Multi-well big data-guided drilling parameter optimization methods, devices, and storage media

Abstract

This application provides a method, apparatus, and storage medium for optimizing drilling parameters guided by multi-well big data, belonging to the field of oil and gas exploration technology. The method includes: acquiring prior information about drilled wells in a target area; acquiring real-time drilling data of the target well during the drilling process; determining geological unit division criteria based on the prior information; identifying the real-time geological unit corresponding to the target well at the current moment using a fuzzy C-means clustering algorithm based on the real-time drilling data, prior information, and geological unit division criteria; determining multiple sets of candidate drilling parameters from the real-time drilling data and historical logging data matched with the real-time geological units; determining the mechanical drilling rate and mechanical specific energy corresponding to each set of candidate drilling parameters using a mechanical drilling rate prediction model and a mechanical specific energy calculation formula; and using a multi-objective optimization algorithm, with the objectives of maximizing the mechanical drilling rate and minimizing the mechanical specific energy, selecting and executing a target drilling parameter combination from the multiple sets of candidate drilling parameters.

Description

Technical Field

[0001] This application relates to the field of oil and gas exploration technology, specifically to a method, equipment, and storage medium for optimizing drilling parameters guided by multi-well big data. Background Technology

[0002] In the field of oil and gas exploration technology, drilling parameter optimization is crucial for improving drilling efficiency and safety. However, traditional methods mainly rely on manual experience or statistical analysis of big data, lacking effective utilization of data from adjacent drilled wells. This fails to achieve real-time adaptive recommendations based on common patterns across multiple wells, resulting in insufficient adaptability of drilling parameters to formation changes and poor robustness under complex geological conditions. Summary of the Invention

[0003] The purpose of this application is to provide a method, apparatus, and storage medium for optimizing drilling parameters guided by multi-well big data.

[0004] To achieve the above objectives, the first aspect of this application provides a method for optimizing drilling parameters guided by multi-well big data, comprising:

[0005] Obtain prior information on drilled wells in the target area, where the target area is the area where the target well is located, and the prior information includes at least historical logging data, historical well logging data, historical formation stratification data, and historical drill bit operating information of the drilled wells; Acquire real-time drilling data of the target well during the drilling process, including real-time logging data and real-time drill bit condition information; Determine the criteria for dividing geological units based on prior information; Based on real-time drilling data, prior information, and geological unit division criteria, a fuzzy C-means clustering algorithm is used to identify the real-time geological unit corresponding to the target well at the current moment. Multiple sets of candidate drilling parameters were identified from real-time drilling data and historical logging data matched with real-time geological units; The mechanical drilling rate and mechanical specific energy corresponding to each group of candidate drilling parameters are determined by the mechanical drilling rate prediction model and the mechanical specific energy calculation formula, respectively. Using a multi-objective optimization algorithm, with the objectives of maximizing mechanical drilling rate and minimizing mechanical specific energy, a target drilling parameter combination is selected from multiple sets of candidate drilling parameters and then executed.

[0006] In this embodiment, the method of identifying the real-time geological unit corresponding to the target well at the current moment using the fuzzy C-means clustering algorithm includes: constructing a multi-dimensional state feature matrix based on real-time logging data and real-time drill bit condition information; initializing multiple cluster centers according to prior information and geological unit division criteria, and initializing the multi-dimensional state feature matrix and the membership matrix of each cluster center; updating the cluster centers based on the membership matrix; updating the membership matrix based on the updated cluster centers; determining the objective function based on the updated cluster centers and the updated membership matrix; repeating the update steps for the cluster centers and the membership matrix until the change in the objective function is less than a preset threshold, and outputting the updated membership matrix; selecting the cluster center corresponding to the largest membership degree in the membership matrix as the real-time geological unit.

[0007] In this embodiment, the objective function is determined according to the following formula:

[0008]

[0009] in, Let be the objective function. Real-time drilling data at time i For cluster centers membership degree m For fuzzy coefficients, For real-time drilling data With cluster center Theoretical weighted distance, The prior weighting factor is calculated based on the formation resistivity and natural gamma. For real-time drilling data The One data point, Cluster center The Data.

[0010] In this embodiment, the membership matrix and cluster centers are updated according to the following formula:

[0011]

[0012] in, Real-time drilling data at time i For cluster centers membership degree For real-time drilling data With cluster center Theoretical weighted distance, For real-time drilling data With cluster center Theoretical weighted distance, m is the fuzzy coefficient.

[0013] In this embodiment of the application, the method further includes, before selecting the target drilling parameter combination from candidate drilling parameters by using a multi-objective optimization algorithm based on the mechanical drilling rate prediction model and the mechanical specific energy calculation formula, with the goal of maximizing the mechanical drilling rate of the drill bit and minimizing the real-time mechanical specific energy, a mechanical drilling rate prediction model is constructed and trained. The mechanical drilling rate prediction model is a multi-level graph neural network, including a first drilling parameter correlation graph within the same geological unit and a second drilling parameter correlation graph between different geological units.

[0014] In this embodiment of the application, the training loss function of the mechanical drilling rate prediction model is:

[0015] in, For sample-weighted loss, n For the sample size, The mean absolute percentage error, To control The hyperparameters of the loss weights, For mean square error loss, For the first i The true value of each sample; For the first i The model prediction value for each sample. This is the training loss function.

[0016] In this embodiment of the application, the method further includes: inputting the target drilling parameter combination into the mechanical drilling rate prediction model to obtain the predicted mechanical drilling rate corresponding to the target drilling parameter combination; calculating the relative error between the predicted mechanical drilling rate and the actual mechanical drilling rate corresponding to the execution of the target drilling parameter combination; and updating the mechanical drilling rate prediction model based on all real-time drilling data obtained during the drilling process of the target well if the relative error is greater than a preset threshold.

[0017] In this embodiment of the application, the formula for calculating mechanical specific energy is:

[0018] in, For mechanical specific energy, For drilling pressure, The diameter of the drill bit. The rotational speed of the turntable. For torque, This refers to the mechanical drilling speed.

[0019] The second aspect of this application provides a drilling parameter optimization device guided by multi-well big data, comprising: The memory is configured to store instructions; The processor is configured to retrieve the instructions from the memory and, when executing the instructions, to implement the aforementioned multi-well big data-guided drilling parameter optimization method.

[0020] A third aspect of this application provides a machine-readable storage medium storing instructions that, when executed by a processor, configure the processor to perform the aforementioned multi-well big data-guided drilling parameter optimization method.

[0021] The above technical solution acquires prior information about adjacent wells drilled to the target well, including historical logging data, historical well logging data, historical formation stratification data, and historical drill bit operating information. Based on this prior information, a geological unit division standard is determined. Then, based on real-time drilling data, prior information, and the geological unit division standard, a fuzzy C-means clustering algorithm is used to identify the real-time geological unit corresponding to the target well at the current moment. Next, multiple sets of candidate drilling parameters are determined from the real-time drilling data and historical logging data matching the real-time geological unit. The mechanical drilling rate (MRR) and mechanical specific energy (MSE) corresponding to each set of candidate drilling parameters are determined using a mechanical drilling rate prediction model and a mechanical specific energy calculation formula. Finally, a multi-objective optimization algorithm is used to maximize the MRR and minimize the MSE to select the target drilling parameter from the multiple sets of candidate drilling parameters. This effectively utilizes the common patterns of multi-well data, enhancing the adaptability and robustness of drilling parameter optimization to unseen formations or parameter combinations.

[0022] Other features and advantages of the embodiments of this application will be described in detail in the following detailed description section. Attached Figure Description

[0023] The accompanying drawings are provided to further illustrate the embodiments of this application and form part of the specification. They are used together with the following detailed description to explain the embodiments of this application, but do not constitute a limitation on the embodiments of this application. In the drawings: Figure 1 The illustration shows a flowchart of a multi-well big data-guided drilling parameter optimization method according to an embodiment of this application; Figure 2 This illustration shows a schematic diagram of the geological unit division results guided by prior information according to an embodiment of this application; Figure 3 The schematic diagram illustrates the structure of a mechanical drilling rate prediction model according to an embodiment of this application; Figure 4 A schematic diagram illustrating the mechanical drilling rate prediction results according to an embodiment of this application is shown.

[0024] Figure 5 The illustration shows a flowchart of a multi-well big data-guided drilling parameter optimization method that can utilize real-time incremental learning of data, according to an embodiment of this application. Figure 6 The diagram illustrates the internal structure of a computer device according to an embodiment of this application. Detailed Implementation

[0025] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are only for illustration and explanation of the embodiments of this application and are not intended to limit the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0026] Figure 1 The illustration shows a schematic flowchart of a multi-well big data-guided drilling parameter optimization method according to an embodiment of this application. Figure 1 As shown in one embodiment of this application, a drilling parameter optimization method guided by multi-well big data is provided, including the following steps: Step 101: Obtain prior information of drilled wells in the target area. The target area is the area where the target well is located. The prior information includes at least historical logging data, historical well logging data, historical formation stratification data, and historical drill bit operating information of drilled wells.

[0027] Step 102: Obtain real-time drilling data of the target well during the drilling process. The real-time drilling data includes real-time logging data and real-time drill bit condition information.

[0028] Step 103: Determine the criteria for dividing geological units based on prior information.

[0029] Step 104: Based on real-time drilling data, prior information, and geological unit division criteria, the fuzzy C-means clustering algorithm is used to identify the real-time geological unit corresponding to the target well at the current moment.

[0030] Step 105: Determine multiple sets of candidate drilling parameters from real-time drilling data and historical logging data matched with real-time geological units.

[0031] Step 106: Determine the mechanical drilling rate and mechanical specific energy corresponding to each group of candidate drilling parameters using the mechanical drilling rate prediction model and the mechanical specific energy calculation formula.

[0032] Step 107: Using a multi-objective optimization algorithm, with the objectives of maximizing mechanical drilling rate and minimizing mechanical specific energy, the target drilling parameter combination is selected from multiple sets of candidate drilling parameters and executed.

[0033] The prior information refers to historical logging data, historical formation data, and historical drill bit condition information of adjacent wells already drilled in the target well area. The logging data includes at least: well depth, drilling pressure, torque, rotary table speed, standpipe pressure, inlet flow rate, inlet density, hook load, equivalent density, outlet conductivity, outlet temperature, and mechanical drilling rate. The logging data includes at least: natural gamma ray, resistivity, and sonic transit time. The formation data includes at least: formation position, top depth, bottom depth, top vertical depth, bottom vertical depth, and formation depth. The drill bit condition information includes at least: drill bit number, drill bit size, drill bit model, depth into the well, depth out of the well, and footage. The processor can acquire the prior information of adjacent wells already drilled and the real-time drilling data of the target well during the drilling process. The actual drilling data includes real-time logging data and real-time drill bit condition information. Specifically, in one embodiment, after acquiring prior information and implementing drilling data, outlier and noisy data in the prior information are removed using methods such as 3σ outlier removal, linear regression interpolation, and sliding filtering. Then, the prior information is normalized using the max-min normalization method to eliminate the influence of different data dimensions. Further, using well depth as a data index, the prior information is organized into a multi-source drilling data table for subsequent geological unit division. Then, the processor can determine the geological unit division criteria based on the prior information. The geological unit division criteria refer to well segment categories with distinguishable drilling response characteristics, divided by fuzzy clustering based on geological parameters and drilling parameters. Specifically, in one embodiment, a fuzzy C-homogeneous clustering algorithm can be used to identify the multi-source drilling data table to obtain multiple geological units corresponding to the prior information and the geological unit division criteria. Furthermore, the processor can identify the real-time geological unit corresponding to the target well at the current moment using a fuzzy C-means clustering algorithm based on real-time drilling data, prior information, and geological unit division criteria. It then determines multiple sets of candidate drilling parameters from the real-time drilling data and historical logging data matched with the real-time geological units. Each set of candidate drilling parameters includes at least the drilling pressure, rotary table speed, and inlet flow rate. Next, the processor can determine the mechanical drilling rate and mechanical specific energy corresponding to each set of candidate drilling parameters using a mechanical drilling rate prediction model and a mechanical specific energy calculation formula. Finally, using a multi-objective optimization algorithm, with the objectives of maximizing the mechanical drilling rate and minimizing the mechanical specific energy, it selects and executes the target drilling parameter combination from the multiple sets of candidate drilling parameters.

[0034] In one embodiment, identifying the real-time geological unit corresponding to the target well at the current moment using the fuzzy C-means clustering algorithm includes: constructing a multi-dimensional state feature matrix based on real-time logging data and real-time drill bit condition information. ,in T For timing length, D The feature dimension is defined; based on prior information and geological unit division criteria, multiple cluster centers are initialized, along with a multidimensional state feature matrix and a membership matrix for each cluster center; the cluster centers are then updated based on the membership matrix.

[0035] in, Real-time drilling data at time i For cluster centers membership degree , m For fuzzy coefficients; Update the membership matrix based on the updated cluster centers:

[0036] in, Real-time drilling data at time i For cluster centers membership degree For real-time drilling data With cluster center Theoretical weighted distance, For real-time drilling data With cluster center Theoretical weighted distance, m For fuzzy coefficients; The objective function is determined based on the updated cluster centers and the updated membership matrix:

[0037]

[0038] in, Let be the objective function. Real-time drilling data at time i For cluster centers membership degree m For fuzzy coefficients, For real-time drilling data With cluster center Theoretical weighted distance, The prior weighting factor is calculated based on the formation resistivity and natural gamma. For real-time drilling data The One data point, Cluster center The One data point; The update steps for cluster centers and membership matrices are repeated until the change in the objective function is less than a preset threshold, at which point the updated membership matrix is ​​output. The cluster center corresponding to the highest membership degree in the membership matrix is ​​selected as the real-time geological unit. Specifically, in one embodiment, for the boundary points of geological units, a time-series window moving average membership degree can be used for label smoothing to reduce state jitter. Figure 2 As shown, in one embodiment, four different geological units were identified during the drilling of the target well: State 1 is a formation with low hardness and a relatively high drilling rate, ranging from 0.5 to 1.0; State 2 is a formation with low hardness and a moderate to low drilling rate, ranging from 0.2 to 0.5; State 3 is a formation with moderate hardness; and State 4 is a formation with relatively high hardness and a relatively low drilling rate. The drilling state classification effect is good.

[0039] In one embodiment, the method further includes constructing and training a mechanical drilling rate prediction model. Specifically, for each geological unit, a parameter correlation graph is constructed within the same geological unit. Parameter correlation diagram between different geological units On the one hand, by constructing a drilling parameter correlation diagram within the same geological unit, the nonlinear mapping relationship between drilling parameters such as drilling pressure and rotational speed and mechanical drilling rate is explored; on the other hand, drilling parameter correlation diagrams between different geological units are established to learn the drilling rate variation patterns under different geological units. Finally, by combining an attention mechanism, the influence weight of each drilling parameter on the drilling rate is revealed, thereby achieving accurate drilling rate modeling across wells and formations and interpretable recommendation of drilling parameters. Figure 3 As shown, the temporal graph convolutional encoder is composed of a temporal encoder, a graph convolutional encoder, and a residual connection layer. The temporal encoder uses a temporal convolutional network (TCN), which expands the receptive field and captures the temporal changes in parameters. The graph convolutional encoder uses a graph attention network (GAT), enabling the model to automatically extract the complex influence relationships between different drilling parameters. Next, based on the attention mechanism module, attention weight coefficients are generated, and the feature subset that contributes the most to the current information transmission is selected. Finally, the current mechanical drilling rate prediction value is output. During the training of the mechanical drilling rate prediction model, the training loss function is:

[0040] in, For sample-weighted loss, n For the sample size, The mean absolute percentage error, To control The hyperparameters of the loss weights, For mean square error loss, For the first i The true value of each sample; For the firsti The model prediction value for each sample. The loss function is used for training. The hyperparameters of the mechanical drilling rate prediction model are optimized using a Bayesian optimizer.

[0041] In one embodiment, after identifying the real-time geological unit corresponding to the target well at the current moment, multiple sets of candidate drilling parameters can be determined from real-time drilling data and historical logging data matched with the real-time geological unit. The mechanical drilling rate and mechanical specific energy corresponding to each set of candidate drilling parameters are then determined using a mechanical drilling rate prediction model and a mechanical specific energy calculation formula. The mechanical specific energy calculation formula is as follows:

[0042] in, For mechanical specific energy, For drilling pressure, The diameter of the drill bit. The rotational speed of the turntable. For torque, This refers to the mechanical drilling speed.

[0043] In one embodiment, a multi-objective optimization algorithm can be used to maximize the rate of drilling (ROD) and minimize the specific energy of the drill bit, selecting a target drilling parameter combination from multiple candidate drilling parameters and executing it. Specifically, in one embodiment, the NSGA-II multi-objective optimization algorithm is used to generate a Pareto front with the objectives of maximizing the ROD and minimizing the specific energy of the drill bit, obtaining feasible solutions, and selecting the target drilling parameter combination based on the TOPSIS integrated decision method. The multi-objective optimization function is as follows:

[0044] in, For a multi-objective optimization function, For mechanical drilling speed, This is the weighting coefficient for the mechanical drilling rate. This is the normalization factor for the mechanical drilling rate. For mechanical specific energy, This is the weighting coefficient for mechanical specific energy. This is the normalization coefficient for mechanical specific energy.

[0045] In one embodiment, such as Figure 4As shown, the target drilling parameter combination is input into the mechanical drilling rate prediction model to obtain the predicted mechanical drilling rate corresponding to the target drilling parameter combination. The relative error between the predicted mechanical drilling rate and the actual mechanical drilling rate corresponding to the execution of the target drilling parameter combination is calculated. If the relative error is greater than a preset threshold, the mechanical drilling rate prediction model is updated based on all real-time drilling data obtained during the drilling of the target well. Specifically, in one embodiment, when the relative error is greater than 85%, the mechanical drilling rate prediction model automatically updates the cluster centers and their model parameter weights. In this clustering process, the parameters are initialized based on the previous cluster centers.

[0046] Through the above embodiments, prior information about adjacent wells drilled to the target well is obtained, including historical logging data, historical well logging data, historical formation stratification data, and historical drill bit operating information. Based on this prior information, a geological unit division standard is determined. Then, based on real-time drilling data, prior information, and the geological unit division standard, a fuzzy C-means clustering algorithm is used to identify the real-time geological unit corresponding to the target well at the current moment. Next, multiple sets of candidate drilling parameters are determined from the real-time drilling data and historical logging data matching the real-time geological unit. The mechanical drilling rate (MRR) and mechanical specific energy (MSE) corresponding to each set of candidate drilling parameters are determined using a mechanical drilling rate prediction model and a mechanical specific energy calculation formula. Finally, a multi-objective optimization algorithm is used to maximize the MRR and minimize the MSE to select the target drilling parameter from the multiple sets of candidate drilling parameters. This effectively utilizes the common patterns of multi-well data, enhancing the adaptability and robustness of drilling parameter optimization to unseen formations or parameter combinations.

[0047] Figure 1 This is a flowchart illustrating a drilling parameter optimization method guided by multi-well big data in one embodiment. It should be understood that, although... Figure 1 The steps in the flowchart are shown sequentially as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise explicitly stated herein, there is no strict order in which these steps are executed, and they can be performed in other orders. Figure 1 At least some of the steps in the process may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least some of the sub-steps or stages of other steps.

[0048] In one embodiment, such as Figure 5As shown, the drilling parameter optimization method guided by multi-well big data includes: acquiring prior information on drilled wells in the target area, as well as real-time drilling data of the target well during the drilling process; guiding dynamic clustering based on prior information to determine the geological unit division criteria; after preprocessing the real-time drilling data, using the fuzzy C-means clustering algorithm to identify the real-time geological unit corresponding to the target well at the current moment based on the real-time drilling data, prior information, and geological unit division criteria, and determining multiple sets of candidate drilling parameters (i.e., multi-well multi-formation sequence dataset) from the real-time drilling data and historical logging data matching the real-time geological units. A mechanical drilling rate prediction model is constructed and trained, which is based on a multi-level graph neural network. The mechanical drilling rate and mechanical specific energy corresponding to each set of candidate drilling parameters are determined by the mechanical drilling rate prediction model and the mechanical specific energy calculation formula, respectively. A multi-objective optimization algorithm is used to select the target drilling parameter combination from multiple sets of candidate drilling parameters with the objectives of maximizing the mechanical drilling rate and minimizing the mechanical specific energy, and then the optimization is performed. The target drilling parameter combination is input into the mechanical drilling rate prediction model to obtain the predicted mechanical drilling rate corresponding to the target drilling parameter combination; the relative error between the predicted mechanical drilling rate and the actual mechanical drilling rate corresponding to the execution of the target drilling parameter combination is calculated; if the relative error is greater than a preset threshold, the mechanical drilling rate prediction model is incrementally updated based on all real-time drilling data obtained during the drilling of the target well.

[0049] In one embodiment, a drilling parameter optimization device guided by multi-well big data is provided (not shown in the figure), comprising: The memory is configured to store instructions; The processor is configured to retrieve the instructions from the memory and, when executing the instructions, to implement the aforementioned multi-well big data-guided drilling parameter optimization method.

[0050] The multi-well big data-guided drilling parameter optimization device includes a processor and a memory. The processor contains a kernel, which retrieves the corresponding program units from the memory. One or more kernels can be configured, and the multi-well big data-guided drilling parameter optimization method is implemented by adjusting the kernel parameters.

[0051] The memory may include non-permanent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.

[0052] This application provides a storage medium storing a program that, when executed by a processor, implements the aforementioned multi-well big data-guided drilling parameter optimization method.

[0053] This application provides a processor for running a program, wherein the program executes the above-described drilling parameter optimization method guided by multi-well big data.

[0054] In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 6 As shown, the computer device includes a processor A01, a network interface A02, and a memory (not shown) connected via a system bus. The processor A01 provides computing and control capabilities. The memory includes internal memory A03 and a non-volatile storage medium A04. The non-volatile storage medium A04 stores an operating system B01 and a computer program B02. The internal memory A03 provides an environment for the operation of the operating system B01 and the computer program B02 stored in the non-volatile storage medium A04. The network interface A02 is used for communication with external terminals via a network connection. When executed by the processor A01, the computer program B02 implements a drilling parameter optimization method guided by multi-well big data.

[0055] Those skilled in the art will understand that Figure 6 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0056] This application provides a computer (electronic) device, which includes a processor, a memory, and a program stored in the memory and executable on the processor. When the processor executes the program, it implements the steps of any of the above-mentioned multi-well big data-guided drilling parameter optimization methods.

[0057] This application also provides a computer program product that, when executed on a data processing device, is suitable for performing steps of an initialization method for drilling parameter optimization guided by multi-well big data.

[0058] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0059] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0060] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0061] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0062] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0063] Memory may include non-persistent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0064] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0065] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0066] The above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A multi-well big data guided drilling parameter optimization method, characterized in that, The method includes: Obtain prior information on drilled wells in the target area, wherein the target area is the area where the target well is located, and the prior information includes at least the historical logging data, historical well logging data, historical formation stratification data, and historical drill bit condition information of the drilled wells; Acquire real-time drilling data of the target well during the drilling process, wherein the real-time drilling data includes real-time logging data and real-time drill bit condition information; Based on the aforementioned prior information, the criteria for dividing geological units are determined; Based on the real-time drilling data, the prior information, and the geological unit division criteria, the fuzzy C-means clustering algorithm is used to identify the real-time geological unit corresponding to the target well at the current moment. Multiple sets of candidate drilling parameters are determined from the real-time drilling data and the historical logging data matched with the real-time geological unit; The mechanical drilling rate and mechanical specific energy corresponding to each group of candidate drilling parameters are determined by the mechanical drilling rate prediction model and the mechanical specific energy calculation formula, respectively. Using a multi-objective optimization algorithm, with the objectives of maximizing the mechanical drilling rate and minimizing the mechanical specific energy, a target drilling parameter combination is selected from multiple sets of candidate drilling parameters and then executed.

2. The drilling parameter optimization method according to claim 1, characterized in that, The step of using fuzzy C-means clustering algorithm to identify the real-time geological unit corresponding to the target well at the current moment includes: A multidimensional state feature matrix is ​​constructed based on the real-time logging data and the real-time drill bit condition information. Based on the prior information and the geological unit division criteria, multiple cluster centers are initialized, and the multidimensional state feature matrix and the membership matrix of each cluster center are initialized. The cluster centers are updated based on the membership matrix; The membership matrix is ​​updated based on the updated cluster centers; The objective function is determined based on the updated cluster centers and the updated membership matrix; Repeat the update steps for the cluster centers and the membership matrix until the change in the objective function is less than a preset threshold, and then output the updated membership matrix. The cluster center corresponding to the largest membership degree in the membership matrix is ​​selected as the real-time geological unit.

3. The drilling parameter optimization method according to claim 2, characterized in that, The objective function is determined according to the following formula: in, Let the objective function be... Real-time drilling data at time i For cluster centers membership degree m For fuzzy coefficients, For real-time drilling data With cluster center Theoretical weighted distance, The prior weighting factor is calculated based on the formation resistivity and natural gamma. For real-time drilling data The One data point, Cluster center The Data points.

4. The drilling parameter optimization method according to claim 3, characterized in that, The membership matrix and the cluster centers are updated according to the following formula: in, Real-time drilling data at time i For cluster centers membership degree For real-time drilling data With cluster center The theoretical weighted distance, For real-time drilling data With cluster center The theoretical weighted distance, m is the fuzzy coefficient.

5. The drilling parameter optimization method according to claim 1, characterized in that, The method further includes, before selecting the target drilling parameter combination from the candidate drilling parameters, using a multi-objective optimization algorithm based on the mechanical drilling rate prediction model and the mechanical specific energy calculation formula, with the objectives of maximizing the mechanical drilling rate of the drill bit and minimizing the real-time mechanical specific energy, constructing and training a mechanical drilling rate prediction model, wherein the mechanical drilling rate prediction model is a multi-level graph neural network, including a first drilling parameter correlation graph within the same geological unit and a second drilling parameter correlation graph between different geological units.

6. The drilling parameter optimization method according to claim 5, characterized in that, The training loss function of the mechanical drilling rate prediction model is: in, For sample-weighted loss, n For the sample size, The mean absolute percentage error, To control The hyperparameters of the loss weights, For mean square error loss, For the first i The true value of each sample; For the first i The model prediction value for each sample. Let be the training loss function.

7. The drilling parameter optimization method according to claim 1, characterized in that, The method further includes: The target drilling parameter combination is input into the mechanical drilling rate prediction model to obtain the predicted mechanical drilling rate corresponding to the target drilling parameter combination; Calculate the relative error between the predicted mechanical drilling rate and the actual mechanical drilling rate corresponding to the execution of the target drilling parameter combination. If the relative error is greater than a preset threshold, update the mechanical drilling rate prediction model based on all the real-time drilling data obtained during the drilling process of the target well.

8. The drilling parameter optimization method according to claim 1, characterized in that, The formula for calculating the mechanical specific energy is: in, The mechanical specific energy, For drilling pressure, The diameter of the drill bit. The rotational speed of the turntable. For torque, This refers to the mechanical drilling speed.

9. A drilling parameter optimization device guided by multi-well big data, characterized in that, include: The memory is configured to store instructions; The processor is configured to retrieve the instructions from the memory and, when executing the instructions, to implement the drilling parameter optimization method guided by multi-well big data according to any one of claims 1 to 8.

10. A machine-readable storage medium storing instructions thereon, characterized in that, When executed by a processor, this instruction causes the processor to be configured to perform a drilling parameter optimization method guided by multi-well big data according to any one of claims 1 to 8.

Images