Coal mine underground near horizontal directional drilling trajectory prediction method and system

Abstract

This invention discloses a method and system for predicting near-horizontal directional drilling trajectories in coal mines, relating to the field of underground coal mine drilling. The method includes: collecting a target environmental feature set; obtaining a target drilling tool and collecting a target operational feature set; constructing a target drilling tool model and simulating it to obtain intelligent simulation results; analyzing the force information of the drilling tool, including various force types with force magnitude indicators; collecting historical drilling records and training an intelligent prediction model; inputting various force types with force magnitude indicators into the intelligent prediction model and outputting a predicted build-up rate; obtaining a preset trajectory and correcting it to obtain the predicted trajectory. This invention solves the problem of low accuracy in drilling trajectory prediction in existing technologies, which leads to non-targeted drilling tool control and affects the scientific and rational nature of drilling. It achieves the technical effect of improving the accuracy of drilling trajectory prediction results, thereby providing a basis and reference for drilling tool control, and ultimately ensuring drilling quality and efficiency.

Description

Technical Field

[0001] This invention relates to the field of underground drilling in coal mines, and more particularly to a method and system for predicting near-horizontal directional drilling trajectories in underground coal mines. Background Technology

[0002] Near-horizontal directional drilling in coal mines has become a major technical approach for geological exploration and gas drainage both domestically and internationally. Current near-horizontal directional drilling trajectory control schemes in coal mines primarily rely on the driller observing the magnitude and direction of the deviation between the actual drilling trajectory and the designed trajectory using a measurement-while-drilling (MWD) system. Based on operational experience, the tool face angle is adjusted when adding drill pipe to track the actual drilling trajectory. This method is highly experience-based and prone to problems such as deviations from the designed trajectory and excessive drill string friction, affecting drilling efficiency and quality. Furthermore, with the increasing depth and diameter of directional boreholes and the increasing complexity and diversity of drilling formations, the inherent limitations of existing sliding directional trajectory control technology and wired MWD technology are becoming increasingly apparent. Therefore, it is necessary to conduct a comprehensive analysis of the drilling tools and the environmental conditions of the actual drilling formation before drilling, and to use computer technology to predict the drilling trajectory.

[0003] However, in the existing technology, relevant technicians monitor and analyze the drilling trajectory of the drill string in real time based on the measurement while drilling system, and then make control adjustments based on historical operating experience. This has technical problems such as being greatly affected by the subjective experience of technicians, low accuracy in predicting the drilling trajectory of the drill string, and further leading to non-targeted control of the drill string, ultimately affecting the scientific and rational nature of drilling. Summary of the Invention

[0004] The purpose of this invention is to provide a method and system for predicting the trajectory of near-horizontal directional drilling in coal mines. This method addresses the technical problem in the existing technology where relevant technicians monitor and analyze the drilling trajectory of the drill bit in real time based on a measurement-while-drilling system, and then control and adjust the drill bit operation in combination with historical operating experience. This method is heavily influenced by the subjective experience of the technicians, resulting in low accuracy in predicting the drilling trajectory. Consequently, the control of the drill bit is not targeted, ultimately affecting the scientific and rational nature of the drilling process.

[0005] In view of the above problems, the present invention provides a method and system for predicting the trajectory of near-horizontal directional drilling in coal mines.

[0006] In a first aspect, the present invention provides a method for predicting the trajectory of near-horizontal directional drilling in coal mines. The method is implemented through a near-horizontal directional drilling trajectory prediction system in coal mines. The method includes: collecting data on the surrounding environment of the target drilling formation to obtain a target environment feature set; obtaining the target drilling tool of the target drilling formation and collecting a target operation feature set of the target drilling tool; constructing a target drilling tool model based on the target operation feature set and performing simulation analysis on the target drilling tool model in conjunction with the target environment feature set to obtain intelligent simulation results; performing force analysis on the target drilling tool based on the intelligent simulation results to obtain drilling tool force information, wherein the drilling tool force information includes multiple force types with force magnitude indicators; collecting historical drilling records and training an intelligent prediction model based on the historical drilling records; inputting the multiple force types with force magnitude indicators into the intelligent prediction model and outputting a predicted build-up rate; obtaining a preset trajectory of the target drilling tool and correcting the preset trajectory based on the predicted build-up rate to obtain a predicted trajectory.

[0007] Secondly, the present invention also provides a near-horizontal directional drilling trajectory prediction system for underground coal mines, used to execute the near-horizontal directional drilling trajectory prediction method for underground coal mines as described in the first aspect, wherein the system includes: a first acquisition module, used to acquire the surrounding environment of the target drilling stratum to obtain a target environment feature set; a second acquisition module, used to obtain the target drilling tool of the target drilling stratum and acquire the target operation feature set of the target drilling tool; and an intelligent simulation module, used to construct a target drilling tool model based on the target operation feature set and to perform simulation analysis on the target drilling tool model in conjunction with the target environment feature set to obtain an intelligent simulation result. The system comprises: a true result module; a force analysis module, used to perform force analysis on the target drill string based on the intelligent simulation results to obtain drill string force information, wherein the drill string force information includes multiple force types with force magnitude indicators; a model training module, used to collect historical drilling records and train an intelligent prediction model based on the historical drilling records; an intelligent prediction module, used to input the multiple force types with force magnitude indicators into the intelligent prediction model and output the predicted build-up rate; and an intelligent acquisition module, used to obtain the preset trajectory of the target drill string and correct the preset trajectory based on the predicted build-up rate to obtain the predicted trajectory.

[0008] One or more technical solutions provided in this invention have at least the following technical effects or advantages:

[0009] By collecting data on the surrounding environment of the target drilling formation, a target environment feature set is obtained; the target drilling tool of the target drilling formation is obtained, and a target operation feature set of the target drilling tool is collected; a target drilling tool model is constructed based on the target operation feature set, and the target drilling tool model is simulated and analyzed in conjunction with the target environment feature set to obtain intelligent simulation results; the target drilling tool is subjected to stress analysis based on the intelligent simulation results to obtain drill tool stress information, wherein the drill tool stress information includes multiple stress types with stress magnitude indicators; historical drilling records are collected, and an intelligent prediction model is trained based on the historical drilling records; the multiple stress types with stress magnitude indicators are input into the intelligent prediction model, and a predicted build-up rate is output; a preset trajectory of the target drilling tool is obtained, and the preset trajectory is corrected based on the predicted build-up rate to obtain a predicted trajectory. By collecting and analyzing the surrounding environment of the target drilling formation and the operation characteristics of the target drilling tool, the technical goal of providing a simulation basis for subsequent simulation of the drilling process is achieved, thereby improving the technical effect of drilling simulation reliability. By analyzing the simulation results of the drilling tool, the various types and magnitudes of forces acting on the tool during drilling are identified, providing a foundation of force information for intelligent analysis of the drill tool's build-up rate. This achieves the technical effect of improving the system's targetedness, accuracy, and reliability in intelligently determining the drill tool's build-up rate. Intelligent analysis yields the predicted build-up rate of the target drill tool, which in turn analyzes the preset trajectory of the drill tool to obtain the actual trajectory prediction result, thus improving the accuracy of the drill tool trajectory prediction. This achieves the goal of intelligent prediction of the drill tool trajectory, improving the accuracy of the drilling trajectory prediction result, providing a basis and reference for drill tool control, and ultimately ensuring drilling quality and efficiency.

[0010] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention, it can be implemented according to the contents of the specification. In order to make the above and other objects, features and advantages of the present invention more obvious and understandable, specific embodiments of the present invention are described below. Attached Figure Description

[0011] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely exemplary. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0012] Figure 1 This is a flowchart illustrating the method for predicting near-horizontal directional drilling trajectories in coal mines according to the present invention.

[0013] Figure 2This is a flowchart illustrating the process of training the intelligent prediction model in the near-horizontal directional drilling trajectory prediction method for coal mines according to the present invention.

[0014] Figure 3 This is a flowchart illustrating the process of selecting multiple integrated prediction models to obtain the intelligent prediction model in the coal mine near-horizontal directional drilling trajectory prediction method of the present invention.

[0015] Figure 4 This is a schematic diagram of the process for evaluating the prediction effect in the near-horizontal directional drilling trajectory prediction method for coal mines of the present invention;

[0016] Figure 5 This is a schematic diagram of the structure of the near-horizontal directional drilling trajectory prediction system for coal mines according to the present invention.

[0017] Explanation of reference numerals in the attached figures:

[0018] The system consists of a first acquisition module M100, a second acquisition module M200, an intelligent simulation module M300, a force analysis module M400, a model training module M500, an intelligent prediction module M600, and an intelligent acquisition module M700. Detailed Implementation

[0019] This invention provides a method and system for predicting near-horizontal directional drilling trajectories in coal mines. It solves the technical problem in existing technologies where technicians rely on real-time monitoring and analysis of the drill string's trajectory using a measurement-while-drilling (MWD) system, combined with historical operational experience, to control and adjust drill string operations. This results in significant reliance on the subjective experience of technicians, low accuracy in drill string trajectory prediction, and ultimately, a lack of targeted drill string control, impacting the scientific and rational nature of drilling. The invention achieves intelligent prediction of drill string trajectories, improving the accuracy of prediction results and providing a foundation for drill string control, ultimately ensuring drilling quality and efficiency.

[0020] The acquisition, storage, use, and processing of data in the technical solution of this invention all comply with the relevant provisions of national laws and regulations.

[0021] The technical solutions of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. It should be understood that the present invention is not limited to the exemplary embodiments described herein. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention. It should also be noted that, for ease of description, only the parts related to the present invention are shown in the accompanying drawings, not all of them.

[0022] Example 1

[0023] Please see the appendix Figure 1 This invention provides a method for predicting the trajectory of near-horizontal directional drilling in coal mines. The method is applied to a near-horizontal directional drilling trajectory prediction system in coal mines, and specifically includes the following steps:

[0024] Step S100: Collect the surrounding environment of the target drilling stratum to obtain the target environment feature set;

[0025] Specifically, the near-horizontal directional drilling trajectory prediction method for coal mines is applied to a near-horizontal directional drilling trajectory prediction system in coal mines. It can collect features of the surrounding environment of the target drilling stratum to provide a data foundation for subsequent drilling operation simulation. The target drilling stratum refers to any stratum in which the near-horizontal directional drilling trajectory prediction system is used to predict the actual drilling trajectory of the drill bit. First, the surrounding environmental information of the target drilling stratum is collected to obtain a target environmental feature set corresponding to the target drilling stratum. For example, geographical environmental information such as latitude and longitude and altitude of a drilling stratum is collected; air environmental information such as temperature, humidity, wind speed, and wind force around the drilling stratum is collected; and relevant underground environmental information such as the structural layers, thickness and composition of each layer, and groundwater content of the drilling stratum is collected to obtain the environmental characteristics of the drilling stratum. By collecting the relevant environmental features of the target drilling stratum and obtaining the target environmental feature set, the technical goal of providing realistic and effective simulation basis data for subsequent drilling operation simulation of the drill bit in this environment is achieved, thereby improving the reliability of the simulation.

[0026] Step S200: Obtain the target drilling tool for the target formation and collect the target operation feature set of the target drilling tool;

[0027] Step S300: Construct a target drilling tool model based on the target operation feature set, and perform simulation analysis on the target drilling tool model in conjunction with the target environment feature set to obtain intelligent simulation results;

[0028] Specifically, before conducting drilling simulation, the drilling tool to be used in this drilling operation is first determined, i.e., the target drilling tool is obtained. Then, the target drilling tool is analyzed to obtain its characteristic information during drilling operations, i.e., the target operation feature set is obtained. Finally, a simulation model of the target drilling tool is constructed, and a simulated drilling environment for the target drilling formation is constructed by combining the drilling operation characteristics of the target drilling tool and the drilling environment characteristics. The intelligent simulation results are then obtained. An example is using MATLAB to simulate the drilling process of the drilling tool and record the relevant simulation data.

[0029] Step S400: Perform stress analysis on the target drill bit based on the intelligent simulation results to obtain drill bit stress information, wherein the drill bit stress information includes various stress types with stress magnitude indicators;

[0030] Specifically, after constructing the corresponding simulation environment and performing drilling simulation, the intelligent simulation results are obtained. Further, the intelligent simulation results are analyzed to obtain the force situation of the target drill bit during the drilling process, such as the gravity from underground structural layers and groundwater, and the frictional force from sand and gravel in the underground structural layers. By analyzing the force on the target drill bit, the force information of the drill bit is obtained. This force information includes various force types with magnitude indicators. By analyzing the simulation results of the drill bit, the various types and magnitudes of forces experienced by the drill bit during drilling are clarified, providing a force information basis for intelligent analysis of the drill bit build-up rate, thus achieving the technical effect of improving the targeting, accuracy, and reliability of the system's intelligent determination of the drill bit build-up rate.

[0031] Step S500: Collect historical drilling records and train an intelligent prediction model based on the historical drilling records;

[0032] Further details are attached. Figure 2 As shown, step S500 of the present invention includes:

[0033] Step S510: Extract the first historical record from the historical drilling records;

[0034] Step S520: Obtain the first historical force information from the first historical record, and preprocess the first historical force information to obtain the first data group;

[0035] Step S530: Analyze the first data group and obtain the first data group identifier based on the analysis results, wherein the first data group identifier refers to the first theoretical slope.

[0036] Step S540: Use the first data set and the first theoretical slope as training data to train the intelligent prediction model.

[0037] Furthermore, the present invention includes the following steps:

[0038] Step S-before 541: Sequentially obtain the first historical preset trajectory and the first historical actual drilling trajectory from the first historical record;

[0039] Step S-before 542: Compare and analyze the first historical preset trajectory and the first historical actual drilling trajectory to obtain the comparison result of the first historical trajectory;

[0040] Step S-before 543: Match the first actual slope rate based on the comparison result of the first historical trajectory;

[0041] Step S-before 544: Adjust the first theoretical slope rate based on the first actual slope rate to obtain the first slope rate.

[0042] Further details are attached. Figure 3 As shown, step S540 of the present invention includes:

[0043] Step S541: Train the first support vector machine based on the training data;

[0044] Step S542: Train a first recurrent neural network based on the training data;

[0045] Step S543: Train the first gradient boosting decision tree based on the training data;

[0046] Step S544: Based on the principle of ensemble learning, build models for the first support vector machine, the first recurrent neural network, and the first gradient boosting decision tree to obtain multiple ensemble prediction models;

[0047] Step S545: Filter the multiple integrated prediction models to obtain the intelligent prediction model.

[0048] Furthermore, the present invention includes the following steps:

[0049] Step S5451: Extract any one of the multiple ensemble prediction models and obtain the primary learner and meta-learner of the arbitrary ensemble prediction model;

[0050] Step S5452: Train the primary learner based on the training data to make predictions and obtain primary prediction results;

[0051] Step S5453: Use the primary prediction result as the input data of the meta-learner to obtain the output data of the meta-learner;

[0052] Step S5454: Use the output data of the meta-learner as the prediction result of any one of the ensemble prediction models.

[0053] Specifically, historical drilling records are collected, and an intelligent prediction model is trained based on these records to intelligently analyze and predict the build-up rate of the target drill bit during the drilling operation.

[0054] First, each historical drilling record is analyzed sequentially. For example, data from a random historical drilling record is analyzed, specifically the historical stress information of the drill string in the first historical record, to obtain the first historical stress information. Then, the first historical stress information is preprocessed, removing force types with forces less than a certain threshold or whose impact on the drill string build-up rate is less than a certain threshold, retaining only the required force types and their corresponding stress magnitudes, thus forming the first data group. Next, based on the first data group, the drill string build-up rate record data from the current historical drilling operation is extracted and analyzed, and marked into the first data group, obtaining the first data group identifier. The first data group identifier refers to the first theoretical build-up rate. Finally, the first data group and the first theoretical build-up rate are used as training data to train the intelligent prediction model.

[0055] Before training and determining the intelligent prediction model using the first data set and the first theoretical build-up rate as training data, the first theoretical build-up rate is adjusted and corrected to improve the model training accuracy. First, the recorded data of the first historical preset trajectory and the first historical actual drilling trajectory in the first historical record are obtained sequentially. The first historical preset trajectory refers to the drilling trajectory that the drill string designed by relevant technicians should execute in this historical drilling operation. The first historical actual drilling trajectory refers to the actual drilling trajectory generated by the drill string during this historical drilling operation. Then, the first historical preset trajectory and the first historical actual drilling trajectory are compared and analyzed to obtain the deviation information of the first historical actual drilling trajectory based on the first historical preset trajectory, i.e., the first historical trajectory comparison result is obtained. Then, the first actual build-up rate is matched based on the first historical trajectory comparison result. For example, based on historical drilling data, different levels of deviation correspond to different levels of build-up rates, thereby generating a list of correspondences between trajectory comparison deviation data and build-up rates, providing a list information basis for quickly determining the actual build-up rate. Finally, the first theoretical build-up rate is adjusted based on the first actual build-up rate to obtain the first build-up rate. The first slope is used to train the intelligent prediction model.

[0056] Furthermore, the first data set and the adjusted first slope are used as training data to train the intelligent prediction model. First, a first support vector machine is trained using the training data. Then, a first recurrent neural network is trained using the training data. Next, a first gradient boosting decision tree is trained using the training data. Following this, based on the principles of ensemble learning, models are built using the first support vector machine, the first recurrent neural network, and the first gradient boosting decision tree to obtain multiple ensemble prediction models. For example, the first support vector machine and the first recurrent neural network can be used as the first layer of prediction structure (i.e., as the primary learner), while the first gradient boosting decision tree can be used as the second layer of prediction structure (i.e., as the meta-learner). Through different model construction schemes, multiple ensemble models are obtained. Finally, predictive analysis is performed on the multiple ensemble prediction models, and the model with the highest prediction accuracy is selected as the intelligent prediction model.

[0057] When analyzing and screening the multiple ensemble prediction models, firstly, one ensemble prediction model is randomly extracted, and its primary learner and meta-learner are obtained. Then, the primary learner is trained and predicted based on the training data to obtain a primary prediction result. Next, the primary prediction result is used as the input data for the meta-learner to obtain its output data. The output data of the meta-learner serves as the prediction result of the arbitrary ensemble prediction model. Finally, the prediction results of each model in the multiple ensemble prediction models are compared, and the model with the smallest deviation between its prediction accuracy and the actual slope rate is determined and selected as the intelligent prediction model.

[0058] By intelligently training and integrating models, and analyzing the prediction accuracy of each integrated model, the model with the highest prediction accuracy is selected as the final intelligent prediction model. This provides an accurate and reliable model basis for subsequent intelligent prediction of the build-up rate of the target drill string, thereby improving the technical effect of improving the accuracy of build-up rate prediction and thus improving the accuracy of trajectory prediction.

[0059] Step S600: Input the various force types with force magnitude identifiers into the intelligent prediction model and output the predicted slope rate;

[0060] Step S700: Obtain the preset trajectory of the target drill string, and correct the preset trajectory based on the predicted build-up rate to obtain the predicted trajectory.

[0061] Specifically, the various force types with force magnitude indicators are input into the intelligent prediction model. Through the prediction analysis of the two-layer prediction structure within the intelligent prediction model—namely, the primary learner and the meta-learner—the corresponding predicted build-up rate is obtained. Further, the preset trajectory of the target drill string is obtained. This preset trajectory refers to the drilling trajectory requirements designed by relevant technicians based on a comprehensive analysis of actual conditions before drilling operations. Finally, the preset trajectory is corrected based on the predicted build-up rate, i.e., the predicted build-up rate is multiplied by the preset trajectory to obtain the predicted trajectory. By obtaining the predicted build-up rate of the target drill string through intelligent analysis, and then analyzing the preset trajectory of the drill string, the actual trajectory prediction result of the drill string is obtained, achieving the technical effect of improving the accuracy of the drill string trajectory prediction result.

[0062] Further details are attached. Figure 4 As shown, the present invention further includes step S800:

[0063] Step S810: Acquire the actual drilling trajectory of the target drill bit;

[0064] Step S820: Compare the actual drilling trajectory with the predicted trajectory to obtain the prediction accuracy;

[0065] Step S830: Calculate the prediction duration and obtain the prediction efficiency based on the prediction duration;

[0066] Furthermore, the present invention includes the following steps:

[0067] Step S831: Obtain the first time, wherein the first time refers to the time when the various force types with force magnitude identifiers are input into the intelligent prediction model;

[0068] Step S832: Obtain the second time, wherein the second time refers to the time when the predicted trajectory is obtained;

[0069] Step S833: Calculate the time difference between the first time and the second time to obtain the predicted duration.

[0070] Step S840: Perform a weighted calculation on the prediction accuracy and the prediction efficiency to obtain a prediction performance evaluation.

[0071] Specifically, after obtaining the predicted inclination rate based on the intelligent prediction model and correcting the preset trajectory based on the predicted inclination rate to obtain the predicted trajectory, in order to objectively, comprehensively, and scientifically evaluate the prediction effect of the system, the actual drilling trajectory information of the target drill bit is collected using relevant intelligent devices, i.e., the actual drilling trajectory is obtained. Then, the actual drilling trajectory is compared with the predicted trajectory obtained by the intelligent analysis of the system to obtain the degree of deviation between the trajectory obtained by the intelligent system and the actual drilling trajectory, i.e., to determine the prediction accuracy. Next, when the various force types with force magnitude indicators are input into the intelligent prediction model, the system automatically obtains the first time, and when the system intelligently analyzes and obtains the predicted trajectory, it obtains the corresponding time information again, i.e., the second time. Then, the time difference can be calculated based on the two times, i.e., the prediction duration of the predicted trajectory obtained by the intelligent analysis of the system, and this prediction duration is used as the prediction efficiency of the system. Finally, the prediction accuracy and the prediction efficiency are weighted and calculated to obtain the evaluation result of the intelligent prediction effect of the system, i.e., the prediction effect evaluation. An exemplary approach is to assign a weight of 0.5 to both prediction accuracy and prediction efficiency, and then combine the two calculation results to obtain a weighted evaluation result of the prediction effect, thereby achieving the goal of a comprehensive and objective evaluation of the system's prediction performance.

[0072] In summary, the method for predicting near-horizontal directional drilling trajectories in coal mines provided by this invention has the following technical advantages:

[0073] By collecting data on the surrounding environment of the target drilling formation, a target environment feature set is obtained; the target drilling tool of the target drilling formation is obtained, and a target operation feature set of the target drilling tool is collected; a target drilling tool model is constructed based on the target operation feature set, and the target drilling tool model is simulated and analyzed in conjunction with the target environment feature set to obtain intelligent simulation results; the target drilling tool is subjected to stress analysis based on the intelligent simulation results to obtain drill tool stress information, wherein the drill tool stress information includes multiple stress types with stress magnitude indicators; historical drilling records are collected, and an intelligent prediction model is trained based on the historical drilling records; the multiple stress types with stress magnitude indicators are input into the intelligent prediction model, and a predicted build-up rate is output; a preset trajectory of the target drilling tool is obtained, and the preset trajectory is corrected based on the predicted build-up rate to obtain a predicted trajectory. By collecting and analyzing the surrounding environment of the target drilling formation and the operation characteristics of the target drilling tool, the technical goal of providing a simulation basis for subsequent simulation of the drilling process is achieved, thereby improving the technical effect of drilling simulation reliability. By analyzing the simulation results of the drilling tool, the various types and magnitudes of forces acting on the tool during drilling are identified, providing a foundation of force information for intelligent analysis of the drill tool's build-up rate. This achieves the technical effect of improving the system's targetedness, accuracy, and reliability in intelligently determining the drill tool's build-up rate. Intelligent analysis yields the predicted build-up rate of the target drill tool, which in turn analyzes the preset trajectory of the drill tool to obtain the actual trajectory prediction result, thus improving the accuracy of the drill tool trajectory prediction. This achieves the goal of intelligent prediction of the drill tool trajectory, improving the accuracy of the drilling trajectory prediction result, providing a basis and reference for drill tool control, and ultimately ensuring drilling quality and efficiency.

[0074] Example 2

[0075] Based on the same inventive concept as the near-horizontal directional drilling trajectory prediction method in the aforementioned embodiments, this invention also provides a near-horizontal directional drilling trajectory prediction system for coal mines. Please refer to the appendix. Figure 5 The system includes:

[0076] The first acquisition module M100 is used to acquire the surrounding environment of the target drilling stratum and obtain the target environment feature set;

[0077] The second acquisition module M200 is used to obtain the target drilling tool of the target drilling formation and to acquire the target operation feature set of the target drilling tool;

[0078] The intelligent simulation module M300 is used to construct a target drilling tool model based on the target operation feature set, and to perform simulation analysis on the target drilling tool model in combination with the target environment feature set to obtain intelligent simulation results;

[0079] The stress analysis module M400 is used to perform stress analysis on the target drill bit based on the intelligent simulation results to obtain drill bit stress information, wherein the drill bit stress information includes a variety of stress types with stress magnitude indicators;

[0080] The model training module M500 is used to collect historical drilling records and train an intelligent prediction model based on the historical drilling records.

[0081] The intelligent prediction module M600 is used to input the various force types with force magnitude identifiers into the intelligent prediction model and output the predicted slope rate.

[0082] The intelligent acquisition module M700 is used to acquire the preset trajectory of the target drill string and correct the preset trajectory based on the predicted build-up rate to obtain the predicted trajectory.

[0083] Furthermore, the model training module M500 in the system is also used for:

[0084] Extract the first historical record from the historical drilling records;

[0085] Obtain the first historical force information from the first historical record, and preprocess the first historical force information to obtain the first data group;

[0086] The first data set is analyzed, and the identifier of the first data set is obtained based on the analysis results, wherein the identifier of the first data set refers to the first theoretical slope.

[0087] The intelligent prediction model is trained by using the first data set and the first theoretical slope as training data.

[0088] Furthermore, the model training module M500 in the system is also used for:

[0089] The first historical preset trajectory and the first historical actual drilling trajectory in the first historical record are obtained sequentially.

[0090] The first historical preset trajectory and the first historical actual drilling trajectory are compared and analyzed to obtain the comparison result of the first historical trajectory;

[0091] Based on the comparison results of the first historical trajectory, the first actual slope rate is matched;

[0092] The first theoretical slope is adjusted based on the first actual slope to obtain the first slope.

[0093] Furthermore, the model training module M500 in the system is also used for:

[0094] The first support vector machine is trained based on the training data;

[0095] A first recurrent neural network is trained based on the training data;

[0096] A first gradient boosting decision tree is trained based on the training data;

[0097] Based on the principle of ensemble learning, models are built for the first support vector machine, the first recurrent neural network, and the first gradient boosting decision tree to obtain multiple ensemble prediction models.

[0098] The intelligent prediction model is obtained by filtering the multiple integrated prediction models.

[0099] Furthermore, the model training module M500 in the system is also used for:

[0100] Extract any one of the multiple ensemble prediction models and obtain the primary learner and meta-learner of the arbitrary ensemble prediction model;

[0101] The primary learner is trained and predicted based on the training data to obtain a primary prediction result.

[0102] The primary prediction results are used as input data for the meta-learner to obtain the output data of the meta-learner;

[0103] The output data of the meta-learner is used as the prediction result of any one of the ensemble prediction models.

[0104] Furthermore, the system also includes an intelligent evaluation module, wherein the intelligent evaluation module is used for:

[0105] The actual drilling trajectory of the target drill bit was obtained;

[0106] The actual drilling trajectory is compared with the predicted trajectory to obtain the prediction accuracy;

[0107] The prediction duration is calculated, and the prediction efficiency is obtained based on the prediction duration;

[0108] The prediction accuracy and prediction efficiency are weighted and calculated to obtain a prediction performance evaluation.

[0109] Furthermore, the system also includes an intelligent evaluation module, wherein the intelligent evaluation module is used for:

[0110] The first time is obtained, where the first time refers to the time when the various force types with force magnitude identifiers are input into the intelligent prediction model;

[0111] Obtain the second time, wherein the second time refers to the time when the predicted trajectory is obtained;

[0112] The predicted duration is obtained by calculating the time difference between the first time and the second time.

[0113] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Figure 1 The method and specific examples for predicting the trajectory of near-horizontal directional drilling in coal mines in Embodiment 1 are also applicable to the near-horizontal directional drilling trajectory prediction system in coal mines in this embodiment. Through the foregoing detailed description of the method for predicting the trajectory of near-horizontal directional drilling in coal mines, those skilled in the art can clearly understand the near-horizontal directional drilling trajectory prediction system in this embodiment. Therefore, for the sake of brevity, it will not be described in detail here. As for the apparatus disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple; relevant details can be found in the method section.

[0114] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0115] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of this invention and its equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for predicting the trajectory of a near-horizontal directional drilling in a coal mine, characterized in that, include: The surrounding environment of the target drilling stratum is collected to obtain the target environmental feature set; Obtain the target drilling tool for the target formation, and collect the target operation feature set of the target drilling tool; A target drilling tool model is constructed based on the target operation feature set, and the target drilling tool model is simulated and analyzed in conjunction with the target environment feature set to obtain intelligent simulation results; Based on the intelligent simulation results, the target drilling tool is subjected to stress analysis to obtain the stress information of the drilling tool, wherein the stress information of the drilling tool includes a variety of stress types with stress magnitude indicators; Collect historical drilling records and train an intelligent prediction model based on the historical drilling records; The various force types with force magnitude indicators are input into the intelligent prediction model, and the predicted slope rate is output. Obtain the preset trajectory of the target drill string, and correct the preset trajectory based on the predicted build-up rate to obtain the predicted trajectory; The step of collecting historical drilling records and training an intelligent prediction model based on those records includes: Extract the first historical record from the historical drilling records; Obtain the first historical force information from the first historical record, and preprocess the first historical force information to obtain the first data group; The first data set is analyzed, and the identifier of the first data set is obtained based on the analysis results, wherein the identifier of the first data set refers to the first theoretical slope. The intelligent prediction model is trained by using the first data set and the first theoretical slope as training data.

2. The method of claim 1, wherein, Before training the intelligent prediction model using the first data set and the first theoretical slope as training data, the method further includes: The first historical preset trajectory and the first historical actual drilling trajectory in the first historical record are obtained sequentially. The first historical preset trajectory and the first historical actual drilling trajectory are compared and analyzed to obtain the comparison result of the first historical trajectory; Based on the comparison results of the first historical trajectory, the first actual slope rate is matched; The first theoretical slope is adjusted based on the first actual slope to obtain the first slope.

3. The method according to claim 2, characterized in that, The step of using the first data set and the first theoretical slope as training data to train the intelligent prediction model includes: The first support vector machine is trained based on the training data; A first recurrent neural network is trained based on the training data; A first gradient boosting decision tree is trained based on the training data; Based on the principle of ensemble learning, models are built for the first support vector machine, the first recurrent neural network, and the first gradient boosting decision tree to obtain multiple ensemble prediction models. The intelligent prediction model is obtained by filtering the multiple integrated prediction models.

4. The method according to claim 3, characterized in that, The process of filtering the multiple integrated prediction models to obtain the intelligent prediction model includes: Extract any one of the multiple ensemble prediction models and obtain the primary learner and meta-learner of the arbitrary ensemble prediction model; The primary learner is trained and predicted based on the training data to obtain a primary prediction result. The primary prediction results are used as input data for the meta-learner to obtain the output data of the meta-learner; The output data of the meta-learner is used as the prediction result of any one of the ensemble prediction models.

5. The method according to claim 1, characterized in that, After obtaining the preset trajectory of the target drill string and correcting the preset trajectory based on the predicted build-up rate to obtain the predicted trajectory, the method further includes: The actual drilling trajectory of the target drill bit was obtained; The actual drilling trajectory is compared with the predicted trajectory to obtain the prediction accuracy; The prediction duration is calculated, and the prediction efficiency is obtained based on the prediction duration; The prediction accuracy and prediction efficiency are weighted and calculated to obtain a prediction performance evaluation.

6. The method according to claim 5, characterized in that, The calculation yields the predicted duration, including: The first time is obtained, where the first time refers to the time when the various force types with force magnitude identifiers are input into the intelligent prediction model; Obtain the second time, wherein the second time refers to the time when the predicted trajectory is obtained; The predicted duration is obtained by calculating the time difference between the first time and the second time.

7. A near-horizontal directional drilling trajectory prediction system for underground coal mines, characterized in that, The system is used to perform the near-horizontal directional drilling trajectory prediction method for coal mines according to any one of claims 1 to 6, the system comprising: The first acquisition module is used to acquire the surrounding environment of the target drilling stratum and obtain the target environment feature set; The second acquisition module is used to obtain the target drilling tool of the target drilling formation and to acquire the target operation feature set of the target drilling tool; The intelligent simulation module is used to construct a target drilling tool model based on the target operation feature set, and to perform simulation analysis on the target drilling tool model in combination with the target environment feature set to obtain intelligent simulation results; The stress analysis module is used to perform stress analysis on the target drilling tool based on the intelligent simulation results to obtain the stress information of the drilling tool, wherein the stress information of the drilling tool includes a variety of stress types with stress magnitude indicators; The model training module is used to collect historical drilling records and train an intelligent prediction model based on the historical drilling records. The intelligent prediction module is used to input the various force types with force magnitude identifiers into the intelligent prediction model and output the predicted slope rate. The intelligent acquisition module is used to acquire the preset trajectory of the target drill string and correct the preset trajectory based on the predicted build-up rate to obtain the predicted trajectory.

Images