Coal mine roof disaster prevention and control method and device based on retrieval enhancement generation

Abstract

The invention provides a coal mine roof disaster prevention and control method and device based on retrieval enhancement generation, and the method comprises the steps: obtaining first information, inputted by a user, of a target coal mine; based on the first information of the target coal mine, a coal industry roof disaster prevention and control information base is retrieved, and retrieval information of the target coal mine is obtained; combining the first information of the target coal mine and the retrieval information of the target coal mine to obtain combined information; inputting the combined information into a large language model to obtain supplementary information of the target coal mine; based on the supplementary information of the target coal mine, identifying a main control factor of the target coal mine through a roof stability main control factor identification model; and based on the main control factors of the target coal mine and the coal mine support technology system, obtaining a support scheme under the main control factors of the target coal mine. The identification of main control factors of the stability of the roof and the acquisition of a support scheme are realized, so as to assist in guiding underground support engineering.

Description

Technical Field

[0001] The present application relates to the technical field of coal mine roof disaster prevention and control, and in particular to a coal mine roof disaster prevention and control method, device, electronic equipment and storage medium based on retrieval enhancement generation. Background Art

[0002] Since the 21st century, the intensity and depth of coal mining in my country have increased significantly, and mine disasters such as coal mine roof collapse, support failure and rock burst have occurred in combination. The prevention and control of deep mine disasters has entered the deep water area, and the bottleneck problem of high-quality prevention and control has become prominent. The construction of intelligent coal mines across the country has entered a stage of rapid advancement. The first batch of intelligent coal mine construction demonstration projects have achieved initial results, and significant progress has been made in information infrastructure, intelligent tunneling, intelligent mining, intelligent disaster prevention, intelligent washing and intelligent transportation, providing replicable experience and models for the construction of intelligent coal mines.

[0003] However, the geological conditions of coal-bearing strata in my country are complex, and there is a common feature of "one mine is the same" or even "one face is the same". How to intelligently apply these replicable experiences and models to each working face or tunnel requires a large amount of coal seam roof data, support design schemes, mine pressure monitoring and tunnel deformation monitoring data, as well as various support failure cases. In particular, during the intelligent construction of coal mines, roof accidents occurred frequently, which is contrary to the concept of safe and efficient coal mining. Therefore, the research and development of intelligent prevention and control technology and systems for coal mine roof disasters is imminent.

[0004] Through decades of coal mining experience, my country has accumulated a large amount of historical data on roof support and a variety of support failure cases, and has done a lot of preliminary work for the characteristics of roof zoning, disaster-causing factors, zoning and classification prevention and control technologies and key parameters, as well as applicable conditions. However, these data and results are still in a fragmented state and have not been systematically sorted and summarized. In order to make these valuable experiences and research results available for subsequent projects, it is necessary to fully carry out systematic and detailed data collection and results summary and refinement to provide favorable support for the construction of intelligent prevention and control technology for coal mine roof disasters. Summary of the invention

[0005] The present application aims to solve one of the technical problems in the related art at least to some extent.

[0006] To this end, the first purpose of this application is to propose a coal mine roof disaster prevention and control method based on retrieval enhanced generation, so as to realize intelligent prevention and control in coal mine roof disaster scenarios, quickly identify the main controlling factors of roof stability and obtain corresponding support plans.

[0007] The second purpose of this application is to propose a coal mine roof disaster prevention and control device based on retrieval enhancement generation.

[0008] The third objective of the present application is to provide an electronic device.

[0009] A fourth objective of the present application is to provide a computer-readable storage medium.

[0010] A fifth object of the present application is to provide a computer program product.

[0011] To achieve the above-mentioned purpose, the first embodiment of the present application proposes a coal mine roof disaster prevention and control method based on retrieval enhancement generation, comprising:

[0012] The method for preventing and controlling coal mine roof disasters based on retrieval enhancement generation is characterized by comprising the following steps:

[0013] Acquire the first information of the target coal mine input by the user;

[0014] Based on the first information of the target coal mine, a coal industry roof disaster prevention and control information database is searched to obtain the search information of the target coal mine; the coal industry roof disaster prevention and control information database includes multiple data sources related to roof disasters and support prevention and control in the coal industry and support schemes under different roof conditions;

[0015] Merging the first information of the target coal mine and the search information of the target coal mine to obtain merged information; and inputting the merged information into a large language model to obtain supplementary information of the target coal mine;

[0016] Based on the supplementary information of the target coal mine, the main controlling factors of the target coal mine are identified through a roof stability main controlling factor identification model;

[0017] Based on the main controlling factors of the target coal mine and the coal mine support technology system, a support plan under the main controlling factors of the target coal mine is obtained; the coal mine support technology system includes support materials under different roof conditions and support methods and parameters of each support material.

[0018] In some implementations, searching a coal industry roof disaster prevention and control information database based on the first information of the target coal mine includes:

[0019] Based on the first information of the target coal mine, the coal industry roof disaster prevention and control information database is retrieved through the knowledge label index system of the coal industry roof disaster prevention and control information database; the knowledge label index system includes top-level classification, subdivided labels and specific attributes, and the knowledge label index system includes three dimensions, which are data source, influencing factors and prevention parameters.

[0020] In some implementations, the method for constructing the roof stability main control factor identification model includes:

[0021] Obtain text information related to roof disasters and support prevention in the coal industry;

[0022] The text information is processed by natural language processing technology to obtain integrated roof disaster and support prevention information;

[0023] By using data mining and analysis technology, multiple key factors affecting roof stability are extracted from the integrated roof disaster and support prevention information;

[0024] Obtaining the contribution of each of the multiple key factors;

[0025] Based on the multiple key factors and the contribution of each key factor, a roof stability main control factor identification model is constructed.

[0026] It can be understood that different key factors are used as labels, and training samples are constructed based on the coal mine characteristics extracted from the coal mine information in different scenarios corresponding to each label and the contribution of each key factor. The target recognition model is trained by the training samples to obtain the recognition model of the main controlling factors of roof stability.

[0027] In some implementations, obtaining the contribution of each of the multiple key factors includes:

[0028] Given a roof stability data, the roof stability evaluation index data calculated under the above key factors is recorded as D = [(x1, y1), (x2, y2), ..., (x n ,y n )], where x is an m-dimensional vector, i.e., it contains m key factors, and y is the roof stability index data;

[0029] Suppose a constructed classification and regression tree CART regression tree has m leaves to divide the key factor x of the input space into m basic units R1, R2, ..., R m , the CART objective minimizes the mean square error mse formula as follows:

[0030]

[0031] Among them, c m represents the predicted value of the mth leaf;

[0032] The predicted value of the roof stability index is set to the mean of the training set elements contained in the leaves, that is:

[0033] C m =ave(y i |x i ∈leaf m )

[0034] A heuristic method is adopted to traverse all the key factor variables and split positions, and the case with the smallest sum of mse of the leaf node is selected as the partition; the jth feature and its value s are selected as the split variable and split point, and the split variable and split point divide the input space of the parent node into two; in each partition, the split variable and split point are selected so that the mse of the main control factor identification model of roof stability on the training set is minimized, that is, the sum of mse of each leaf is minimized;

[0035] After determining the segmentation key factor j, its information gain is calculated, and the negative logarithm of mse is used for weighting to obtain the importance ranking of the multiple key factors, and the contribution of each key factor is obtained according to the importance ranking of the multiple key factors.

[0036] In some implementations, the obtaining of a support scheme under the main control factors of the target coal mine based on the main control factors of the target coal mine and the coal mine support technology system includes:

[0037] Determining key support parameters of the target coal mine based on the main control factors of the target coal mine;

[0038] Based on the coal mine support technology system, determine the target support scheme template;

[0039] Based on the key support parameters of the target coal mine, the support parameters of the target support scheme template are adjusted to obtain the support scheme under the main control factors of the target coal mine.

[0040] In some implementations, the method for constructing the coal mine support technology system includes:

[0041] Obtain multi-dimensional data of different mines, including geological conditions, physical and mechanical parameters of surrounding rocks, laws of mine pressure manifestation, displacement data, and successful and failed support cases;

[0042] Based on the multi-dimensional data, extract the causes of roof accidents and key support parameters;

[0043] Based on the causes of roof accidents and key support parameters, a zoning and classified coal mine support technology system under diverse roof conditions is constructed.

[0044] In some implementations, after obtaining the support scheme under the main control factors of the target coal mine, the method includes:

[0045] The support plan of the target coal mine is added to the coal industry roof disaster prevention and control information database.

[0046] To achieve the above-mentioned purpose, the second embodiment of the present application proposes a coal mine roof disaster prevention and control device based on retrieval enhancement generation, comprising:

[0047] A user information acquisition module, used to acquire first information of a target coal mine input by a user;

[0048] A related knowledge retrieval module is used to retrieve a coal industry roof disaster prevention and control information database based on the first information of the target coal mine to obtain the retrieval information of the target coal mine; the coal industry roof disaster prevention and control information database includes multiple data sources related to roof disasters and support prevention and control in the coal industry and support schemes under different roof conditions;

[0049] A model question-answering module is used to merge the first information of the target coal mine and the search information of the target coal mine to obtain merged information; and input the merged information into a large language model to obtain supplementary information of the target coal mine;

[0050] A main control factor identification module, used to identify the main control factors of the target coal mine based on the supplementary information of the target coal mine and through a roof stability main control factor identification model;

[0051] A support scheme acquisition module is used to acquire a support scheme under the main control factors of the target coal mine based on the main control factors of the target coal mine and the coal mine support technology system; the coal mine support technology system includes support materials under different roof conditions and support methods and parameters of each support material.

[0052] In some implementations, the relevant knowledge retrieval module is specifically used to:

[0053] Based on the first information of the target coal mine, the coal industry roof disaster prevention and control information database is retrieved through the knowledge label index system of the coal industry roof disaster prevention and control information database; the knowledge label index system includes top-level classification, subdivided labels and specific attributes, and the knowledge label index system includes three dimensions, which are data source, influencing factors and prevention parameters.

[0054] In some implementations, the device further includes a model building module, and when building the roof stability main control factor identification model, the model building module is used to:

[0055] Obtain text information related to roof disasters and support prevention in the coal industry;

[0056] The text information is processed by natural language processing technology to obtain integrated roof disaster and support prevention information;

[0057] By using data mining and analysis technology, multiple key factors affecting roof stability are extracted from the integrated roof disaster and support prevention information;

[0058] Obtaining the contribution of each of the multiple key factors;

[0059] Based on the multiple key factors and the contribution of each key factor, a roof stability main control factor identification model is constructed.

[0060] In some implementations, when obtaining the contribution of each of the multiple key factors, the model building module is used to:

[0061] Given a roof stability data, the roof stability evaluation index data calculated under the above key factors is recorded as D = [(x1, y1), (x2, y2), ..., (x n ,y n )], where x is an m-dimensional vector, i.e., it contains m key factors, and y is the roof stability index data;

[0062] Suppose a constructed classification and regression tree CART regression tree has m leaves to divide the key factor x of the input space into m basic units R1, R2, ..., R m , the CART objective minimizes the mean square error mse formula as follows:

[0063]

[0064] Among them, c m represents the predicted value of the mth leaf;

[0065] The predicted value of the roof stability index is set to the mean of the training set elements contained in the leaves, that is:

[0066] C m =ave(y i |x i ∈leaf m )

[0067] A heuristic method is adopted to traverse all the key factor variables and split positions, and the case with the smallest sum of mse of the leaf node is selected as the partition; the jth feature and its value s are selected as the split variable and split point, and the split variable and split point divide the input space of the parent node into two; in each partition, the split variable and split point are selected so that the mse of the main control factor identification model of roof stability on the training set is minimized, that is, the sum of mse of each leaf is minimized;

[0068] After determining the segmentation key factor j, its information gain is calculated, and the negative logarithm of mse is used for weighting to obtain the importance ranking of the multiple key factors, and the contribution of each key factor is obtained according to the importance ranking of the multiple key factors.

[0069] In some implementations, the support scheme acquisition module is specifically used to:

[0070] Determining key support parameters of the target coal mine based on the main control factors of the target coal mine;

[0071] Based on the coal mine support technology system, determine the target support scheme template;

[0072] Based on the key support parameters of the target coal mine, the support parameters of the target support scheme template are adjusted to obtain the support scheme under the main control factors of the target coal mine.

[0073] In some implementations, the device further includes a system construction module, and the system construction module is used to:

[0074] Obtain multi-dimensional data of different mines, including geological conditions, physical and mechanical parameters of surrounding rocks, laws of mine pressure manifestation, displacement data, and successful and failed support cases;

[0075] Based on the multi-dimensional data, extract the causes of roof accidents and key support parameters;

[0076] Based on the causes of roof accidents and key support parameters, a zoning and classified coal mine support technology system under diverse roof conditions is constructed.

[0077] In some implementations, the support scheme acquisition module is further used to:

[0078] The support plan of the target coal mine is added to the coal industry roof disaster prevention and control information database.

[0079] To achieve the above-mentioned purpose, the third aspect embodiment of the present application proposes an electronic device, comprising: a processor, and a memory communicatively connected to the processor; the memory stores computer-executable instructions; the processor executes the computer-executable instructions stored in the memory to implement the method described in the first aspect.

[0080] To achieve the above-mentioned purpose, the fourth aspect of the present application proposes a computer-readable storage medium, in which computer-readable storage medium is stored computer-executable instructions, and when the computer-executable instructions are executed by a processor, they are used to implement the method described in the first aspect.

[0081] To achieve the above-mentioned purpose, the fifth aspect of the present application proposes a computer program product, including a computer program, which implements the method described in the first aspect when executed by a processor.

[0082] The coal mine roof disaster prevention and control method, device, electronic device and storage medium based on retrieval enhancement generation provided in the present application integrate and analyze various data sources related to roof disaster prevention and control in the coal industry, construct a coal industry roof disaster prevention and control information database and a related knowledge label index system, as well as a roof stability main control factor identification model and a coal support technology system, to realize large model retrieval based on retrieval enhancement generation; according to the large model retrieval results, combined with the roof stability main control factor identification model and the coal support technology system, the main control factors of roof stability are identified and the support plan is obtained, thereby assisting in guiding underground support projects.

[0083] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0084] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0085] Figure 1 A schematic diagram of a flow chart of a coal mine roof disaster prevention and control method based on retrieval enhancement generation provided in an embodiment of the present application;

[0086] Figure 2 A schematic diagram of a knowledge tag indexing system provided in an embodiment of the present application;

[0087] Figure 3 A block diagram of a coal mine roof disaster prevention and control device based on retrieval enhancement generation provided in an embodiment of the present application;

[0088] Figure 4 A block diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0089] Embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limiting the present application.

[0090] Terminology explanation:

[0091] The Retrieval-augmented Generation (RAG) model combines language models and information retrieval techniques. Specifically, when the model needs to generate text or answer questions, it first retrieves relevant information from a large collection of documents, and then uses the retrieved information to guide text generation, thereby improving the quality and accuracy of predictions.

[0092] The following describes the coal mine roof disaster prevention and control method, device and equipment based on retrieval enhancement generation in an embodiment of the present application with reference to the accompanying drawings.

[0093] Figure 1 A flow chart of a method for preventing and controlling coal mine roof disasters based on retrieval enhancement generation provided in an embodiment of the present application.

[0094] It should be noted that the executor of the coal mine roof disaster prevention and control method based on retrieval enhancement generation in the embodiment of the present application is the coal mine roof disaster prevention and control device based on retrieval enhancement generation in the embodiment of the present application, and the coal mine roof disaster prevention and control device based on retrieval enhancement generation can be configured in an electronic device so that the electronic device can execute the coal mine roof disaster prevention and control function based on retrieval enhancement generation.

[0095] like Figure 1 As shown, the coal mine roof disaster prevention and control method based on retrieval enhancement generation includes the following steps:

[0096] Step 101: Acquire first information of a target coal mine input by a user.

[0097] The first information includes information of multiple dimensions of the coal mine obtained by the user based on the on-site scene of the target coal mine and related data, such as roof conditions, mine pressure information, geological information, support information, etc.

[0098] Step 102, based on the first information of the target coal mine, search the coal industry roof disaster prevention and control information database to obtain the search information of the target coal mine; the coal industry roof disaster prevention and control information database includes a variety of data sources related to roof disasters and support prevention and control in the coal industry and support plans under different roof conditions.

[0099] As an implementation method, based on the first information of the target coal mine, the coal industry roof disaster prevention and control information database is retrieved, including: based on the first information of the target coal mine, the coal industry roof disaster prevention and control information database is retrieved through the knowledge label index system of the coal industry roof disaster prevention and control information database; the knowledge label index system includes top-level classification, subdivided labels and specific attributes, and the knowledge label index system includes three dimensions, namely data source, influencing factors and prevention parameters.

[0100] In some embodiments, the method for constructing a knowledge tag index system includes: systematically sorting out and labeling multi-dimensional information such as historical data, support failure cases, geological conditions, support plans, and roof disaster occurrences; determining the structure of the knowledge tag index system, including top-level classification, subdivided tags, and specific attributes, and designing a multi-level, multi-dimensional knowledge tag index system to ensure the efficiency and accuracy of data retrieval.

[0101] For example, the knowledge tag index system is as follows: Figure 2 As shown in the figure, the knowledge label index system includes three dimensions: data source, influencing factors and prevention and control parameters.

[0102] Step 103, merging the first information of the target coal mine and the search information of the target coal mine to obtain merged information; and inputting the merged information into the large language model to obtain supplementary information of the target coal mine.

[0103] That is to say, the description information of the target coal mine is improved through the retrieval results, and the improved description information is input into the large language model to obtain the output result fed back by the large language model, thereby realizing the large language model output based on retrieval enhancement generation.

[0104] Step 104, based on the supplementary information of the target coal mine, the main controlling factors of the target coal mine are identified through the roof stability main controlling factor identification model.

[0105] As an implementation method, the main controlling factor identification model of roof stability obtains the required characteristic information from the supplementary information of the target coal mine, identifies the main controlling factors of the target coal mine based on the characteristic information, and obtains the identification result.

[0106] In some embodiments, a method for constructing a roof stability main control factor identification model includes: obtaining text information related to roof disasters and support prevention and control in the coal industry;

[0107] The text information is processed through natural language processing technology to obtain integrated roof disaster and support prevention information; through data mining and analysis technology, multiple key factors affecting roof stability are extracted from the integrated roof disaster and support prevention information; the contribution of each key factor among multiple key factors is obtained; based on multiple key factors and the contribution of each key factor, a roof stability main control factor identification model is constructed.

[0108] Optionally, calculating the contribution of each key factor to the roof stability may be achieved by calculating the information gain of each key factor.

[0109] In some embodiments, obtaining the contribution of each of the multiple key factors includes:

[0110] Given a roof stability data, the roof stability evaluation index data calculated under each key factor is recorded as D = [(x1, y1), (x2, y2), ..., (x n ,y n )], where x is an m-dimensional vector, i.e., it contains m key factors, and y is the roof stability index data;

[0111] The goal of the regression problem is to construct a function f(x) that can fit the elements in the data set D so that the mse is minimized, that is:

[0112]

[0113] This application uses CART for regression, and the goal is to minimize mse. Suppose a constructed classification and regression tree CART regression tree has m leaves, which means that CART divides the key factor x of the input space into m basic units R1, R2, ..., R m , that is, CART will have at most M different predicted values; the formula for minimizing the mean square error mse of the CART objective is as follows:

[0114]

[0115] Among them, c m represents the predicted value of the mth leaf;

[0116] The goal of minimizing the mse of the CART population is equivalent to minimizing the loss of each leaf. To minimize the mse of a leaf, the predicted value of the top plate stability index needs to be set to the mean of the training set elements contained in the leaf, that is:

[0117] c m =ave(y i |x i ∈leaf m )

[0118] Therefore, in each division, when selecting the splitting variable and splitting point (that is, selecting the feature and the split that divides the feature space into two), the MSE of the model on the training set is minimized, that is, the sum of the MSE of each leaf is minimized.

[0119] A heuristic method is adopted to traverse all the key factor variables and split positions, and the case with the smallest sum of mse of the leaf node is selected as the partition; the jth feature and its value s are selected as the split variable and split point, and the split variable and split point divide the input space of the parent node into two; in each partition, the split variable and split point are selected so that the mse of the main control factor identification model of roof stability on the training set is minimized, that is, the sum of mse of each leaf is minimized;

[0120] After determining the segmentation key factor j, its information gain is calculated and weighted using the negative logarithm of mse to obtain the importance ranking of multiple key factors, and based on the importance ranking of multiple key factors, the contribution of each key factor is obtained.

[0121] Exemplarily, multiple data sources related to roof disasters and support prevention and control in the coal industry are collected, including but not limited to: laws and regulations, national standards, industry standards, enterprise standards, patents, journals, dissertations, published books, newspapers, accident investigation reports, internal documents of enterprises, etc. The collected data are cleaned, classified and integrated using data processing technologies such as Z-score standardization and natural language processing (NLP). Using data mining and analysis technology, the data characteristics of key factors affecting roof disaster prevention and control are extracted from the integrated data, including key factors such as unclear geological structure, mine pressure manifestation and imbalance, insufficient support strength, and inadequate on-site construction and personnel management. That is, the key factors affecting roof stability include geological structure, mine pressure manifestation, support strength, construction management, etc. Using these key information, the contribution of each key factor affecting roof disasters can be identified, and the contribution of each key factor to roof stability can be determined by calculating information gain. Based on the extracted key factors, a roof stability main control factor identification model is established; the model is trained and optimized using machine learning algorithms to continuously improve the recognition accuracy of the model, thereby achieving accurate identification and prevention of roof disasters.

[0122] Step 105, based on the main controlling factors of the target coal mine and the coal mine support technology system, obtain the support plan under the main controlling factors of the target coal mine; the coal mine support technology system includes support materials under different roof conditions and support methods and parameters of each support material.

[0123] As an implementation method, based on the main controlling factors of the target coal mine and the coal mine support technology system, a support plan under the main controlling factors of the target coal mine is obtained; including: based on the main controlling factors of the target coal mine, determining the key support parameters of the target coal mine; based on the coal mine support technology system, determining the target support plan template; based on the key support parameters of the target coal mine, adjusting the support parameters of the target support plan template to obtain the support plan under the main controlling factors of the target coal mine.

[0124] As an implementation method, the support parameters of the target support scheme template are adjusted to obtain the support scheme under the main control factors of the target coal mine, including: based on the support parameters and support targets of the adjusted target support scheme template, the target optimization algorithm is used to obtain the support scheme of the target coal mine. That is, machine learning and optimization algorithms can be used to intelligently design support schemes that adapt to different roof conditions. Through simulation and measured data, the support scheme is continuously optimized to improve the support effect and safety, so as to reduce the occurrence of roof accidents.

[0125] In some embodiments, a method for constructing a coal mine support technology system includes: obtaining multi-dimensional data from different mines, the multi-dimensional data including geological conditions, physical and mechanical parameters of surrounding rocks, mine pressure manifestation laws, displacement data, and successful and failed support cases; based on the multi-dimensional data, analyzing the causes of roof accidents, and extracting key support parameters for roof support under different conditions; based on the key support parameters for roof support under these different conditions, constructing a zoned and classified coal mine support technology system under diversified roof conditions.

[0126] For example, the framework of the coal mine support technology system covers a variety of support materials, methods and parameters such as anchor rods, anchor cables, anchor nets, and masonry.

[0127] Exemplarily, the support scheme of the target coal mine obtained includes obtaining parameter values ​​of key support parameters such as tunnel anchor cables.

[0128] In some embodiments, after obtaining the support plan under the main controlling factors of the target coal mine, it includes: adding the support plan of the target coal mine to the coal industry roof disaster prevention and control information database, thereby continuously expanding the support plans in the coal industry roof disaster prevention and control information database to provide more accurate information for subsequent retrieval.

[0129] The coal mine roof disaster prevention and control method based on retrieval enhancement generation in the embodiment of the present application integrates, analyzes and mines various data sources related to roof disaster prevention and control in the coal industry, constructs a coal industry roof disaster prevention and control information database and a related knowledge tag index system, as well as a roof stability main control factor identification model and a coal support technology system, to achieve large model retrieval based on retrieval enhancement generation; based on the large model retrieval results, combined with the roof stability main control factor identification model and the coal support technology system, the main control factors of roof stability are identified and the support plan is obtained, thereby assisting in guiding underground support projects.

[0130] In order to implement the above-mentioned embodiment, the present application also proposes a coal mine roof disaster prevention and control device based on retrieval enhancement generation. Figure 3 A schematic diagram of a structure of a coal mine roof disaster prevention and control device based on retrieval enhancement generation provided in an embodiment of the present application. Figure 3As shown, the coal mine roof disaster prevention and control device based on retrieval enhancement generation may include: a user information acquisition module 301, a related knowledge retrieval module 302, a model question and answer module 303, a main control factor identification module 304 and a support plan acquisition module 305.

[0131] The user information acquisition module 301 is used to acquire the first information of the target coal mine input by the user;

[0132] The relevant knowledge retrieval module 302 is used to retrieve the coal industry roof disaster prevention and control information database based on the first information of the target coal mine to obtain the search information of the target coal mine; the coal industry roof disaster prevention and control information database includes multiple data sources related to roof disasters and support prevention and control in the coal industry and support schemes under different roof conditions;

[0133] The model question answering module 303 is used to merge the first information of the target coal mine and the search information of the target coal mine to obtain the merged information; and input the merged information into the large language model to obtain the supplementary information of the target coal mine;

[0134] A main control factor identification module 304 is used to identify the main control factors of the target coal mine based on the supplementary information of the target coal mine and through a roof stability main control factor identification model;

[0135] The support scheme acquisition module 305 is used to acquire the support scheme under the main control factors of the target coal mine based on the main control factors of the target coal mine and the coal mine support technology system; the coal mine support technology system includes support materials under different roof conditions and support methods and parameters of each support material.

[0136] Furthermore, in a possible implementation of the embodiment of the present application, the relevant knowledge retrieval module 302 is specifically used to:

[0137] Based on the first information of the target coal mine, the coal industry roof disaster prevention and control information database is retrieved through the knowledge label index system of the coal industry roof disaster prevention and control information database; the knowledge label index system includes top-level classification, subdivided labels and specific attributes, and the knowledge label index system includes three dimensions, namely data source, influencing factors and prevention parameters.

[0138] Furthermore, in a possible implementation of the embodiment of the present application, the device further includes a model building module 306. When building the roof stability main control factor identification model, the model building module 306 is used to:

[0139] Obtain text information related to roof disasters and support prevention in the coal industry;

[0140] The text information is processed through natural language processing technology to obtain integrated roof disaster and support prevention information;

[0141] Through data mining and analysis technology, multiple key factors affecting roof stability are extracted from the integrated roof disaster and support prevention information;

[0142] Obtain the contribution of each key factor among multiple key factors;

[0143] Based on multiple key factors and the contribution of each key factor, a main controlling factor identification model for roof stability is constructed.

[0144] Furthermore, in a possible implementation of the embodiment of the present application, when obtaining the contribution of each key factor among the multiple key factors, the model building module 306 is used to:

[0145] Given a roof stability data, the roof stability evaluation index data calculated under each key factor is recorded as D = [(x1, y1), (x2, y2), ..., (x n ,y n )], where x is an m-dimensional vector, i.e., it contains m key factors, and y is the roof stability index data;

[0146] Suppose a constructed classification and regression tree CART regression tree has m leaves to divide the key factor x of the input space into m basic units R1, R2, ..., R m , the CART objective minimizes the mean square error mse formula as follows:

[0147]

[0148] Among them, c m represents the predicted value of the mth leaf;

[0149] The predicted value of the roof stability index is set to the mean of the training set elements contained in the leaves, that is:

[0150] C m =ave(y i |x i ∈leaf m )

[0151] A heuristic method is adopted to traverse all the key factor variables and split positions, and the case with the smallest sum of mse of the leaf node is selected as the partition; the jth feature and its value s are selected as the split variable and split point, and the split variable and split point divide the input space of the parent node into two; in each partition, the split variable and split point are selected so that the mse of the main control factor identification model of roof stability on the training set is minimized, that is, the sum of mse of each leaf is minimized;

[0152] After determining the segmentation key factor j, its information gain is calculated and weighted using the negative logarithm of mse to obtain the importance ranking of multiple key factors, and based on the importance ranking of multiple key factors, the contribution of each key factor is obtained.

[0153] Furthermore, in a possible implementation of the embodiment of the present application, the support scheme acquisition module 305 is specifically used to:

[0154] Determine the key support parameters of the target coal mine based on the main controlling factors of the target coal mine;

[0155] Determine the target support scheme template based on the coal mine support technology system;

[0156] Based on the key support parameters of the target coal mine, the support parameters of the target support scheme template are adjusted to obtain the support scheme under the main control factors of the target coal mine.

[0157] Furthermore, in a possible implementation of the embodiment of the present application, the device further includes a system construction module 307. When constructing the coal mine support technology system, the system construction module 307 is used to:

[0158] Obtain multi-dimensional data of different mines, including geological conditions, physical and mechanical parameters of surrounding rocks, laws of mine pressure manifestation, displacement data, and successful and failed support cases;

[0159] Based on multi-dimensional data, the causes of roof accidents and key support parameters are extracted;

[0160] Based on the causes of roof accidents and key support parameters, a zoning and classified coal mine support technology system under diverse roof conditions is constructed.

[0161] Furthermore, in a possible implementation of the embodiment of the present application, the support scheme acquisition module 305 is further configured to:

[0162] Add the support plan of the target coal mine to the coal industry roof disaster prevention and control information database.

[0163] It should be noted that the aforementioned explanation of the embodiment of the coal mine roof disaster prevention and control method based on retrieval enhancement generation is also applicable to the coal mine roof disaster prevention and control device based on retrieval enhancement generation in this embodiment, and will not be repeated here.

[0164] In order to implement the above embodiment, the present application also proposes an electronic device. Figure 4 , Figure 4 Schematic diagram of the structure of the electronic device provided in the embodiment of the present application. Figure 4As shown, the electronic device 400 includes: a processor 401, and a memory 402 communicatively connected to the processor 401; the memory 402 stores computer-executable instructions; the processor 401 executes the computer-executable instructions stored in the memory to implement the method provided by the aforementioned embodiment.

[0165] In order to implement the above embodiments, the present application also proposes a computer-readable storage medium, in which computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, they are used to implement the methods provided by the above embodiments.

[0166] In order to implement the above embodiments, the present application also proposes a computer program product, including a computer program, which implements the methods provided by the above embodiments when executed by a processor.

[0167] The collection, storage, use, processing, transmission, provision and disclosure of user personal information involved in this application are in compliance with relevant laws and regulations and do not violate public order and good morals.

[0168] It should be noted that personal information from users should be collected for legitimate and reasonable purposes and should not be shared or sold outside of these legitimate uses. In addition, such collection / sharing should be carried out after receiving the user's informed consent, including but not limited to notifying the user to read the user agreement / user notice and sign the agreement / authorization including authorization of relevant user information before the user uses the function. In addition, any necessary steps should be taken to protect and safeguard access to such personal information data and ensure that others who have access to personal information data comply with its privacy policy and procedures.

[0169] The present application is expected to provide an implementation scheme for users to selectively block the use or access of personal information data. That is, the present disclosure is expected to provide hardware and / or software to prevent or block access to such personal information data. Once the personal information data is no longer needed, the risk can be minimized by limiting data collection and deleting the data. In addition, when applicable, such personal information is de-identified to protect the privacy of the user.

[0170] In the description of the aforementioned embodiments, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0171] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In the description of this application, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0172] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, fragment or portion of code comprising one or more executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present application belong.

[0173] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute the instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purpose of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses. More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or processing in other suitable ways if necessary, and then stored in a computer memory.

[0174] It should be understood that the various parts of the present application can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0175] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.

[0176] In addition, each functional unit in each embodiment of the present application may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0177] The storage medium mentioned above may be a read-only memory, a magnetic disk or an optical disk, etc. Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limiting the present application. A person of ordinary skill in the art may change, modify, replace and modify the above embodiments within the scope of the present application.

Claims

1. A method for preventing and controlling coal mine roof disasters based on retrieval enhancement generation, characterized in that: The following steps are involved: Acquire the first information of the target coal mine input by the user; Based on the first information of the target coal mine, a coal industry roof disaster prevention and control information database is searched to obtain the search information of the target coal mine; the coal industry roof disaster prevention and control information database includes multiple data sources related to roof disasters and support prevention and control in the coal industry and support schemes under different roof conditions; Merging the first information of the target coal mine and the search information of the target coal mine to obtain merged information; and inputting the combined information into a large language model to obtain supplementary information of the target coal mine; Based on the supplementary information of the target coal mine, the main controlling factors of the target coal mine are identified through a roof stability main controlling factor identification model; Based on the main controlling factors of the target coal mine and the coal mine support technology system, a support plan under the main controlling factors of the target coal mine is obtained; the coal mine support technology system includes support materials under different roof conditions and support methods and parameters of each support material.

2. The method according to claim 1, characterized in that The searching of a coal industry roof disaster prevention and control information database based on the first information of the target coal mine includes: Based on the first information of the target coal mine, the coal industry roof disaster prevention and control information database is retrieved through the knowledge label index system of the coal industry roof disaster prevention and control information database; the knowledge label index system includes top-level classification, subdivided labels and specific attributes, and the knowledge label index system includes three dimensions, which are data source, influencing factors and prevention parameters.

3. The method according to claim 1, characterized in that The method for constructing the roof stability main control factor identification model includes: Obtain text information related to roof disasters and support prevention in the coal industry; The text information is processed by natural language processing technology to obtain integrated roof disaster and support prevention information; By using data mining and analysis technology, multiple key factors affecting roof stability are extracted from the integrated roof disaster and support prevention information; Obtaining the contribution of each of the multiple key factors; Based on the multiple key factors and the contribution of each key factor, a roof stability main control factor identification model is constructed.

4. The method according to claim 3, characterized in that The obtaining of the contribution of each of the multiple key factors includes: Given a roof stability data, the roof stability evaluation index data calculated under the above key factors is recorded as D = [(x1, y1), (x2, y2), ..., (x n ,y n )], where x is an m-dimensional vector, i.e., it contains m key factors, and y is the roof stability index data; Suppose a constructed classification and regression tree CART regression tree has m leaves to divide the key factor x of the input space into m basic units R1, R2, ..., R m , the CART objective minimizes the mean square error mse formula as follows: Among them, c m represents the predicted value of the mth leaf; The predicted value of the roof stability index is set to the mean of the training set elements contained in the leaves, that is: c m =ave(and i |x i ∈leaf m ) A heuristic method is adopted to traverse all the key factor variables and split positions, and the case with the smallest sum of mse of the leaf node is selected as the partition; the jth feature and its value s are selected as the split variable and split point, and the split variable and split point divide the input space of the parent node into two; in each partition, the split variable and split point are selected so that the mse of the main control factor identification model of roof stability on the training set is minimized, that is, the sum of mse of each leaf is minimized; After determining the segmentation key factor j, its information gain is calculated, and the negative logarithm of mse is used for weighting to obtain the importance ranking of the multiple key factors, and the contribution of each key factor is obtained according to the importance ranking of the multiple key factors.

5. The method according to claim 1, characterized in that The step of obtaining a support scheme under the main control factors of the target coal mine based on the main control factors of the target coal mine and the coal mine support technology system comprises: Determining key support parameters of the target coal mine based on the main control factors of the target coal mine; Based on the coal mine support technology system, determine the target support scheme template; Based on the key support parameters of the target coal mine, the support parameters of the target support scheme template are adjusted to obtain the support scheme under the main control factors of the target coal mine.

6. The method according to claim 1, characterized in that The method for constructing the coal mine support technology system comprises: Obtain multi-dimensional data of different mines, including geological conditions, physical and mechanical parameters of surrounding rocks, laws of mine pressure manifestation, displacement data, and successful and failed support cases; Based on the multi-dimensional data, extract the causes of roof accidents and key support parameters; Based on the causes of roof accidents and key support parameters, a zoning and classified coal mine support technology system under diverse roof conditions is constructed.

7. The method according to claim 1, characterized in that After obtaining the support scheme under the main control factors of the target coal mine, the method includes: The support plan of the target coal mine is added to the coal industry roof disaster prevention and control information database.

8. A coal mine roof disaster prevention and control device based on retrieval enhancement generation, characterized in that: include: A user information acquisition module, used to acquire first information of a target coal mine input by a user; A related knowledge retrieval module is used to retrieve a coal industry roof disaster prevention and control information database based on the first information of the target coal mine to obtain the retrieval information of the target coal mine; the coal industry roof disaster prevention and control information database includes multiple data sources related to roof disasters and support prevention and control in the coal industry and support schemes under different roof conditions; A model question-answering module is used to merge the first information of the target coal mine and the search information of the target coal mine to obtain merged information; and input the merged information into a large language model to obtain supplementary information of the target coal mine; A main control factor identification module, used to identify the main control factors of the target coal mine based on the supplementary information of the target coal mine and through a roof stability main control factor identification model; A support scheme acquisition module is used to acquire a support scheme under the main control factors of the target coal mine based on the main control factors of the target coal mine and the coal mine support technology system; the coal mine support technology system includes support materials under different roof conditions and support methods and parameters of each support material.

9. An electronic device, characterized in that: include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, which are used to implement the method according to any one of claims 1 to 7 when executed by a processor.

Images