in the formation, and analyzes the relationship between the drill bit type and the anti-drilling characteristic parameters of the rock in the drilled formation during the drilling process through a drill bit selection neural network model combined with the actual use of the drill bit in the drilled well, thereby realizing the selection of drill bit types for the entire well section, reducing manpower and material costs, and providing a reference for safe and efficient drilling in the work area. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 It is a flow chart of the drill bit selection method based on neural network of the present invention;

[0029] Figure 2 It is a schematic diagram of the neural network structure of the present invention;

[0030] Figure 3 It is a cross-sectional schematic diagram of the anti-drilling characteristics of a certain well of the present invention;

[0031] Figure 4 This is a schematic diagram of the neural network drill bit selection and application of the present invention. DETAILED DESCRIPTION

[0032] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments; based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0033] It should be noted that the words "front", "rear", "left", "right", "up" and "down" used in the following description refer to directions in the drawings, and the words "inside" and "outside" refer to directions toward or away from the geometric center of a specific component, respectively.

[0034] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by those skilled in the art of the present invention. The terms used herein in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more related listed items.

[0035] like Figure 1 As shown, the drill bit selection method based on neural network provided by the present invention comprises:

[0036] (1) Collect geological, well logging, well logging, drilling and completion data of the actual strata in the study area and rock samples obtained by downhole coring, hereinafter referred to as rock samples.

[0037] (2) According to the relevant provisions and requirements of SY / T 5336-1996 "Routine Analysis Methods for Rock Cores" and GB / T50266-99 "Standards for Engineering Rock Mass Test Methods", the core samples were made into cylindrical rock specimens with a diameter of 2.5 cm and a height of 5.0 cm. The errors of the specimen height and diameter were not more than 0.3 mm; the non-parallelism of the two end faces of the specimen was not more than 0.05 mm; the end face of the specimen was perpendicular to the specimen axis, and the maximum deviation was not more than 0.25°.

[0038] (3) According to the relevant requirements of SY / T 5336-1996 "Routine analysis method of rock core" and SY / T 6351-1998 "Laboratory determination of acoustic wave properties of rock", the rock samples were subjected to indoor acoustic wave testing and basic physical property testing to obtain the acoustic wave time difference (AC) and density (DEN) of the rock samples;

[0039] (4) The rock samples were subjected to triaxial compression test, hardness test, abrasiveness test, and drillability test. The drillability test took the confining pressure into consideration and obtained the uniaxial compressive strength (UCS), cohesion (C), and internal friction angle of the rock samples. Hardness (H d ), abrasiveness index (G d ), drillability grade (K d ), so as to obtain the anti-drilling characteristic parameters of the rock sample.

[0040] (5) Based on rock physics and rock mechanics, the rock physical response mechanism of the anti-drilling characteristic parameters of rock samples was studied, and a well logging prediction calculation model for studying the anti-drilling characteristic parameters of formations was established. Combined with the geological, logging, logging, drilling and completion data in the actual formation and the anti-drilling characteristic parameters of rock samples, a single well profile of the anti-drilling characteristic parameters of the formation and rock samples can be further obtained.

[0041] (6) Based on the drilling and completion reports, drilling logs, well histories, and other information of the wells that have been drilled in the work area, the types and models of drill bits used in the work area of ​​the study area and their corresponding footage and drilling speed are compiled.

[0042] (7) Based on the information on drill bit types obtained from the study area, the drilling speed and footage information were statistically analyzed, and the drill bit type for the study formation was selected based on the average drilling speed and average footage of the actual drill bits in the study area. Combined with the formation data in the single well profile of the anti-drilling characteristic parameters, the corresponding information such as natural gamma, acoustic wave time difference, density, mud content, compressive strength, cohesion, internal friction angle, hardness, abrasiveness and drillability grade corresponding to the well section drilled by each type of drill bit in the formation was extracted, that is, the anti-drilling characteristic parameters of the formation, and then combined with the anti-drilling characteristic parameters of the rock sample, so as to construct a sample database of the optimal drill bits in the study area.

[0043] (8) According to the drill bit IADC code, the drill bits with the same or similar performance indicators in the above (7) are classified into one category, and the drill bits of different types are numbered with different training codes. The anti-drilling characteristic parameters corresponding to the well sections drilled by different types of drill bits in the formation and the anti-drilling characteristic parameters of the rock samples are combined as the anti-drilling parameter table for drill bit selection (i.e., the database of the corresponding relationship between the formation anti-drilling characteristic parameters and the drill bit types), and this is used as the neural network input parameter, i.e., the self-learning sample.

[0044] (9) Construct a neural network model for drill bit selection. The specific network structure is as follows: Figure 2 As shown in the figure, the number of network layers and nodes are set as follows according to the actual input parameters: the neural network has 5 layers, including 3 hidden layers, the number of nodes in the network input layer is set to 10, the number of nodes in the first hidden layer is set to 20, the number of nodes in the second hidden layer is set to 11, the number of nodes in the third hidden layer is set to 4, and the number of nodes in the output layer is set to 1. The number of network iterations is set to 100,000 times, the display interval is set to 50, the learning rate is set to 0.001, the target accuracy is set to 0.0001, the activation function 1 is set to logsig, the activation function 2 is set to logsig, the activation function 3 is set to purelin, and the error calculation function is set to traingdx.

[0045] (10) The self-learning samples are input to complete the network training. After the network continuously iterates to reach the specified accuracy, the connection weights of the nodes between each layer and the thresholds of each layer are obtained. The combination of the weights and thresholds between the nodes in each layer is the drill selection model.

[0046] (11) Input the formation anti-drilling characteristic parameters of the well with the drill bit type to be predicted (such as natural gamma, acoustic wave time difference, density, mud content, compressive strength, cohesion, internal friction angle, hardness, abrasiveness and drillability grade) into the trained network, obtain the drill bit type predicted by the neural network, compare it with the actual drill bit type, make an accuracy judgment, and determine whether the selection accuracy meets the requirements.

[0047] (12) The drill bit selection prediction method based on neural network can predict the drill bit type of the whole well section with the optimal sample database of the drilled wells, so as to judge the adaptability of the drill bit type to the drilled formation. If the predicted drill bit type of a certain depth section of the well is inconsistent with the actual use, combined with the drill bit type and its mechanical drilling speed and drilling size of the same layer section of the drilled well, the actual drilling effect is good, indicating that the drill bit type of the original well is well adapted to the formation; if the actual drilling effect is poor, and the predicted drill bit type has a good drilling effect in the same layer section of the adjacent well, indicating that the drill bit type of the original well is poorly adapted to the formation, then the predicted drill bit type is recommended for this depth section. On this basis, the drill bit type that is adapted to the formation characteristics of the whole well section is selected, which can be used as the basis for the selection of drill bits for the same formation with similar lithology in the new drilling design, so as to improve the drilling mechanical speed and average footage. With the formation parameters required for prediction as input, the trained neural network model can predict the drill bit type for the entire well section and compare it with the actual drill bit type used. Combined with the formation's anti-drilling characteristics and drilling effect, and considering the drilling effect of the same drill bit type in adjacent wells, the drill bit type suitable for the formation can be selected, which will provide support for the optimization of drill bits for new wells adjacent to the work area.

[0048] The following is an explanation based on actual cases:

[0049] 1) Collect geological, well logging, well logging, drilling and completion data of the study area and rock samples obtained by downhole coring;

[0050] (2) According to the relevant provisions and requirements of SY / T 5336-1996 "Routine Analysis Methods for Rock Cores" and GB / T50266-99 "Standards for Engineering Rock Mass Test Methods", the core samples are made into cylindrical rock specimens with a diameter of 2.5 cm and a height of 5.0 cm. The errors of the specimen height and diameter are both within 0.3 mm; the non-parallelism of the two end faces of the specimen is no more than 0.05 mm; the end face of the specimen is perpendicular to the specimen axis, and the maximum deviation is no more than 0.25°;

[0051] (3) According to the relevant requirements of SY / T 5336-1996 "Routine analysis method of rock core" and SY / T 6351-1998 "Laboratory determination of acoustic wave properties of rock", the rock samples were subjected to indoor acoustic wave testing and basic physical property testing to obtain the acoustic wave time difference (AC) and density (DEN) of the rock samples;

[0052] (4) The rock samples were subjected to triaxial compression test, hardness test, abrasiveness test, and drillability test. The drillability test took the confining pressure into consideration and obtained the uniaxial compressive strength (UCS), cohesion (C), and internal friction angle of the rock samples. Hardness (H d ), abrasiveness index (G d ), drillability grade (Kd ), so as to obtain the anti-drilling characteristic parameters of the rock sample;

[0053] (5) Based on rock physics and rock mechanics, the rock physical response mechanism of rock sample anti-drilling characteristic parameters is studied, and a well logging prediction calculation model for studying formation drilling characteristic parameters is established (as shown in Table 1). Combined with the logging data and mud logging data of the drilled wells, the single well profile of the drilling characteristic parameters can be further obtained, such as Figure 3 As shown. Figure 3 The anti-drilling characteristic parameters of a certain formation can be proposed, and it is also found that the lithology of the formation changes frequently and the distribution regularity of the anti-drilling characteristic parameters is poor, which brings difficulties to the optimization of the drill bit type for the formation.

[0054] Table 1 Calculation model of drilling characteristic parameters of each formation

[0055]

[0056] (6) Based on the drilling and completion reports, drilling logs, well histories and other information of the wells drilled in the work area, the types and models of drill bits used in the work area of ​​the study area and their corresponding footage and drilling speed are sorted out;

[0057] (7) Based on the information on drill bit types obtained from the study area, the drilling speed and footage information were statistically analyzed, and the drill bit type for the study formation was selected based on the average drilling speed and average footage of the actual drill bits in the study area. Combined with the formation data in the single well profile of the anti-drilling characteristic parameters, the corresponding information such as natural gamma, acoustic wave time difference, density, mud content, compressive strength, cohesion, internal friction angle, hardness, abrasiveness and drillability grade corresponding to the well section drilled by each type of drill bit in the formation was extracted, that is, the anti-drilling characteristic parameters of the formation, and then combined with the anti-drilling characteristic parameters of the rock sample, so as to construct a sample database of the optimal drill bits in the study area.

[0058] (8) According to the drill bit IADC code, the drill bits with the same or similar performance indicators in the above (7) are classified into one category, and the drill bits of different types are numbered with different training codes as shown in Table 2. The anti-drilling characteristic parameters corresponding to the well sections drilled by different types of drill bits in the formation and the anti-drilling characteristic parameters of the rock samples are used as the anti-drilling parameter table for drill bit selection (i.e., the database of the corresponding relationship between formation parameters and drill bit types), and this is used as the neural network input parameter, i.e., the self-learning sample as shown in Table 3.

[0059] Table 2 Neural network training codes for various drill bit IADC codes

[0060]

[0061] Table 3 Self-learning samples (part)

[0062]

[0063]

[0064] (9) Construct a drill bit selection neural network model. The number of network layers and nodes are set as follows according to the actual input parameters: the neural network has 5 layers, including 3 hidden layers. The number of nodes in the network input layer is set to 10, the number of nodes in the first hidden layer is set to 20, the number of nodes in the second hidden layer is set to 11, the number of nodes in the third hidden layer is set to 4, and the number of nodes in the output layer is set to 1. Set the number of network iterations to 100,000 times, the display interval to 50, the learning rate to 0.001, the target accuracy to 0.0001, the activation function 1 to logsig, the activation function 2 to logsig, the activation function 3 to purelin, and the error calculation function to traingdx;

[0065] (10) Input the self-learning samples and complete the network training. After the network continuously iterates and runs to reach the specified accuracy, the connection weights of the nodes between each layer and the thresholds of each layer are obtained. The combination of the weights and thresholds between the nodes of each layer is the drill bit selection model as shown in Tables 4 to 7.

[0066] Table 4 Input layer-hidden layer 1 weights and thresholds

[0067]

[0068] Table 5 Weights and thresholds of hidden layer 1-hidden layer 2

[0069]

[0070]

[0071] Table 6 Weights and thresholds of hidden layer 2 to hidden layer 3

[0072]

[0073]

[0074] Table 7 Hidden layer 3 - output layer weights and thresholds

[0075]

[0076] (11) The formation parameters of the well whose drill bit type is to be predicted (such as natural gamma, acoustic time difference, density, mud content, compressive strength, cohesion, internal friction angle, hardness, abrasiveness and drillability grade) are input into the trained network to obtain the drill bit type predicted by the neural network. The comparison with the actual drill bit type is shown in Table 8. The predicted drill bit type is consistent with the actual drill bit type, and the prediction accuracy reaches 90%, which illustrates the feasibility of this method in predicting the drill bit type.

[0077] Table 8 Neural network back-judgment result verification table

[0078]

[0079]

[0080] (12) The drill bit selection prediction method based on neural network can predict the drill bit type of the whole well section with the preferred sample database of the drilled wells, so as to judge the adaptability of the drill bit type to the drilled formation. If the predicted drill bit type of a certain depth section of the well is inconsistent with the actual use, combined with the drill bit type and its mechanical drilling speed and drilling size of the same layer section of the well, the latter has a good drilling effect, indicating that the drill bit type of the original well has better adaptability to the formation, while the latter has a poor drilling effect, and the predicted drill bit type has a good drilling effect in the same layer section of the adjacent well, indicating that the drill bit type of the original well has poor adaptability to the formation, then the predicted drill bit type is recommended for this depth section. On this basis, the drill bit type that is suitable for the formation characteristics of the whole well section is selected, which can be used as the basis for the selection of drill bits for the same formation with similar lithology in the new drilling design, so as to improve the drilling mechanical speed and average footage. With the formation parameters required for prediction as input, the drill bit type prediction of the whole well section can be realized through the trained neural network model, such as Figure 4 As shown, by comparing with the actual drill bit type in use, combining the anti-drilling characteristics and drilling effect of the formation, considering the drilling effect of the same drill bit type in adjacent wells, the drill bit type suitable for the formation is selected, which will provide support for the optimization of drill bits for new wells adjacent to the work area.

[0081] The following are device embodiments of the present invention, which can be used to perform method embodiments of the present invention. For details not disclosed in the device embodiments, please refer to the method embodiments of the present invention.

[0082] In another embodiment of the present invention, a neural network-based drill bit selection system is provided. The neural network-based drill bit selection system can be used to implement the above-mentioned neural network-based drill bit selection method. Specifically, the neural network-based drill bit selection system includes a rock sample parameter module, a single well profile module, a single well profile module, a model module and a selection module.

[0083] Among them, the rock sample parameter module is used to collect formation data, obtain rock samples by coring downhole, and obtain the anti-drilling characteristic parameters of rock samples.

[0084] The single well profile module is used to obtain the single well profile of the common anti-drilling characteristic parameters of the rock sample and the formation according to the anti-drilling characteristic parameters of the rock sample and the formation data.

[0085] The formation parameter module is used to extract the anti-drilling characteristic parameters corresponding to the well sections drilled by different types of drill bits in the formation, combined with the single well profile of the anti-drilling characteristic parameters.

[0086] The model module is used to establish a drill bit selection neural network model. The anti-drilling characteristic parameters corresponding to the well sections drilled by different types of drill bits in the formation and the anti-drilling characteristic parameters of the rock samples are used as self-learning samples and input into the drill bit selection neural network model to train the drill bit selection neural network model.

[0087] The selection module is used to input the formation anti-drilling characteristic parameters of the well with the drill bit type to be selected into the trained drill bit selection neural network model to obtain the drill bit type used in different depth sections.

[0088] In another embodiment of the present invention, a terminal device is provided, which includes a processor and a memory, wherein the memory is used to store a computer program, the computer program includes program instructions, and the processor is used to execute the program instructions stored in the computer storage medium. The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. It is the computing core and control core of the terminal, which is suitable for implementing one or more instructions, and is specifically suitable for loading and executing one or more instructions to implement the corresponding method flow or corresponding function; the processor described in the embodiment of the present invention can be used for the operation of the drill bit selection method based on the neural network, including: S1, collecting formation data, coring downhole to obtain rock samples, and obtaining the anti-drilling characteristic parameters of the rock samples; S2, obtaining according to the anti-drilling characteristic parameters of the rock samples and the formation data A single well profile of the common anti-drilling characteristic parameters of the rock sample and the formation; S3, combining the single well profile of the anti-drilling characteristic parameters, extracting the anti-drilling characteristic parameters corresponding to the well sections drilled by different types of drill bits in the formation; S4, establishing a drill bit selection neural network model, taking the anti-drilling characteristic parameters corresponding to the well sections drilled by different types of drill bits in the formation and the anti-drilling characteristic parameters of the rock sample as self-learning samples, inputting them into the drill bit selection neural network model, and training the drill bit selection neural network model; S5, inputting the formation anti-drilling characteristic parameters of the well of the drill bit type to be selected into the trained drill bit selection neural network model, and obtaining the drill bit types used in different depth sections.

[0089] In another embodiment, the present invention further provides a computer-readable storage medium (Memory), which is a memory device in a terminal device for storing programs and data. It is understandable that the computer-readable storage medium here can include both the built-in storage medium in the terminal device and the extended storage medium supported by the terminal device. The computer-readable storage medium provides a storage space, which stores the operating system of the terminal. In addition, one or more instructions suitable for being loaded and executed by the processor are also stored in the storage space, and these instructions can be one or more computer programs (including program codes). It should be noted that the computer-readable storage medium here can be a high-speed RAM memory or a non-volatile memory, such as at least one disk memory.

[0090] The processor can load and execute one or more instructions stored in the computer-readable storage medium to implement the corresponding steps of the drill bit selection method based on the neural network in the above-mentioned embodiment; one or more instructions in the computer-readable storage medium are loaded by the processor and the following steps are executed: S1, collecting formation data, coring downhole to obtain rock samples, and obtaining the anti-drilling characteristic parameters of the rock samples; S2, according to the anti-drilling characteristic parameters of the rock samples and the formation data, obtaining a single well profile of the common anti-drilling characteristic parameters of the rock samples and the formation; S3, combining the anti-drilling characteristic parameters of the rock samples and the formation data; Characteristic parameter single well profile, extract the anti-drilling characteristic parameters corresponding to the well sections drilled by different types of drill bits in the formation; S4, establish a drill bit selection neural network model, take the anti-drilling characteristic parameters corresponding to the well sections drilled by different types of drill bits in the formation and the anti-drilling characteristic parameters of the rock samples as self-learning samples, input them into the drill bit selection neural network model, and train the drill bit selection neural network model; S5, input the formation anti-drilling characteristic parameters of the well of the drill bit type to be selected into the trained drill bit selection neural network model, and obtain the drill bit types used in different depth sections.

[0091] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0092] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0093] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0094] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0095] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0096] It should be understood that the above description is for illustration and not for limitation. Many embodiments and many applications beyond the examples provided will be apparent to those skilled in the art upon reading the above description. Therefore, the scope of the present teachings should not be determined with reference to the above description, but rather with reference to the foregoing claims and the full scope of equivalents to which such claims are entitled. For the purpose of comprehensiveness, all articles and references, including disclosures of patent applications and publications, are incorporated herein by reference. The omission of any aspect of the subject matter disclosed herein in the foregoing claims is not intended to be a waiver of such subject matter, nor should it be considered that the applicant has not considered such subject matter to be part of the disclosed inventive subject matter.

Claims

1. A drill bit selection method based on neural network, characterized in that: The process includes: S1, collect formation data, take coring samples downhole to obtain rock samples, and obtain the anti-drilling characteristic parameters of the rock samples; S2, obtaining a single well profile of the common anti-drilling characteristic parameters of the rock sample and the formation according to the anti-drilling characteristic parameters of the rock sample and the formation data; S3, combining the single well profile of the anti-drilling characteristic parameters, extracting the anti-drilling characteristic parameters corresponding to the well sections drilled by different types of drill bits in the formation; S4, establishing a drill bit selection neural network model, taking the anti-drilling characteristic parameters corresponding to the well sections drilled by different types of drill bits in the formation and the anti-drilling characteristic parameters of the rock samples as self-learning samples, inputting them into the drill bit selection neural network model, and training the drill bit selection neural network model; S5, inputting the formation anti-drilling characteristic parameters of the well of the drill bit type to be selected into the trained drill bit selection neural network model to obtain the drill bit types used in different depth sections.

2. The drill bit selection method based on neural network according to claim 1, characterized in that: The anti-drilling characteristic parameters include natural gamma, sonic time difference, density, mud content, compressive strength, cohesion, internal friction angle, hardness, abrasiveness and drillability grade.

3. The drill bit selection method based on neural network according to claim 1, characterized in that: Before S3, the drilling and completion reports, drilling logs, well histories and other information of the wells drilled in the work area were used to sort out the types and models of drill bits used in the study area and their corresponding footage and drilling speed. Based on the information such as drill bit types sorted out in the study area, a statistical analysis was conducted on the information such as drilling speed and footage. The drill bit type for the study formation was selected based on the average drilling speed and average footage of the actual drill bits in the study area.

4. The drill bit selection method based on neural network according to claim 1, characterized in that: After S3 is completed, drill bits with the same or similar performance indicators are classified into one category according to the drill bit IADC code, and the classified drill bits are numbered with different training codes to establish a database of the corresponding relationship between formation anti-drilling characteristic parameters and drill bit types.

5. The drill bit selection method based on neural network according to claim 1, characterized in that: In S5, the number of network layers and nodes of the drill bit selection neural network model are set as follows: the neural network has a total of 5 layers, including 3 hidden layers. The number of nodes in the network input layer is set to 10, the number of nodes in the first hidden layer is set to 20, the number of nodes in the second hidden layer is set to 11, the number of nodes in the third hidden layer is set to 4, and the number of nodes in the output layer is set to 1; the number of network iterations is set to 100,000 times, the display interval is set to 50, the learning rate is set to 0.001, the target accuracy is set to 0.0001, the activation function 1 is set to logsig, the activation function 2 is set to logsig, the activation function 3 is set to purelin, and the error calculation function is set to traingdx.

6. The drill bit selection method based on neural network according to claim 1, characterized in that: In S5, after obtaining the drill type result, it is compared with the actual drill type used to determine whether the selection accuracy meets the requirements.

7. The drill bit selection method based on neural network according to claim 1, characterized in that: In S5, after obtaining the drill bit type result, if the predicted drill bit type in a certain depth section of the well is inconsistent with the actual use, combined with the drill bit type and its mechanical drilling speed and drilling size of the same layer section that has been drilled, the actual drilling effect is good, indicating that the drill bit type of the original well has good adaptability to the formation; if the actual drilling effect is poor, and the predicted drill bit type has a good drilling effect in the same layer section of the adjacent well, indicating that the drill bit type of the original well has poor adaptability to the formation, then the predicted drill bit type is recommended for this depth section. On this basis, the drill bit type that is adapted to the formation characteristics of the entire well section is selected, which is used as the basis for selecting drill bits for formations with similar lithology in the same formation group in new drilling design.

8. A drill bit selection system based on neural network, characterized in that: include: The rock sample parameter module is used to collect formation data, obtain rock samples by coring downhole, and obtain the anti-drilling characteristic parameters of rock samples; The single well profile module is used to obtain the single well profile of the common anti-drilling characteristic parameters of the rock sample and the formation according to the anti-drilling characteristic parameters of the rock sample and the formation data; The formation parameter module is used to extract the anti-drilling characteristic parameters corresponding to the well sections drilled by different types of drill bits in the formation in combination with the single well profile of the anti-drilling characteristic parameters; A model module is used to establish a drill bit selection neural network model, and the anti-drilling characteristic parameters corresponding to the well sections drilled by different types of drill bits in the formation and the anti-drilling characteristic parameters of the rock samples are used as self-learning samples, which are input into the drill bit selection neural network model to train the drill bit selection neural network model; The selection module is used to input the formation anti-drilling characteristic parameters of the well with the drill bit type to be selected into the trained drill bit selection neural network model to obtain the drill bit type used in different depth sections.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the drill bit selection method based on neural network as described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the drill bit selection method based on a neural network as described in any one of claims 1 to 7 are implemented.

Images