Drilling parameter optimization method and device based on high-precision rock debris mineral and mechanical properties, electronic equipment and computer program product

Abstract

The invention relates to the technical field of well drilling, in particular to a well drilling parameter optimization method and device based on high-precision rock debris mineral and mechanical characteristics, electronic equipment and a computer program product, and the method comprises the steps that model loss is introduced, a regression model is trained through sample data, candidate well drilling parameter data are input into the trained regression model; and by taking the output result minimization of the regression model as a target, constructing a target function, optimizing the target function based on a local optimal algorithm and a global optimal algorithm, and determining target drilling parameter data of the drilling equipment. The historical time is used for guiding the regression model to conduct supervised learning, the prediction accuracy of the regression model is improved, minimization of the output result of the regression model is used as the target of the target function, that is, maximization of the operation efficiency is used as the target of the target function to solve and optimize the target drilling parameters, and compared with manual setting, the method has the advantage that the efficiency is greatly improved. The optimization effect of the drilling parameters can be improved, so that the drilling operation efficiency and the operation safety are effectively improved.

Description

Technical Field

[0001] This invention relates to the field of drilling technology, specifically a method, apparatus, electronic equipment, and computer program product for optimizing drilling parameters based on high-precision analysis of rock cuttings minerals and mechanical properties. Background Technology

[0002] With the continued growth of global energy demand, the exploration and development of oil and gas resources are facing unprecedented challenges. Complex and variable geological conditions, the high-temperature and high-pressure environment of deep wells, and ever-increasing cost control requirements are driving drilling technology towards intelligence and automation. The optimization of drilling parameters, as a key link in improving drilling efficiency and reducing costs, is becoming increasingly important.

[0003] Currently, traditional drilling parameter optimization methods are usually based on engineers' experience and judgment. This method is highly subjective and difficult to adapt to the complex and ever-changing geological conditions during drilling, resulting in poor optimization results. Furthermore, improperly set drilling parameters can reduce drilling efficiency and increase operational risks. Summary of the Invention

[0004] This invention provides a method, apparatus, electronic device, and computer program product for optimizing drilling parameters based on high-precision rock cuttings mineral and mechanical properties. It overcomes the shortcomings of the prior art and can effectively solve the problems of strong subjectivity and difficulty in adapting to complex and changing geological conditions during drilling that exist in existing traditional drilling parameter optimization methods.

[0005] One of the technical solutions of this invention is achieved through the following measures: a drilling parameter optimization method based on high-precision rock cuttings mineral and mechanical properties, comprising:

[0006] Acquire several sample data, each of which includes historical mineral composition data, historical rock mechanical parameter data, historical drilling parameter data of drilling equipment, and a corresponding historical time identifier. The historical time is the time taken by the drilling equipment to break a unit volume of rock.

[0007] Introduce model loss and train the regression model using sample data;

[0008] All candidate drilling parameter data are input into the trained regression model. The objective function is constructed with the goal of minimizing the output of the regression model. The objective function is then optimized based on local and global optimum algorithms to determine the target drilling parameter data for the drilling equipment.

[0009] The following are further optimizations and / or improvements to the above-mentioned technical solution:

[0010] The above-mentioned introduction of model loss and training of the regression model using sample data includes:

[0011] A number of sample data are divided into a training sample set and a test sample set according to a certain ratio;

[0012] The regression model is trained using the training sample set in conjunction with the training termination condition;

[0013] The historical mineral composition data, historical rock mechanics parameter data, and historical drilling parameter data corresponding to each sample data in the test sample set are input into the regression model for regression to obtain the prediction time. The model loss is determined based on the difference between the prediction time and the historical time, and the regression model is tested and optimized based on the model loss.

[0014] The historical timeframes for obtaining the data for each sample mentioned above include:

[0015] Obtain multiple reference drilling depths for the drilling equipment, and the reference drilling time corresponding to each reference drilling depth;

[0016] The first fitting curve is obtained by fitting multiple reference drilling depths and their corresponding reference drilling times.

[0017] The initial drilling time is determined based on the first fitted curve;

[0018] Historical rock-breaking volume is determined based on historical drilling depth and historical drilling diameter.

[0019] The historical time corresponding to the sample data is determined by the quotient of the initial drilling time and the historical rock breaking volume.

[0020] The historical mineral composition data and historical rock mechanical parameter data obtained for each sample above include:

[0021] Acquire three-dimensional images of microcores corresponding to multiple reference drilling depths, as well as reference mineral composition data corresponding to each reference drilling depth;

[0022] The number of samples is obtained, and historical drilling depths are uniformly sampled based on the number of samples to obtain multiple sampled drilling depths;

[0023] The data of multiple reference drilling depths and their corresponding reference mineral components are fitted to obtain a second fitting curve. Based on the multiple sampled drilling depths and the second fitting curve, the historical mineral component data corresponding to each sampled drilling depth is determined.

[0024] For each reference drilling depth, the reference rock mechanics parameters corresponding to each reference drilling depth are determined based on the corresponding three-dimensional image and the corresponding reference mineral composition data.

[0025] Multiple reference drilling depths and their corresponding reference rock mechanics parameters are fitted to obtain a third fitting curve. Based on multiple sampled drilling depths and the third fitting curve, the historical rock mechanics parameter data corresponding to each sampled drilling depth are determined.

[0026] The above-mentioned optimization of the objective function based on local and global optimum algorithms determines the target drilling parameter data for the drilling equipment, including:

[0027] Obtain the parameter value range, create multiple parameter pools for the objective function, and randomly generate candidate drilling parameter data within the parameter value range in each parameter pool;

[0028] The candidate drilling parameter data generated in each parameter pool are input into the regression model for regression to obtain multiple candidate times.

[0029] Based on the candidate time, the globally optimal parameter data is determined among all candidate drilling parameter data, and the locally optimal parameter data for each parameter pool is determined among the candidate drilling parameter data in each parameter pool.

[0030] Determine whether the candidate time corresponding to the globally optimal parameter data is within a predetermined threshold or whether the number of adjustments has reached a predetermined number;

[0031] Whether or not it responds, based on the global optimal parameter data and the local optimal parameter data, the candidate drilling parameter data currently generated in each parameter pool is adjusted, and new candidate drilling parameter data is generated again in each parameter pool and re-input into the regression model;

[0032] Therefore, the currently determined global optimal parameter data is set as the target drilling parameter data for the drilling equipment.

[0033] The above-mentioned method adjusts the candidate drilling parameter data currently generated in each parameter pool based on globally optimal and locally optimal parameter data, and then generates new candidate drilling parameter data in each parameter pool, including:

[0034] Based on the global optimal parameter data and the local optimal parameter data, the parameter adjustment amount corresponding to each parameter pool is updated respectively;

[0035] Based on the updated parameter adjustments, the candidate drilling parameter data currently generated in each parameter pool are adjusted, and new candidate drilling parameter data are generated again in each parameter pool.

[0036] The second technical solution of the present invention is achieved through the following measures: a drilling parameter optimization device based on high-precision rock cuttings mineral and mechanical properties, comprising:

[0037] The sample acquisition unit acquires several sample data, each of which includes historical mineral composition data, historical rock mechanical parameter data, historical drilling parameter data of the drilling equipment, and a corresponding historical time identifier. The historical time is the time taken by the drilling equipment to break a unit volume of rock.

[0038] The training unit introduces model loss and uses sample data to train the regression model;

[0039] The parameter optimization unit inputs all candidate drilling parameter data into the trained regression model, constructs an objective function with the goal of minimizing the output of the regression model, and optimizes the objective function based on local optimum and global optimum algorithms to determine the target drilling parameter data for the drilling equipment.

[0040] The following are further optimizations and / or improvements to the above-mentioned technical solution:

[0041] The above parameter optimization unit includes:

[0042] The initial processing module obtains the parameter value range, creates multiple parameter pools for the objective function, and randomly generates candidate drilling parameter data within the parameter value range in each parameter pool;

[0043] The regression module inputs the candidate drilling parameter data generated in each parameter pool into the regression model for regression to obtain multiple candidate times.

[0044] The optimization solution module determines the globally optimal parameter data among all candidate drilling parameter data based on candidate time, and determines the locally optimal parameter data for each parameter pool among the candidate drilling parameter data in each parameter pool.

[0045] Determine whether the candidate time corresponding to the globally optimal parameter data is within a predetermined threshold or whether the number of adjustments has reached a predetermined number;

[0046] Whether or not it responds, based on the global optimal parameter data and the local optimal parameter data, the candidate drilling parameter data currently generated in each parameter pool is adjusted, and new candidate drilling parameter data is generated again in each parameter pool and re-input into the regression model;

[0047] Therefore, the currently determined global optimal parameter data is set as the target drilling parameter data for the drilling equipment.

[0048] The third technical solution of the present invention is achieved through the following measures: an electronic device, including a processor and a memory, wherein a computer program is stored in the memory, and the computer program is loaded and executed by the processor to implement the steps in the drilling parameter optimization method based on high-precision rock cuttings minerals and mechanical properties.

[0049] The fourth technical solution of the present invention is achieved through the following measures: a computer program product, including a computer program or instructions, characterized in that, when the computer program or instructions are executed by a processor, they implement the steps in the drilling parameter optimization method based on high-precision rock cuttings minerals and mechanical properties.

[0050] This invention offers at least the following advantages: By acquiring sample data containing historical mineral composition data, historical rock mechanics parameter data, and historical drilling parameter data of drilling equipment, and then inputting these data into a regression model for regression, the regression model can predict the prediction time. Historical time can serve as an ideal generation result for the prediction time. Therefore, by using historical time to guide supervised learning of the regression model, and continuously reducing model loss through iterative training, the regression model learns to understand the feature information contained in multi-dimensional mineral composition, rock mechanics parameters, and drilling parameters, thereby improving the regression performance. The model's prediction accuracy is determined by the prediction time, which is the time taken by the drilling equipment to break a unit volume of rock. Therefore, the prediction time accurately represents the operating efficiency of the drilling equipment. The objective function is to minimize the output of the regression model, which is equivalent to minimizing the prediction time, i.e., maximizing the operating efficiency. Since the target drilling parameters of the drilling equipment are determined by optimizing the objective function, the optimized target drilling parameters can maximize the operating efficiency of the drilling equipment. Compared with manual setting, this can improve the optimization effect of drilling parameters, thereby effectively improving drilling operation efficiency and safety. Attached Figure Description

[0051] Appendix Figure 1 This is a schematic diagram of a drilling parameter optimization method according to the present invention.

[0052] Appendix Figure 2 This is a schematic diagram of the method for obtaining historical time data of each sample in this invention.

[0053] Appendix Figure 3 This is a schematic diagram of the method for reconstructing the first fitted curve in this invention.

[0054] Appendix Figure 4 This is a schematic diagram of the method for obtaining historical mineral composition data and historical rock mechanical parameter data of various samples in this invention.

[0055] Appendix Figure 5 This is an optional schematic diagram of a first fitting curve provided in an embodiment of the present invention.

[0056] Appendix Figure 6 This is an optional schematic diagram of a second fitting curve provided in an embodiment of the present invention.

[0057] Appendix Figure 7 This is an optional schematic diagram of a third fitting curve provided in an embodiment of the present invention.

[0058] Appendix Figure 8 This is a schematic diagram of the method for determining the target drilling parameter data of drilling equipment in this invention.

[0059] Appendix Figure 9 This is a schematic diagram of the method for adjusting the candidate drilling parameter data currently generated in each parameter pool in this invention.

[0060] Appendix Figure 10 This is a schematic diagram of a drilling parameter optimization device according to the present invention. Detailed Implementation

[0061] The present invention is not limited to the following embodiments, and the specific implementation can be determined according to the technical solution of the present invention and the actual situation.

[0062] It should be noted that in various specific embodiments of this application, when processing data related to the characteristics of the target object, such as target object attribute information or attribute information sets, is required, the permission or consent of the target object will be obtained first. Furthermore, the collection, use, and processing of this data will comply with relevant laws, regulations, and standards. The target object can be a user. In addition, when embodiments of this application need to obtain target object attribute information, separate permission or consent from the target object will be obtained through pop-ups or redirection to a confirmation page. Only after obtaining the target object's separate permission or consent will the necessary target object-related data for the normal operation of the embodiments of this application be obtained.

[0063] In the description of this application, "several" means one or more, "multiple" means two or more, "greater than", "less than", "exceeding" etc. are understood to exclude the number itself, and "above", "below", "within" etc. are understood to include the number itself.

[0064] It should be noted that although functional modules are divided in the device schematic diagram and a logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the device or the order in the flowchart. The terms "first," "second," etc., in the specification, claims, or the aforementioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0065] To facilitate understanding of the technical solutions provided in the embodiments of this application, some key terms used in the embodiments of this application will be explained below:

[0066] Artificial intelligence (AI) is the theory, methods, technology, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to achieve optimal results. In other words, AI is a comprehensive technology within computer science that attempts to understand the essence of intelligence and produce a new kind of intelligent machine that can react in a way similar to human intelligence. AI studies the design principles and implementation methods of various intelligent machines, enabling them to possess the functions of perception, reasoning, and decision-making.

[0067] Artificial intelligence (AI) is a comprehensive discipline encompassing a wide range of fields, including both hardware and software technologies. Fundamental AI technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing, operating / interactive systems, and mechatronics. AI software technologies primarily include computer vision, speech processing, natural language processing, as well as machine learning / deep learning, autonomous driving, and intelligent transportation.

[0068] Machine Learning (ML) is a multidisciplinary field involving probability theory, statistics, approximation theory, convex analysis, and algorithm complexity theory. It specifically studies how computers can simulate or implement human learning behavior to acquire new knowledge or skills and reorganize existing knowledge structures to continuously improve their performance. Machine learning is the core of artificial intelligence and the fundamental way to endow computers with intelligence; its applications span all areas of artificial intelligence. Machine learning and deep learning typically include techniques such as artificial neural networks, belief networks, reinforcement learning, transfer learning, inductive learning, and instruction-based learning.

[0069] Currently, traditional drilling parameter optimization methods are usually based on engineers' experience and judgment. This method is highly subjective and difficult to adapt to the complex and ever-changing geological conditions during drilling, resulting in poor optimization results. Furthermore, improperly set drilling parameters can reduce drilling efficiency and increase operational risks.

[0070] To address the issue of poor performance in manually optimizing drilling parameters, this application provides a method, apparatus, electronic device, and computer program for optimizing drilling parameters based on high-precision rock cuttings mineral and mechanical properties. The method includes: acquiring several sample data sets, each including historical mineral composition data, historical rock mechanical parameter data, historical drilling parameter data of the drilling equipment, and a corresponding historical time identifier, where the historical time is the time taken by the drilling equipment to break a unit volume of rock; introducing model loss and training a regression model using the sample data; inputting candidate drilling parameter data into the trained regression model, constructing an objective function with the goal of minimizing the output of the regression model, and optimizing the objective function based on local and global optimum algorithms to determine the target drilling parameter data for the drilling equipment.

[0071] According to the solution provided in this application, sample data including historical mineral composition data, historical rock mechanics parameter data, and historical drilling parameter data of drilling equipment are obtained. This data is then input into a regression model for regression. The regression model can predict the prediction time, and historical time can serve as the ideal generation result for the prediction time. Therefore, historical time is used to guide the regression model in supervised learning. Through iterative training, the model loss is continuously reduced, enabling the regression model to learn and understand the feature information contained in multi-dimensional mineral composition, rock mechanics parameters, and drilling parameters, thereby improving the prediction accuracy of the regression model. Since the prediction time is the time taken by the drilling equipment to break a unit volume of rock, the prediction time can accurately characterize the operating efficiency of the drilling equipment. Minimizing the output result of the regression model is then used as the objective function, which is equivalent to minimizing the prediction time, i.e., maximizing operating efficiency. Since the target drilling parameter data of the drilling equipment is determined by optimizing the objective function, the optimized target drilling parameter data can maximize the operating efficiency of the drilling equipment. Compared with manual setting, this improves the optimization effect of drilling parameters, thereby effectively improving drilling operation efficiency and safety.

[0072] The drilling parameter optimization method, apparatus, electronic equipment, and computer program based on high-precision rock cuttings minerals and mechanical properties provided in this application are specifically described through the following embodiments. First, the drilling parameter optimization method based on high-precision rock cuttings minerals and mechanical properties in this application embodiment is described.

[0073] The drilling parameter optimization method based on high-precision rock cuttings mineral and mechanical properties provided in this application relates to the field of computer technology. The drilling parameter optimization method provided in this application can be applied to a terminal, a server, or software running on either a terminal or a server. In some embodiments, the terminal can be a smartphone, tablet, laptop, desktop computer, etc.; the server can be configured as an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms; the software can be an application implementing the drilling parameter optimization method, but is not limited to the above forms.

[0074] This application can be used in a wide variety of general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0075] The embodiments of this application will be further described below with reference to the accompanying drawings.

[0076] Example 1: As shown in the attached document Figure 1 As shown, this invention discloses a drilling parameter optimization method based on high-precision rock cuttings mineral and mechanical properties. This drilling parameter optimization method can be executed by a server, a terminal, or a server in conjunction with a terminal. The drilling parameter optimization method includes, but is not limited to, the following steps S110 to S130:

[0077] Step S110: Obtain several sample data, wherein each sample data includes historical mineral composition data, historical rock mechanical parameter data, historical drilling parameter data of drilling equipment, and a corresponding historical time identifier. The historical time is the time taken by the drilling equipment to break a unit volume of rock.

[0078] It should be noted that the sample data can come from data recorded during the drilling process of multiple wells.

[0079] The sample data includes historical mineral composition data, historical rock mechanics parameter data, and historical drilling parameter data of drilling equipment;

[0080] Among them, historical mineral composition data is obtained by conducting historical mineral composition tests on drilling sites, and one or more historical mineral composition data can be obtained; historical rock mechanical parameter data is obtained through finite element simulation calculation and analysis of mechanical parameters, and one or more historical rock mechanical parameter data can be obtained.

[0081] Historical mineral composition data can be collected from aspects such as content, grain size, crystal structure, and chemical composition. In terms of mineral types, historical mineral composition data can include quartz, feldspar, mica, carbonate minerals, clay minerals, and sulfate minerals. For drilling missions aimed at oil exploration, historical mineral composition data can include organic matter content, barite, bitumen, and kerogen. Historical rock mechanical parameters can include density, elastic modulus, and Poisson's ratio. Based on this, historical mineral composition data and historical rock mechanical parameter data can essentially describe the complex and variable geological environment.

[0082] Historical drilling parameter data for drilling equipment may include drill bit type, drilling pressure, displacement, drilling speed, etc.

[0083] Among them, historical time is the time taken by drilling equipment to break a unit volume of rock, and the time taken by drilling equipment to break a unit volume of rock is the average time taken to break a unit volume of rock from the start of drilling to a certain depth.

[0084] It should be noted that the sample data also includes historical drilling depth and the corresponding historical drilling diameter. Historical drilling depth represents the depth reached by the drilling equipment that is the source of the sample data at a certain moment in the same well that is also the source of the sample data. Historical drilling diameter is the diameter of the tangential circle where the well location at the historical drilling depth is located. The corresponding historical drilling area can be calculated from the historical drilling diameter.

[0085] Step S120: Introduce model loss and train the regression model using sample data.

[0086] The regression model can be random forest regression, gradient boosting regression, XGBoost, or neural network, and the embodiments in this application are not limited to these.

[0087] Based on this, during training, historical mineral composition data, historical rock mechanics parameter data, and historical drilling parameter data of the training samples are input into the regression model for regression to obtain the predicted time. This allows us to find the intrinsic relationship between historical mineral composition data, historical rock mechanics parameter data, and historical drilling parameter data and the predicted time, thereby improving the reliability of the predicted time. Here, the predicted time is the time taken by the drilling equipment to break a unit volume of rock, which is the average time taken to break a unit volume of rock from the start of drilling to a certain depth.

[0088] Specifically, the training process may include, but is not limited to, the following: introducing a model loss and training the regression model using sample data, including:

[0089] (1) Divide a number of sample data into a training sample set and a test sample set according to the proportion;

[0090] (2) Train the regression model using the training sample set in conjunction with the training termination condition; where the training termination condition can be the number of training iterations.

[0091] (3) Input the historical mineral composition data, historical rock mechanics parameter data, and historical drilling parameter data corresponding to each sample data in the test sample set into the regression model for regression to obtain the predicted time. Determine the model loss based on the difference between the predicted time and the historical time, and test and optimize the regression model based on the model loss. The model loss here can be a loss function. If the loss value of the loss function is less than a threshold, the entire model training process ends. If it is not less than the threshold, the regression model parameters are adjusted and retraining is performed.

[0092] During this training process, historical time is used to guide the regression model in supervised learning. Through iterative training, the model loss is continuously reduced, enabling the regression model to learn and understand the feature information contained in multi-dimensional mineral composition, rock mechanical parameters, and drilling parameters, thereby improving the prediction accuracy of the regression model.

[0093] Step S130: Input all candidate drilling parameter data into the trained regression model, construct an objective function with the goal of minimizing the output of the regression model, and optimize the objective function based on local optimum and global optimum algorithms to determine the target drilling parameter data of the drilling equipment.

[0094] It should be noted that the output of the regression model is the predicted time, and the model formula is as follows:

[0095]

[0096] Furthermore, the formula for the objective function is as follows:

[0097] J = minHθ x

[0098] Among them, H θ For the regression model, θ represents the model parameters, and x represents the input parameters, which include historical mineral composition data, historical rock mechanics parameter data, and historical drilling parameter data. The prediction time is J, where J is the objective function, and minH is the objective function. θ x represents minimizing the prediction time of the regression model's output value.

[0099] Specifically, in order to obtain the optimized value of the objective function, it is necessary to solve the objective function. The method for solving the objective function provided in this application is an iterative method, that is, to optimize the objective function based on local optimum and global optimum algorithms. Specifically, a certain range is selected in the domain of x as the input of the objective function, and the input of the objective function is adjusted through iteration. The final result is the target drilling parameter data.

[0100] Based on this, since the prediction time is the time it takes for the drilling equipment to break a unit volume of rock, the prediction time can accurately characterize the operating efficiency of the drilling equipment. By taking the minimization of the output of the regression model as the objective function, it is equivalent to taking the minimization of the prediction time as the objective function, that is, taking the maximization of operating efficiency as the objective function. Since the target drilling parameter data of the drilling equipment is determined by optimizing the objective function, the optimized target drilling parameter data can maximize the operating efficiency of the drilling equipment. Compared with manual setting, it can improve the optimization effect of drilling parameters, thereby effectively improving drilling operation efficiency and operation safety.

[0101] Example 2: As shown in the attached document Figure 2 As shown, this embodiment is a further optimization of the above embodiment. Specifically, obtaining the historical time of each sample data includes:

[0102] Step S210: Obtain multiple reference drilling depths of the drilling equipment and the reference drilling time corresponding to each reference drilling depth.

[0103] Step S22: Fit multiple reference drilling depths and their corresponding reference drilling times to obtain the first fitting curve.

[0104] It should be noted that when drilling, the drilling equipment records the current depth at regular intervals. These recorded depths and the corresponding time are respectively the reference drilling depth and reference drilling time.

[0105] It should be noted that the reference appendix Figure 5The reference drilling depth can be used as the ordinate of the first fitting curve, and the reference drilling time can be used as the abscissa of the first fitting curve. The number of reference drilling depths can be determined based on past experience and experiments. The goal is to fit the first fitting curve relatively smoothly.

[0106] Based on this, fitting the reference drilling depth and reference drilling time to obtain the first fitting curve can provide richer sample data for the regression model, thereby improving the training effect of the regression model.

[0107] Step S230: Determine the initial drilling time based on the first fitted curve.

[0108] Among them, the historical drilling depth can be compared with the depth coordinate of the first fitted curve to obtain the corresponding initial drilling time, that is, the time consumed by the drilling equipment to reach the historical drilling depth.

[0109] Based on this, determining the initial drilling by using historical drilling depth and the first fitted curve can reduce the difficulty of obtaining the historical time corresponding to the sample data and improve the efficiency of creating sample datasets.

[0110] Step S240: Determine the historical rock-breaking volume based on historical drilling depth and historical drilling diameter.

[0111] Step S250: Determine the historical time corresponding to the sample data based on the quotient of the initial drilling time and the historical rock breaking volume.

[0112] It should be noted that during drilling, as the depth increases, the required drill bit becomes smaller, and the corresponding well diameter also decreases. For example, at the beginning of drilling, the first drill bit used in the depth range of 0 to 100 meters is the largest, while the second drill bit used in the depth range of 100 to 200 meters will be smaller than the first drill bit. Consequently, the well diameter in the depth range of 0 to 100 meters will be larger than that in the depth range of 100 to 200 meters.

[0113] Specifically, the historical rock breaking volume needs to be calculated by taking multiple rock breaking volumes across multiple depth ranges, and then adding these multiple rock breaking volumes together to obtain the historical rock breaking volume.

[0114] The volume of a single rock break is equal to the historical drilling depth multiplied by the area of ​​the circle corresponding to the historical drilling diameter.

[0115] For example, see attached document. Figure 5 The historical rock-breaking volume corresponding to a historical drilling depth of 100 meters is depth 100 meters * (well diameter 20 meters / 2)^2 * π = 10000π (cubic meters), and the historical time is 20 / 10000π.

[0116] Reference Appendix Figure 8The historical rock-breaking volume corresponding to a historical drilling depth of 200 meters is 10000π cubic meters + (depth 200 meters - depth 100 meters) * (well diameter 10 meters / 2)^2 * π = 12500π (cubic meters), and the historical time is 15 / 12500π.

[0117] Based on this, by determining the historical rock-breaking volume based on historical drilling depth and historical drilling diameter, and then determining the historical time corresponding to the sample data based on the quotient of the initial drilling time and the historical rock-breaking volume, the drilling efficiency of the drilling equipment is quantified, providing more suitable data for training the regression model, which can improve the efficiency and training effect of the regression model.

[0118] Furthermore, if the first fitted curve fails to fit successfully or cannot effectively determine the initial drilling time in the above embodiments, the first fitted curve can be reconstructed, as shown in the attached figure. Figure 3 As shown, it includes, but is not limited to, the following steps:

[0119] Step S310: Obtain the non-drilling time corresponding to each drilling depth range.

[0120] Step S320: Adjust the corresponding segmented curves based on each non-drilling time.

[0121] Step S330: Reconstruct the first fitted curve based on the adjusted piecewise curve.

[0122] Step S340: Determine the initial drilling time based on the historical drilling depth and the reconstructed first fitted curve.

[0123] The main non-drilling time is caused by processes such as changing drill bits, cleaning the bottom of the well, enlarging the wellbore, and tripping in and out of the drill string.

[0124] The first fitted curve includes multiple segmented curves. The smoothness of the connection point of each segmented curve is clearly distinguishable from other parts of the scaled curve. The segmented curves need to be obtained in such a way that the depth does not increase with the significant increase of time. Such curve segments can be identified as non-drilling time and need to be removed.

[0125] Based on this, by reconstructing the original first fitted curve by subtracting non-drilling time, the rationality of the first fitted curve and the accuracy of the determined initial drilling time can be improved.

[0126] In one possible implementation, the adjusted piecewise curves are connected sequentially, and the corresponding time coordinates are reduced by the amount of non-drilling time.

[0127] In another possible implementation, the adjusted piecewise curve is stretched using the two endpoints of the original piecewise curve as a reference, so that all the piecewise curves are connected in sequence.

[0128] Example 3: As shown in the attached document Figure 4 As shown, this embodiment is a further optimization of the above embodiment. Specifically, it obtains historical mineral composition data and historical rock mechanical parameter data for each sample, including:

[0129] Step S410: Obtain three-dimensional images of microcores corresponding to multiple reference drilling depths, as well as reference mineral composition data corresponding to each reference drilling depth.

[0130] Step S420: Obtain the number of samples, and uniformly sample the historical drilling depths based on the number of samples to obtain multiple sampled drilling depths.

[0131] Step S430: Fit multiple reference drilling depths and corresponding reference mineral composition data to obtain a second fitting curve. Based on multiple sampled drilling depths and the second fitting curve, determine the historical mineral composition data corresponding to each sampled drilling depth.

[0132] It should be noted that there are multiple sets of historical mineral composition data and historical rock mechanical parameter data.

[0133] For the fitting of the second fitted curve, refer to the appendix. Figure 6 The reference drilling depth can be used as the horizontal axis, and the corresponding proportion of reference mineral components can be used as the vertical axis. The number of reference drilling depths can be determined based on past experience and experiments. The goal is to fit the second fitting curve more smoothly.

[0134] It should be noted that the reference mineral composition includes one or more mineral components, and the reference mineral composition includes the mass of each mineral component. For any mineral component, multiple reference drilling depths and their corresponding masses can be fitted to obtain the corresponding second fitting curve. Therefore, by fitting multiple reference drilling depths and their corresponding reference mineral composition data, the second fitting curve corresponding to each mineral component can be obtained.

[0135] Uniform sampling starts from the wellhead and samples the well at regular intervals until the historical drilling depth is reached. The sampling distance is equal to the quotient of the historical drilling depth and the number of samples minus one. The number of samples can be determined based on past experience and experiments. The purpose is to ensure that the historical mineral composition data obtained from the sampling can comprehensively represent the overall mineral composition of the well.

[0136] Based on this, by uniformly sampling the historical drilling depths using the fitted second curve, multiple sampled drilling depths are obtained, and the historical mineral composition data corresponding to each sampled drilling depth is determined. This enables the sampled historical mineral composition data to more comprehensively represent the overall mineral composition of the well, thereby improving the accuracy of the regression model in determining the prediction time.

[0137] Step S440: For each reference drilling depth, determine the reference rock mechanics parameters corresponding to each reference drilling depth based on the corresponding three-dimensional image and the corresponding reference mineral composition data.

[0138] Step S450: Fit multiple reference drilling depths and their corresponding reference rock mechanics parameters to obtain a third fitting curve. Based on multiple sampled drilling depths and the third fitting curve, determine the historical rock mechanics parameter data corresponding to each sampled drilling depth.

[0139] Among them, a computed tomography (CT) device can be used to acquire three-dimensional images of microcores corresponding to each reference drilling depth, and reference mineral composition data can be obtained for each reference drilling depth based on the second fitting curve.

[0140] Among them, reference appendix Figure 7 The reference drilling depth can be used as the abscissa of the third fitting curve, and the corresponding reference rock mechanics parameters can be used as the ordinate of the third fitting curve. The number of reference drilling depths can be determined based on past experience and experiments. The goal is to fit the third fitting curve more smoothly.

[0141] It should be noted that the reference rock mechanics parameters include one or more mineral components. The reference rock mechanics parameters include each physical quantity that describes the rock mechanics properties. For any physical quantity, multiple reference drilling depths and their corresponding physical quantities can be fitted to obtain the corresponding third fitting curve. Therefore, by fitting multiple reference drilling depths and their corresponding reference rock mechanics parameter data, the third fitting curve corresponding to each physical quantity describing the rock mechanics properties can be obtained.

[0142] Based on this, a third fitting curve is obtained by fitting reference drilling depth and reference rock mechanics parameters. Then, the historical rock mechanics parameter data corresponding to the sampled drilling depth is determined according to the third fitting curve. This enables the sampled historical rock mechanics parameter data to more comprehensively represent the overall rock mechanics parameter characteristics of the well, thereby improving the accuracy of the regression model in determining the prediction time.

[0143] In one possible implementation, reference rock mechanics parameters for each reference drilling depth are determined based on the corresponding 3D images and corresponding reference mineral composition data. Specifically, for each reference drilling depth, a 3D porous mineral structure model is constructed based on the corresponding 3D images and corresponding reference mineral composition data. Each 3D porous mineral structure model is then sliced ​​to obtain multiple 2D porous mineral structure models. Based on each 2D porous mineral structure model, a corresponding digital core geometric model is constructed. Numerical simulations of the compressive fracturing process of each digital core geometric model are then performed using the finite element method and the discrete element method to obtain the reference rock mechanics parameters for each digital core geometric model.

[0144] Example 4: As shown in the appendix Figure 8 As shown, this embodiment is a further optimization of the above embodiment. Specifically, it optimizes the objective function based on local and global optimization algorithms to determine the target drilling parameter data for the drilling equipment, including:

[0145] Step S510: Obtain the parameter value range, create multiple parameter pools for the objective function, and randomly generate candidate drilling parameter data within the parameter value range in each parameter pool.

[0146] Among them, the parameter range is set according to the operating conditions of the drilling equipment. The operating conditions of the drilling equipment can include aspects such as wellhead safety, drilling platform stability, drilling fluid performance, and bottom hole pressure.

[0147] The parameter pool can be a database used to store candidate drilling data, and the parameter range can be used to indicate the upper and lower limits of candidate drilling parameter data.

[0148] Each parameter pool describes a potential optimal solution for a specific drilling parameter of the drilling rig relative to the objective function. The parameter range limits the variation range of the candidate drilling parameter data, ensuring that the candidate drilling parameter data always maintains a certain degree of rationality. For example, if a candidate drilling parameter data is generated in a parameter pool, with data including drilling pressure (x), drilling speed (y), and displacement (z), and the parameter range limits the domain of drilling pressure to X, drilling speed to Y, and displacement to Z, then regardless of how drilling pressure (x), drilling speed (y), and displacement (z) change, they will always remain within their respective domains.

[0149] Based on this, by creating multiple parameter pools for the objective function and randomly generating candidate drilling parameter data within the parameter value range for each parameter pool, the rationality of the candidate drilling parameter data can be maintained, and the error rate of subsequent optimization results based on the candidate drilling parameter data can be reduced.

[0150] Step S520: Input the currently generated candidate drilling parameter data in each parameter pool into the regression model for regression to obtain multiple candidate times.

[0151] The regression model can output the corresponding predicted time by inputting drilling parameters. In this case, it is assumed that the predicted time output by the regression model is consistent with the actual time when the actual drilling parameters are input into the drilling equipment.

[0152] Based on this, the candidate time corresponding to the candidate drilling parameter data can be obtained through regression model, without the need to determine the candidate time through experiments, which can improve optimization efficiency.

[0153] Step S530: Based on the candidate time, determine the globally optimal parameter data among all candidate drilling parameter data, and determine the locally optimal parameter data for each parameter pool among the candidate drilling parameter data in each parameter pool.

[0154] Among them, the shorter the candidate time, the better the corresponding candidate drilling parameter data. The candidate times can be sorted from smallest to largest, and the drilling parameter data corresponding to the shortest candidate time can be selected as the global optimal parameter data.

[0155] In this context, locally optimal parameter data refers to the candidate drilling parameter data corresponding to the shortest candidate time that each parameter pool can find throughout the entire optimization process. For example, the locally optimal parameter data of a certain parameter pool is A, and the candidate time corresponding to A is a, where a is the shortest candidate time that the parameter pool can obtain during the optimization process. The current candidate drilling parameter data is B, and the candidate time corresponding to B is b. In the next optimization, the candidate drilling parameter data is C, and the candidate time corresponding to C is c, where b is greater than a. Therefore, B cannot be used as the locally optimal parameter data of the parameter pool. When c is less than a, the locally optimal parameter data of the parameter pool will be replaced by C. When c is greater than a, the locally optimal parameter data of the parameter pool will still be A.

[0156] Based on this, by selecting the globally optimal parameter data from all candidate drilling parameter data, and determining the locally optimal parameter data for each parameter pool from the candidate drilling parameter data in each parameter pool, the global optimum can be derived from the local optimum. This decomposes the problem into multiple local problems, which can be solved in parallel, thereby improving the efficiency of obtaining the globally optimal solution.

[0157] Step S540: Determine whether the candidate time corresponding to the globally optimal parameter data is at a predetermined threshold or whether the number of adjustments has reached a predetermined number.

[0158] Step S550: In response to whether or not, based on the global optimal parameter data and the local optimal parameter data, adjust the candidate drilling parameter data currently generated in each parameter pool, generate new candidate drilling parameter data in each parameter pool, and re-input them into the regression model.

[0159] In step S560, the response is to determine the currently determined global optimal parameter data as the target drilling parameter data for the drilling equipment.

[0160] In this process, the candidate drilling parameter data generated in each parameter pool is adjusted once. Generally, there is still a considerable gap between the data and the true global optimum. Therefore, multiple adjustments are required to make the global optimum parameter data approach the true global optimum solution. Thus, the candidate drilling parameter data generated in each parameter pool is adjusted again, and new candidate drilling parameter data is generated in each parameter pool. This process continues until the candidate time corresponding to the global optimum parameter data is less than a predetermined threshold or the number of adjustments reaches a predetermined number. Only then is the global optimum parameter data output as the target drilling parameter for the drilling equipment.

[0161] Based on this, by making multiple adjustments to the candidate drilling parameter data currently generated in each parameter pool, it is possible to quickly approximate the true optimal drilling parameters and improve the accuracy of the target drilling parameters.

[0162] Example 5: As shown in the attached document Figure 9 As shown, this embodiment is a further optimization of the above embodiment. Specifically, based on the globally optimal parameter data and the locally optimal parameter data, the candidate drilling parameter data currently generated in each parameter pool is adjusted, and new candidate drilling parameter data is generated again in each parameter pool, including:

[0163] Step S610: Based on the global optimal parameter data and the local optimal parameter data, update the parameter adjustment amount corresponding to each parameter pool.

[0164] It should be noted that each parameter pool is equipped with randomly generated parameter adjustment values.

[0165] The random range for parameter adjustment is derived from past experience and experiments and has a certain degree of rationality. To make the parameter adjustment more reasonable, it needs to be updated according to the following parameter adjustment formula:

[0166] v i (t+1)=w·v i (t)+c1·r1·(p best,i -x i (t))+c2·r2·(g best -x i (t))

[0167] Among them, v i (t+1) is the adjusted parameter adjustment amount, w is the inertia weight, and v i (t) is the current parameter adjustment amount, w·v i (t) is the product of the inertia weight and the current parameter adjustment, representing the weighted average of the current parameter adjustment, indicating that a portion of the current parameter adjustment can be retained, c1·r1·(p best,i -x i (t) is the individual learning term, representing the tendency of candidate drilling parameter data in each parameter pool to move towards locally optimal parameter data based on individual experience, c2·r2·(g best -x i (t) is the social learning term, representing the trend of candidate drilling parameter data in the parameter pool moving towards the globally optimal parameter data based on collective experience. Here, c1 and c2 represent learning factors, determining the degree of influence of individual and collective experience on parameter adjustment. r1 and r2 are random numbers between 0 and 1, providing randomness to aid in exploring new regions. p best,i It represents the locally optimal parameter data of the i-th parameter pool, g. best It is the globally optimal parameter data, x i (t) represents the current candidate drilling parameter data for the i-th parameter pool.

[0168] Based on this, by updating the parameter adjustment amount corresponding to each parameter pool based on the global optimal parameter data and the local optimal parameter data, and adapting it to each optimization, the candidate drilling parameter data will not be too small or over-adjusted, which can improve the accuracy of the parameter adjustment amount in adjusting the candidate drilling parameter data.

[0169] Step S620: Based on the updated parameter adjustment amounts, adjust the currently generated candidate drilling parameter data in each parameter pool, and generate new candidate drilling parameter data in each parameter pool.

[0170] The new candidate drilling parameter data can be represented by the following formula:

[0171] x i (t+1)=x i (t)+v i (t+1)

[0172] Where, x i (t+1) represents candidate drilling parameter data, x i (t) represents the candidate drilling parameter data before adjustment, v i (t+1) is the updated parameter adjustment amount obtained in step S610. Here, the new candidate drilling parameter data can be calculated using the above formula.

[0173] Based on this, by adding the updated parameter adjustment amount to the candidate drilling parameter data before adjustment to obtain new candidate drilling parameter data, the candidate drilling parameter data can more fully traverse the path from the initial candidate drilling parameter data to the local optimal parameter data, thereby improving the accuracy of the local optimal parameter data.

[0174] Example 6: As shown in the appendix Figure 10 As shown in the figure, this invention discloses a drilling parameter optimization device based on high-precision rock cuttings mineral and mechanical properties, comprising:

[0175] The sample acquisition unit acquires several sample data, each of which includes historical mineral composition data, historical rock mechanical parameter data, historical drilling parameter data of the drilling equipment, and a corresponding historical time identifier. The historical time is the time taken by the drilling equipment to break a unit volume of rock.

[0176] The training unit introduces model loss and uses sample data to train the regression model;

[0177] The parameter optimization unit inputs all candidate drilling parameter data into the trained regression model, constructs an objective function with the goal of minimizing the output of the regression model, and optimizes the objective function based on local optimum and global optimum algorithms to determine the target drilling parameter data for the drilling equipment.

[0178] The parameter optimization unit includes:

[0179] The initial processing module obtains the parameter value range, creates multiple parameter pools for the objective function, and randomly generates candidate drilling parameter data within the parameter value range in each parameter pool;

[0180] The regression module inputs the candidate drilling parameter data generated in each parameter pool into the regression model for regression to obtain multiple candidate times.

[0181] The optimization solution module determines the globally optimal parameter data among all candidate drilling parameter data based on candidate time, and determines the locally optimal parameter data for each parameter pool among the candidate drilling parameter data in each parameter pool.

[0182] Determine whether the candidate time corresponding to the globally optimal parameter data is within a predetermined threshold or whether the number of adjustments has reached a predetermined number;

[0183] Whether or not it responds, based on the global optimal parameter data and the local optimal parameter data, the candidate drilling parameter data currently generated in each parameter pool is adjusted, and new candidate drilling parameter data is generated again in each parameter pool and re-input into the regression model;

[0184] Therefore, the currently determined global optimal parameter data is set as the target drilling parameter data for the drilling equipment.

[0185] Example 7: This embodiment of the invention discloses an electronic device, including a processor and a memory. The memory stores a computer program, which is loaded and executed by the processor to implement the steps in the drilling parameter optimization method based on high-precision rock cuttings mineral and mechanical properties.

[0186] The processor described above can be a central processing unit (CPU), a general-purpose processor, a digital signal processor (DSP), an ASIC, an FPGA, or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. It can also be a combination that implements computational functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc. The memory can include, but is not limited to, various media capable of storing computer programs, such as USB flash drives, read-only memory, portable hard drives, magnetic disks, or optical disks.

[0187] 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 implemented 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. The solutions in the embodiments of this application can be implemented in various computer languages, such as the object-oriented programming language Java and the interpreted scripting language JavaScript.

[0188] 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 flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0189] 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.

[0190] Example 8: This embodiment of the invention discloses a computer program product, including a computer program or instructions, characterized in that, when the computer program or instructions are executed by a processor, they implement the steps in a drilling parameter optimization method based on high-precision rock cuttings mineral and mechanical properties.

[0191] The above technical features constitute the preferred embodiment of the present invention, which has strong adaptability and optimal implementation effect. Unnecessary technical features can be added or removed according to actual needs to meet the requirements of different situations.

Claims

1. A drilling parameter optimization method based on high-precision rock cuttings mineral and mechanical properties, characterized in that, include: Acquire several sample data, each of which includes historical mineral composition data, historical rock mechanical parameter data, historical drilling parameter data of drilling equipment, and a corresponding historical time identifier. The historical time is the time taken by the drilling equipment to break a unit volume of rock. Introduce model loss and train the regression model using sample data; All candidate drilling parameter data are input into the trained regression model. The objective function is constructed with the goal of minimizing the output of the regression model. The objective function is then optimized based on local and global optimum algorithms to determine the target drilling parameter data for the drilling equipment.

2. The drilling parameter optimization method based on high-precision rock cuttings mineral and mechanical properties according to claim 1, characterized in that, The process of introducing model loss and training the regression model using sample data includes: A number of sample data are divided into a training sample set and a test sample set according to a certain ratio; The regression model is trained using the training sample set in conjunction with the training termination condition; The historical mineral composition data, historical rock mechanics parameter data, and historical drilling parameter data corresponding to each sample data in the test sample set are input into the regression model for regression to obtain the prediction time. The model loss is determined based on the difference between the prediction time and the historical time, and the regression model is tested and optimized based on the model loss.

3. The drilling parameter optimization method based on high-precision rock cuttings mineral and mechanical properties according to claim 1 or 2, characterized in that, The historical time for obtaining each sample data includes: Obtain multiple reference drilling depths for the drilling equipment, and the reference drilling time corresponding to each reference drilling depth; The first fitting curve is obtained by fitting multiple reference drilling depths and their corresponding reference drilling times. The initial drilling time is determined based on the first fitted curve; Historical rock-breaking volume is determined based on historical drilling depth and historical drilling diameter. The historical time corresponding to the sample data is determined by the quotient of the initial drilling time and the historical rock breaking volume.

4. The drilling parameter optimization method based on high-precision rock cuttings mineral and mechanical properties according to claim 1 or 2, characterized in that, The historical mineral composition data and historical rock mechanical parameter data obtained from each sample data include: Acquire three-dimensional images of microcores corresponding to multiple reference drilling depths, as well as reference mineral composition data corresponding to each reference drilling depth; The number of samples is obtained, and historical drilling depths are uniformly sampled based on the number of samples to obtain multiple sampled drilling depths; The data of multiple reference drilling depths and their corresponding reference mineral components are fitted to obtain a second fitting curve. Based on the multiple sampled drilling depths and the second fitting curve, the historical mineral component data corresponding to each sampled drilling depth is determined. For each reference drilling depth, the reference rock mechanics parameters corresponding to each reference drilling depth are determined based on the corresponding three-dimensional image and the corresponding reference mineral composition data. Multiple reference drilling depths and their corresponding reference rock mechanics parameters are fitted to obtain a third fitting curve. Based on multiple sampled drilling depths and the third fitting curve, the historical rock mechanics parameter data corresponding to each sampled drilling depth are determined.

5. The drilling parameter optimization method based on high-precision rock cuttings mineral and mechanical properties according to any one of claims 1 to 4, characterized in that, The optimization of the objective function based on local and global optimum algorithms to determine the target drilling parameter data for the drilling equipment includes: Obtain the parameter value range, create multiple parameter pools for the objective function, and randomly generate candidate drilling parameter data within the parameter value range in each parameter pool; The candidate drilling parameter data generated in each parameter pool are input into the regression model for regression to obtain multiple candidate times. Based on the candidate time, the globally optimal parameter data is determined among all candidate drilling parameter data, and the locally optimal parameter data for each parameter pool is determined among the candidate drilling parameter data in each parameter pool. Determine whether the candidate time corresponding to the globally optimal parameter data is within a predetermined threshold or whether the number of adjustments has reached a predetermined number; Whether or not it responds, based on the global optimal parameter data and the local optimal parameter data, the candidate drilling parameter data currently generated in each parameter pool is adjusted, and new candidate drilling parameter data is generated again in each parameter pool and re-input into the regression model; Therefore, the currently determined global optimal parameter data is set as the target drilling parameter data for the drilling equipment.

6. The drilling parameter optimization method based on high-precision rock cuttings mineral and mechanical properties according to claim 5, characterized in that, The process involves adjusting the currently generated candidate drilling parameter data in each parameter pool based on globally optimal and locally optimal parameter data, and then generating new candidate drilling parameter data in each parameter pool, including: Based on the global optimal parameter data and the local optimal parameter data, the parameter adjustment amount corresponding to each parameter pool is updated respectively; Based on the updated parameter adjustments, the candidate drilling parameter data currently generated in each parameter pool are adjusted, and new candidate drilling parameter data are generated again in each parameter pool.

7. A drilling parameter optimization device based on high-precision rock cuttings mineral and mechanical properties, applying the method described in any one of claims 1 to 6, characterized in that, include: The sample acquisition unit acquires several sample data, each of which includes historical mineral composition data, historical rock mechanical parameter data, historical drilling parameter data of the drilling equipment, and a corresponding historical time identifier. The historical time is the time taken by the drilling equipment to break a unit volume of rock. The training unit introduces model loss and uses sample data to train the regression model; The parameter optimization unit inputs all candidate drilling parameter data into the trained regression model, constructs an objective function with the goal of minimizing the output of the regression model, and optimizes the objective function based on local optimum and global optimum algorithms to determine the target drilling parameter data for the drilling equipment.

8. The drilling parameter optimization device based on high-precision rock cuttings mineral and mechanical properties according to claim 7, characterized in that, The parameter optimization unit includes: The initial processing module obtains the parameter value range, creates multiple parameter pools for the objective function, and randomly generates candidate drilling parameter data within the parameter value range in each parameter pool; The regression module inputs the candidate drilling parameter data generated in each parameter pool into the regression model for regression to obtain multiple candidate times. The optimization solution module determines the globally optimal parameter data among all candidate drilling parameter data based on candidate time, and determines the locally optimal parameter data for each parameter pool among the candidate drilling parameter data in each parameter pool. Determine whether the candidate time corresponding to the globally optimal parameter data is within a predetermined threshold or whether the number of adjustments has reached a predetermined number; Whether or not it responds, based on the global optimal parameter data and the local optimal parameter data, the candidate drilling parameter data currently generated in each parameter pool is adjusted, and new candidate drilling parameter data is generated again in each parameter pool and re-input into the regression model; Therefore, the currently determined global optimal parameter data is set as the target drilling parameter data for the drilling equipment.

9. An electronic device, characterized in that, It includes a processor and a memory, the memory storing a computer program, which is loaded and executed by the processor to implement the steps in the drilling parameter optimization method based on high-precision rock cuttings mineral and mechanical properties as claimed in any one of claims 1 to 6.

10. A computer program product, comprising a computer program or instructions, characterized in that, When the computer program or instructions are executed by the processor, they implement the steps in the drilling parameter optimization method based on high-precision rock cuttings mineral and mechanical properties as claimed in any one of claims 1 to 6.

Images