Coal mine intelligent mining occupational appraisal training system

Abstract

The invention relates to the technical field of occupational appraisal training, in particular to a coal mine intelligent mining occupational appraisal training system. According to the invention, through training, workers in the coal mining industry can fully understand the knowledge content of the intelligent system of the fully mechanized coal mining face and gradually master the operation, use and maintenance of the intelligent system of the fully mechanized coal mining face through operation drilling. The fully mechanized coal mining face intelligent platform is developed according to coal mine intelligent system platform software, an actual fully mechanized coal mining face intelligent mining environment is built through three-dimensional visualization and virtual reality technologies, and a coal mine intelligent mining real scene is displayed. And the theoretical knowledge learning module is combined to learn related knowledge of the fully-mechanized coal mining equipment, and the training assessment module assesses the training effect of the workers, so that the whole-process intelligent management of the intelligent mining skill cultivation of the coal mine is realized, and the training effect of the workers is greatly improved.

Description

Technical Field

[0001] The invention relates to the technical field of occupational identification and training, and in particular to an occupational identification and training system for intelligent coal mining. Background Art

[0002] With the continuous application of automation and information technology, coal mine production has gradually become intelligent, improving production efficiency and safety. The fully mechanized mining face is the core link of coal mining, which requires workers to have professional knowledge and skills; and with the application of the fully mechanized mining face intelligent system, the requirements for workers have also increased. However, in traditional coal mine operation training, it mainly relies on theoretical teaching and on-site practical operation, which makes it difficult to systematically cultivate the comprehensive ability of operators.

[0003] Therefore, there is an urgent need to develop an intelligent coal mining occupation identification and training system to overcome the shortcomings of the existing technology. Summary of the invention

[0004] The purpose of the present invention is to provide a coal mine intelligent mining occupation identification and training system to solve the problems raised in the above background technology.

[0005] The coal mine intelligent mining occupation appraisal training system includes: fully mechanized mining face intelligent platform, theoretical knowledge learning module, training assessment module and training effect evaluation module; The intelligent platform of fully mechanized mining face is used to build the actual intelligent mining environment of fully mechanized mining face and display the real scene of intelligent mining in coal mines; Theoretical knowledge learning module is used to provide the staff with relevant knowledge content of the intelligent platform of the fully mechanized mining face, so that the staff can learn relevant knowledge content at any time without being restricted by time and place; The training and assessment module is used to provide staff with a simulated operating environment so that they can conduct simulated exercises; it also records and evaluates the theoretical knowledge and practical operating ability of staff throughout the process, and gives a quantitative operating ability score; The training effectiveness evaluation module is used to comprehensively analyze and evaluate the training effectiveness of staff based on their theoretical learning and practical training, and provide targeted improvement suggestions.

[0006] Furthermore, the intelligent platform of fully mechanized mining face includes a three-dimensional visualization module, a virtual reality interaction module, a data communication module and a fault diagnosis module; The 3D visualization module is used to construct the 3D visualization scene of the intelligent platform of the fully mechanized mining face and realize the 3D visualization display of the fully mechanized mining face equipment and environment; Virtual reality interaction module, which is used to use virtual reality technology to realize the interactive operation between the staff and the intelligent platform of the fully mechanized mining face; A data communication module is used to communicate data with the actual fully mechanized mining working face, receive and store sensor data and equipment operating condition information of the actual fully mechanized mining working face; The fault diagnosis module is used to simulate common faults in actual fully mechanized mining working faces and display them through the 3D visualization module and virtual reality interaction module to guide the staff to diagnose and eliminate faults; Through data communication, the data communication module synchronizes the sensor data and equipment operating condition information of the actual fully mechanized mining working face to the coal mine intelligent mining professional identification and training system. The three-dimensional visualization module, virtual reality interaction module and fault diagnosis module call the actual fully mechanized mining working face information saved in the coal mine intelligent mining professional identification and training system to realize the linkage between simulation training and the actual fully mechanized mining working face.

[0007] Furthermore, the fault diagnosis module establishes a fault simulation model based on the operating data and fault data of the actual fully mechanized mining working face, automatically generates the characteristic parameters and occurrence patterns of various common faults, as well as the abnormal operating condition data of the equipment when the fault occurs, and simulates and displays the possible faults that may occur in the actual fully mechanized mining working face through the three-dimensional visualization module and the virtual reality interaction module, guiding the staff to diagnose and eliminate the faults.

[0008] Furthermore, the theoretical knowledge learning module generates a staff portrait for the staff through the theoretical knowledge recommendation model. The characteristic dimensions of the staff portrait include knowledge base, learning ability and operational skills. A personalized training plan is dynamically generated based on the staff portrait, and the learning content, progress, difficulty and presentation method are adjusted in real time during the training process. The theoretical knowledge recommendation model is built using machine learning algorithms and historical training data. It intelligently generates customized theoretical learning plans based on the personal characteristics of staff members and continuously optimizes itself as learning behavior data accumulates. The theoretical knowledge learning module uses the three-dimensional visualization module in the intelligent platform of the comprehensive mining working face to display the working principles and operating essentials of the comprehensive mining working face, and displays the equipment structure and process flow in a virtual environment, thereby strengthening the staff's intuitive understanding of theoretical knowledge.

[0009] Furthermore, the theoretical knowledge recommendation model analyzes the learning behavior data of the staff, identifies the weak links of the staff in theoretical knowledge learning, automatically optimizes the personalized training plan, and conducts targeted intensive training for the staff; The specific steps of the theoretical knowledge recommendation model to achieve personalized training program optimization are as follows: S101: The theoretical knowledge recommendation model analyzes the learning behavior data of the staff and obtains the learning behavior sequence of the staff , where x trepresents the learning behavior feature vector of the staff at time node t, and time node t is a time point in the staff's learning process; S102: Use the gated recurrent unit GRU network to model the knowledge state transformation of the staff, using h t Indicates the staff's mastery of each knowledge point; S103: Use the attention mechanism to aggregate the representations of different knowledge points, consider the dependencies between different knowledge points, and obtain the comprehensive knowledge state representation S of the staff member t ; ; in, represents the attention weight of knowledge point i at time node t; h t Indicates the staff's mastery of each knowledge point; Indicates the staff's mastery of knowledge point i at time node t; Represents the dependency matrix between knowledge point i and knowledge point j; Represents the attention weight of knowledge point j at time node t; Indicates the staff's mastery of knowledge point j at time node t; S104: A multi-layer perceptron is used to predict the staff's mastery of each knowledge point, and the difficulty of different knowledge points is reflected by the knowledge point difficulty vector. The specific calculation formula for the mastery of the knowledge point is: ; in, It represents the probability of staff mastering knowledge point i at time node t, which is used to optimize the subsequent personalized training strategy; Represents the learning behavior sequence and knowledge points that the staff has mastered before Under the condition of , predict whether the staff has mastered the knowledge point i at time node t; represents the set of knowledge points that the staff has mastered before time node t; σ represents the probability of converting the result of linear combination into the knowledge point mastered by the staff; is a weight vector, which indicates that at time point t, for knowledge point i, the model is used to measure the importance of the input feature; It represents the knowledge status of the staff member on knowledge point i at time point t; represents the state vector at time point t; Represents the knowledge point difficulty vector, which is used to reflect the difficulty of different knowledge points; is a constant used to adjust the output value so that when all input features are zero, the model can still predict a non-zero probability value; S105: By analyzing the staff's mastery of various knowledge points, identifying the staff's weak links in theoretical knowledge learning, automatically optimizing personalized training plans, and conducting targeted intensive training for the staff.

[0010] Furthermore, the training and assessment module includes an operation practice unit, which combines with the intelligent platform of the comprehensive mining working face to provide the staff with a virtual environment practice operation. The virtual environment practice operation includes routine working condition simulation and abnormal working condition simulation. The routine working condition simulation guides the staff to conduct standardized operation training based on the actual work flow of the comprehensive mining working face; the abnormal working condition simulation randomly generates emergencies in the comprehensive mining working face to test the staff's emergency response and handling capabilities.

[0011] Furthermore, the operation practice unit also conducts real-time analysis and diagnosis of the staff's operating behavior based on artificial intelligence algorithms. By capturing the staff's operations in the virtual environment, it judges the standardization and effectiveness of each operation action, and gives corresponding prompts and feedback to help the staff promptly discover and correct incorrect operating habits. At the same time, the operation practice unit can connect to multiple staff virtual reality terminals at the same time, allowing the staff to carry out team system operation training and practice information sharing, task coordination and emergency equipment capabilities in the comprehensive mining face.

[0012] Furthermore, the training assessment module also includes an assessment unit, which records the actual operation ability of the staff throughout the process and uses the training results evaluation model to automatically evaluate and score the staff's operation behavior; The training outcome evaluation model is based on the key steps and requirements in the operating procedures of the fully mechanized mining face. It determines whether each operation of the staff is standardized and complies with the safe production process, and makes quantitative scores. Among them, the training outcome evaluation model uses weighted summation to achieve automatic evaluation and scoring. The specific calculation formula is: ; in, is the scoring result of the training outcome evaluation; S is the scoring item of the operation steps, which is used to reflect whether the staff operates in accordance with the requirements of the regulations; α is the step score weight, which is used to reflect the importance of the operation steps; D is the safety hazard deduction item, which is used to cultivate the safety awareness of the staff by deducting points for safety hazards; β is the safety hazard weight, which is used to reflect the importance of safety hazards; P is the personalized adjustment factor, which is used to dynamically adjust the scoring results according to the specific situation of the staff, so as to improve the pertinence of the training outcome evaluation.

[0013] Furthermore, the training effect evaluation module adopts an incremental learning algorithm. As the theoretical learning and practical training data of staff members continue to accumulate, the theoretical knowledge recommendation model and personalized training strategy in the theoretical knowledge learning module are continuously optimized; The specific execution steps of the training effect evaluation module are: S201: continuously collect the training data generated by the staff in the theoretical learning module and the operation practice unit, and perform preprocessing operations on the training data to improve the data quality; S202: Filter out key features that affect the training effect from the preprocessed training data using statistical analysis and machine learning methods. The key features include the mastery of important knowledge points, the gap between the learning progress and the average level, and the error rate during the practical operation. S203: using an incremental learning algorithm, based on the original theoretical knowledge recommendation model and personalized training strategy, and using the newly collected training data to fine-tune and update the original theoretical knowledge recommendation model and personalized training strategy; S204: Based on the fine-tuned and updated theoretical knowledge recommendation model and personalized training strategy in step S203, the theoretical knowledge recommendation model is optimized, and the optimization includes adjusting the recommendation weights of knowledge points and optimizing the learning path; S205: improving the personalized training strategy based on the optimized theoretical knowledge recommendation model, and formulating a targeted personalized training plan based on the actual learning situation; S206: Continue to collect the latest training data of the staff and execute step S203 regularly.

[0014] Furthermore, it also includes a training effect visualization module, which uses data visualization technology to intuitively present each staff member's theoretical knowledge mastery, practical skill level, and comprehensive ability improvement, providing a global analysis perspective for training managers and supporting comparative analysis with talent training data of other coal mining companies in the industry.

[0015] Beneficial effects of the present invention: The present invention uses three-dimensional and virtual simulation technologies to construct a real comprehensive mining working face environment, allowing staff to conduct theoretical learning and practical training in a simulated scene, making comprehensive mining training close to actual production; at the same time, it adopts personalized training strategies to dynamically adjust theoretical knowledge recommendations and training plans according to the learning situation of the staff, so as to meet the training needs of different staff in a targeted manner and greatly improve the training effect.

[0016] The present invention simulates various sudden abnormal situations of fully mechanized mining working faces in a virtual environment to test the emergency response and handling capabilities of the staff. This not only helps the staff to become familiar with emergency plans and operating procedures, but also cultivates their safety awareness and improves their ability to deal with emergencies in actual work. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 is a system structure diagram of the present invention; Figure 2 A flowchart of specific operation steps for optimizing the personalized training program in the present invention; Figure 3 This is a flow chart for optimizing the theoretical knowledge recommendation model and personalized training strategy in the present invention. DETAILED DESCRIPTION

[0018] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0019] As attached Figure 1-3 As shown, the present invention provides a coal mine intelligent mining occupation identification and training system, including a comprehensive mining face intelligent platform, a theoretical knowledge learning module, a training assessment module, a training effect evaluation module and a training effect visualization module.

[0020] The intelligent platform for fully mechanized mining working faces is used to build an actual intelligent mining environment for fully mechanized mining working faces and display the real scenes of intelligent mining in coal mines.

[0021] The intelligent platform for the fully mechanized mining face is developed entirely in accordance with the coal mine intelligent system platform software. It has a complete interface, dynamic data display, and comprehensive operation and application functions. It can simulate and control two intelligent coal mining modes, horizontal and inclined tunnels, and can automatically run and stop the operation with one button.

[0022] In addition to being developed in accordance with the coal mine intelligent system platform software, the intelligent platform for the comprehensive mining working face also includes a three-dimensional visualization module, a virtual reality interaction module, a data communication module and a fault diagnosis module.

[0023] The 3D visualization module is used to build a 3D visualization scene of the intelligent platform of the fully mechanized mining face, and realize the 3D visualization display of the fully mechanized mining face equipment and environment. The fully mechanized mining equipment is displayed through a 3D stereoscopic picture, and the structure of the fully mechanized mining equipment is clearly displayed. After clicking on the fully mechanized mining equipment, the fully mechanized mining equipment is rotated 360° in a separate picture, which clearly shows the structure of the fully mechanized mining equipment. In addition, the working principle of the fully mechanized mining equipment is synchronously explained in combination with the theoretical knowledge learning module.

[0024] The virtual reality interaction module is used to use virtual reality technology to realize the interactive operation between the staff and the intelligent platform of the fully mechanized mining face. The fault diagnosis module lists the common faults in the fully mechanized mining face intelligent system in the form of a catalog. Click on each fault in the catalog. The virtual reality interaction module will display the location of the clicked fault in the overall virtual model diagram, and explain the causes and treatment methods of the fault in combination with the theoretical knowledge learning module.

[0025] The data communication module is used to communicate data with the actual fully mechanized mining working face, receive and store sensor data and equipment operating condition information of the actual fully mechanized mining working face, including but not limited to coal mining machine operating condition display, hydraulic support operating condition display, working face voice system status display, working face video display, etc.

[0026] The fault diagnosis module is used to simulate common faults in actual fully mechanized mining working faces, and display them through the three-dimensional visualization module and virtual reality interaction module to guide staff to diagnose and eliminate faults.

[0027] The fault diagnosis module establishes a fault simulation model based on the actual fully mechanized mining face operation data and fault data, automatically generates the characteristic parameters and occurrence patterns of various common faults, as well as the abnormal operating condition data of the equipment when the fault occurs, and simulates and displays the possible faults that may occur in the actual fully mechanized mining face through the three-dimensional visualization module and the virtual reality interaction module, guiding the staff to diagnose and eliminate the faults.

[0028] Through data communication, the data communication module synchronizes the sensor data and equipment operating condition information of the actual fully mechanized mining working face to the coal mine intelligent mining occupation identification and training system. The three-dimensional visualization module, virtual reality interaction module and fault diagnosis module call the saved actual fully mechanized mining working face information to realize the linkage between simulation training and the actual fully mechanized mining working face.

[0029] The theoretical knowledge learning module is used to provide staff with relevant knowledge content of the intelligent platform of the comprehensive mining working face, so that staff can learn relevant knowledge content at any time without being restricted by time and place.

[0030] The theoretical knowledge learning module provides staff with the opportunity to learn theoretical knowledge at any time in the form of pictures, texts and videos. The theoretical knowledge includes at least: the use and operation manual of the comprehensive mining face intelligent system, the hardware manual of the comprehensive mining face intelligent system, the software manual of the comprehensive mining face intelligent system, the introduction of comprehensive mining equipment, the introduction of comprehensive mining process, customized reports, as well as policies, standards and technical requirements.

[0031] Among them, the theoretical knowledge learning module generates staff portraits for staff through the theoretical knowledge recommendation model. The characteristic dimensions of the staff portraits include knowledge base, learning ability and operational skills. Personalized training plans are dynamically generated based on the staff portraits, and the learning content, progress, difficulty and presentation methods are adjusted in real time during the training process. The theoretical knowledge recommendation model is constructed using machine learning algorithms and historical training data. It intelligently generates customized theoretical learning plans based on the personal characteristics of the staff, and continuously optimizes itself with the accumulation of learning behavior data to improve the pertinence and effectiveness of theoretical training.

[0032] The theoretical knowledge learning module demonstrates the working principle and operating essentials of the intelligent platform of the comprehensive mining working face through the three-dimensional visualization module in the intelligent platform of the comprehensive mining working face, displays the equipment structure and process flow in a virtual environment, and strengthens the staff's intuitive understanding of theoretical knowledge.

[0033] The intelligent platform of the comprehensive mining face displays the comprehensive mining equipment in a three-dimensional picture, clearly showing the structure of the comprehensive mining equipment; when clicking on the picture of the comprehensive mining equipment, the theoretical knowledge learning module will explain the working principle of the comprehensive mining equipment components synchronously through voice and text, and the three-dimensional picture of the comprehensive mining equipment will rotate 360° to clearly show the structure of the comprehensive mining equipment. At the same time, the theoretical knowledge learning module will list the common faults in the intelligent system of the comprehensive mining face in the form of a catalog. Click on one of the faults, and the location of the fault will be displayed in the comprehensive mining face model established by the virtual reality interaction module, and the cause of the fault and the treatment method will be explained.

[0034] A theoretical knowledge recommendation model is also deployed in the theoretical knowledge learning module. The theoretical knowledge recommendation model analyzes the learning behavior data of staff, identifies their weak links, automatically optimizes personalized training plans, and conducts targeted intensive training for staff.

[0035] As attached Figure 2 As shown in the figure, the specific operation steps of the theoretical knowledge recommendation model to achieve personalized training program optimization are: S101: The theoretical knowledge recommendation model analyzes the learning behavior data of the staff and obtains the learning behavior sequence of the staff , where x t The feature vector representing the learning behavior of the staff at time node t, including but not limited to the number of clicks, learning time, and the amount of wrong questions, is used to reflect the learning status and habits of the staff, and the time node t is a time point in the staff training process; S102: Use the gated recurrent unit GRU network to model the knowledge state transformation of the staff, using h t Indicates the staff's mastery of each knowledge point; ; in, It is the update gate of the gated recurrent unit GRU network; is the candidate hidden state; It is a hidden state, that is, the staff's mastery of each knowledge point; represents the hidden state vector at the previous time step t-1, which represents the knowledge state of the worker before entering the current learning stage; S103: Use the attention mechanism to aggregate the representations of different knowledge points, consider the dependencies between different knowledge points, and obtain the comprehensive knowledge state representation of the staff member ; ; in, represents the attention weight of knowledge point i at time node t, which is used to reflect the importance of knowledge point i to the staff at time node t; It represents the staff's mastery of knowledge point i at time node t. By weighted summing it, we can get the overall knowledge status of the staff at the current moment; Represents the dependency matrix between knowledge point i and knowledge point j, which is used to reflect the degree of association between i and j; represents the attention weight of knowledge point j at time node t, which is used to reflect the importance of knowledge point j to the staff at time node t; Indicates the staff's mastery of knowledge point j at time node t, by considering the dependencies between different knowledge points and representing different knowledge points Perform weighted summation to obtain a more comprehensive knowledge state representation. Comprehensive knowledge state representation of staff By integrating the staff's mastery of various knowledge points and the dependencies between different knowledge points, a more comprehensive representation of the staff's knowledge status is provided for subsequent personalized training program recommendations; S104: A multi-layer perceptron is used to predict the staff's mastery of each knowledge point, and the difficulty of different knowledge points is reflected by the knowledge point difficulty vector d, wherein the specific calculation formula for the knowledge point mastery is: ; in, It represents the probability of staff mastering knowledge point i at time node t, which is used to optimize the subsequent personalized training strategy; Represents the learning behavior sequence and knowledge points that the staff has mastered before Under the condition of , predict whether the staff has mastered the knowledge point i at time node t; It is used to indicate the staff's mastery of knowledge point i at time node t; represents the set of knowledge points that the staff has mastered before time node t; σ represents the probability of converting the result of linear combination into the knowledge point mastered by the staff; is a weight vector, which indicates that at time point t, for knowledge point i, the model is used to measure the importance of the input feature; It represents the knowledge status of the staff member on knowledge point i at time point t; represents the state vector at time point t; Represents the knowledge point difficulty vector, which is used to reflect the difficulty of different knowledge points; is a constant used to adjust the output value so that when all input features are zero, the model can still predict a non-zero probability value; S105: By analyzing the staff's mastery of various knowledge points, identifying the staff's weak links, automatically optimizing personalized training plans, and conducting targeted intensive training for the staff.

[0036] The training and assessment module is used to provide staff with a simulated operating environment so that they can conduct simulated exercises in the simulated operating environment; it also records and evaluates the theoretical knowledge and practical operating capabilities of staff throughout the process and gives a quantitative operating capability score.

[0037] The training and assessment module includes an operation practice unit. The operation practice unit combines with the intelligent platform of the comprehensive mining working face to provide the staff with a virtual environment for practice operations. The virtual environment practice operations include routine working condition simulation and abnormal working condition simulation. The routine working condition simulation guides the staff to conduct standardized operation training based on the actual work flow of the comprehensive mining working face. Standardized operation training includes but is not limited to equipment operation and production scheduling; the abnormal working condition simulation randomly generates emergencies in the comprehensive mining working face, including equipment failures, disasters and accidents, etc., to test the emergency response and handling capabilities of the staff.

[0038] The operation practice unit also conducts real-time analysis and diagnosis of the staff's operating behavior based on artificial intelligence algorithms. By capturing the staff's operations in a virtual environment, it judges the standardization and effectiveness of each operation action, and gives corresponding prompts and feedback to help the staff promptly discover and correct incorrect operating habits. At the same time, the operation practice unit can connect to the virtual reality terminals of multiple staff members at the same time, enabling the staff to carry out team system operation training and practice information sharing, task coordination and emergency equipment capabilities in the comprehensive mining face.

[0039] The training assessment module also includes an assessment unit, which records the actual operating ability of the staff throughout the process and uses the training achievement evaluation model to automatically evaluate and score the staff's operating behavior; the training achievement evaluation model is based on the key steps and requirements in the comprehensive mining face operating procedures, determines whether each operation of the staff is standardized and complies with the safe production process, and performs quantitative scoring; the training achievement evaluation model uses a weighted summation algorithm to achieve automatic evaluation and scoring. The specific calculation formula is: ; in, is the scoring result of the training outcome evaluation; S is the scoring item of the operation steps, which is used to reflect whether the staff operates in accordance with the requirements of the regulations. If the staff's operation complies with the requirements of the regulations, the score is 1, and if the operation does not comply with the requirements of the regulations, the score is 0; α is the step score weight, which is used to reflect the importance of the operation step; D is the safety hazard deduction item, that is, the potential safety hazard that can be caused by the staff's operation. By deducting points for safety hazards, the safety awareness of the staff is cultivated. The corresponding deduction value can be set according to the severity of the safety implications; β is the safety hazard weight, which is used to reflect the importance of safety hazards; P is a personalized adjustment factor, which is used to dynamically adjust the scoring results according to the specific situation of the staff, so as to improve the pertinence of the training outcome evaluation.

[0040] The training success model adopts a weighted sum algorithm. By setting step scores, safety hazard deductions and personalized adjustment factors, it can comprehensively evaluate the standardization, safety and personal characteristics of staff operations, and more objectively and accurately reflect the actual operational capabilities of staff.

[0041] The actual operation process of the comprehensive mining face often includes multiple steps, each of which is an important indicator for evaluating the ability of the staff. By assigning different weights to these important indicators and summing them up, the performance of the staff in the entire operation process can be fully reflected. In the training result evaluation model, adding safety hazard deduction items can make the staff aware of the importance of safe operation, so that they pay more attention to safe behavior; at the same time, relying solely on the operation step scores cannot fully reflect the actual operation level of the staff. Including safety hazard deduction items in the training result evaluation indicators can more objectively evaluate the comprehensive ability of the staff. The personalized adjustment factor gives certain points rewards or deductions according to the learning progress and performance of the staff. It can enable the training results evaluation model to dynamically adjust the evaluation results according to the specific situation of the staff, improve the pertinence of the model evaluation, and also help stimulate the learning enthusiasm of the staff and promote the continuous improvement of personal ability.

[0042] The assessment unit also has functions such as online practice, formal examinations, score inquiry and wrong question summary. The question bank content comprehensively covers all aspects of the intelligent system of the comprehensive mining working face, and can realize random selection of questions and test papers, which is convenient for unified organization and management.

[0043] The question bank includes at least knowledge-based, application-based, and standard-based questions. Knowledge-based questions include but are not limited to basic electrical knowledge of electrical control equipment in comprehensive mining working faces, basic theories of electro-hydraulic control, theories of intelligent comprehensive mining, system structure theory, and software theory. Application-based questions include but are not limited to electrical control hardware and software applications, comprehensive mining electric control system applications, and comprehensive mining intelligent applications. Standard-based questions include but are not limited to relevant standards and operating specifications for comprehensive mining intelligent applications.

[0044] The training effectiveness evaluation module is used to comprehensively analyze and evaluate the training effectiveness of staff based on their theoretical learning and practical training, and provide targeted improvement suggestions.

[0045] The training effect evaluation module adopts an incremental learning algorithm. As the theoretical learning and practical training data of staff members continue to accumulate, the theoretical knowledge recommendation model and personalized training strategy in the theoretical knowledge learning module are continuously optimized; as shown in the attached Figure 3 As shown, the specific steps include the following: S201: continuously collect the training data generated by the staff in the theoretical learning module and the operation practice unit, wherein the training data includes but is not limited to the learning time, learning progress, exercise scores, practical scores, and the mastery of knowledge points, and then store the generated data in the database of the coal mine intelligent mining occupation identification training system; then perform preprocessing operations including cleaning, denoising and normalization on the stored training data to improve the data quality, and then organize and summarize the preprocessed training data according to the staff, knowledge points and time dimensions for subsequent analysis and processing; S202: Filter out key features that affect the training effect from the preprocessed training data using statistical analysis and machine learning methods. The key features include but are not limited to the mastery of important knowledge points, the gap between the learning progress and the average level, and the error rate during the practical operation; S203: using an incremental learning algorithm, based on the original theoretical knowledge recommendation model and personalized training strategy, and using the newly collected training data to fine-tune and update the original theoretical knowledge recommendation model and personalized training strategy, wherein the incremental learning algorithm may include but is not limited to online learning and transfer learning algorithms, wherein the purpose of fine-tuning and updating the original theoretical knowledge recommendation model and personalized training strategy is to enable the model and strategy to adapt to the constantly changing learning data; S204: Based on the fine-tuned and updated theoretical knowledge recommendation model in step S203, the theoretical knowledge recommendation model is optimized. The specific optimization includes adjusting the recommendation weights of knowledge points and optimizing the learning path. The optimized theoretical knowledge recommendation model can provide personalized theoretical learning content for staff more accurately. S205: Improve the personalized training strategy based on the optimized theoretical knowledge recommendation model, adjust the time allocation, exercise difficulty, and assessment frequency of theoretical learning and practical training according to the actual learning situation of each staff member, and formulate a targeted personalized training plan; S206: Continuously collect the latest training data of the staff and regularly execute step S203, so that the theoretical knowledge recommendation model and personalized training strategy can be continuously optimized with the training progress of the staff and adapt to the changes in the learning ability and knowledge structure of the staff.

[0046] In this embodiment, the coal mine intelligent mining occupational identification training system also includes a training effect visualization module, which intuitively presents each staff member's theoretical knowledge mastery, practical skill level, and comprehensive ability improvement through data visualization technology, provides a global analysis perspective for training managers, and supports comparative analysis with talent training data of other coal mining companies in the industry.

[0047] The above-mentioned embodiments only express the preferred implementation modes of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present invention. It should be pointed out that, for a person skilled in the art, several modifications, improvements and substitutions can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the present invention patent shall be subject to the attached claims.

Claims

1. The intelligent coal mining occupation identification and training system is characterized by: include: Intelligent platform for fully mechanized mining working face, theoretical knowledge learning module, training assessment module and training effect evaluation module; The intelligent platform of fully mechanized mining face is used to build the actual intelligent mining environment of fully mechanized mining face and display the real scene of intelligent mining in coal mines; Theoretical knowledge learning module is used to provide the staff with relevant knowledge content of the intelligent platform of the fully mechanized mining face, so that the staff can learn relevant knowledge content at any time without being restricted by time and place; The training and assessment module is used to provide staff with a simulated operating environment so that they can conduct simulated exercises; it also records and evaluates the theoretical knowledge and practical operating ability of staff throughout the process, and gives a quantitative operating ability score; The training effectiveness evaluation module is used to comprehensively analyze and evaluate the training effectiveness of staff based on their theoretical learning and practical training, and provide targeted improvement suggestions.

2. The coal mine intelligent mining occupation identification and training system according to claim 1 is characterized in that: The intelligent platform of fully mechanized mining face includes 3D visualization module, virtual reality interaction module, data communication module and fault diagnosis module; The 3D visualization module is used to construct the 3D visualization scene of the intelligent platform of the fully mechanized mining face and realize the 3D visualization display of the fully mechanized mining face equipment and environment; Virtual reality interaction module, which is used to use virtual reality technology to realize the interactive operation between the staff and the intelligent platform of the fully mechanized mining face; A data communication module is used to communicate data with the actual fully mechanized mining working face, receive and store sensor data and equipment operating condition information of the actual fully mechanized mining working face; The fault diagnosis module is used to simulate common faults in actual fully mechanized mining working faces and display them through the 3D visualization module and virtual reality interaction module to guide the staff to diagnose and eliminate faults; Through data communication, the data communication module synchronizes the sensor data and equipment operating condition information of the actual fully mechanized mining working face to the coal mine intelligent mining professional identification and training system. The three-dimensional visualization module, virtual reality interaction module and fault diagnosis module call the actual fully mechanized mining working face information saved in the coal mine intelligent mining professional identification and training system to realize the linkage between simulation training and the actual fully mechanized mining working face.

3. The coal mine intelligent mining occupation identification and training system according to claim 2 is characterized in that: The fault diagnosis module establishes a fault simulation model based on the operating data and fault data of the actual fully mechanized mining working face, automatically generates the characteristic parameters and occurrence patterns of various common faults, as well as the abnormal operating condition data of the equipment when the fault occurs. Through the three-dimensional visualization module and the virtual reality interaction module, it simulates and displays the possible faults that may occur in the actual fully mechanized mining working face, and guides the staff to diagnose and eliminate the faults.

4. The coal mine intelligent mining occupation identification and training system according to claim 1 is characterized in that: The theoretical knowledge learning module generates a staff portrait for the staff through the theoretical knowledge recommendation model. The characteristic dimensions of the staff portrait include knowledge base, learning ability and operational skills. A personalized training plan is dynamically generated based on the staff portrait, and the learning content, progress, difficulty and presentation method are adjusted in real time during the training process. The theoretical knowledge recommendation model is built using machine learning algorithms and historical training data. It intelligently generates customized theoretical learning plans based on the personal characteristics of staff members and continuously optimizes itself as learning behavior data accumulates. The theoretical knowledge learning module uses the three-dimensional visualization module in the intelligent platform of the comprehensive mining working face to display the working principles and operating essentials of the comprehensive mining working face, and displays the equipment structure and process flow in a virtual environment, thereby strengthening the staff's intuitive understanding of theoretical knowledge.

5. The coal mine intelligent mining occupation identification and training system according to claim 4 is characterized in that: The theoretical knowledge recommendation model analyzes the learning behavior data of staff, identifies the weak links of staff in theoretical knowledge learning, automatically optimizes personalized training plans, and conducts targeted intensive training for staff; The specific steps of the theoretical knowledge recommendation model to achieve personalized training program optimization are as follows: S101: The theoretical knowledge recommendation model analyzes the learning behavior data of the staff and obtains the learning behavior sequence of the staff , where x t represents the learning behavior feature vector of the staff at time node t, and time node t is a time point in the staff's learning process; S102: Use the gated recurrent unit GRU network to model the knowledge state transformation of the staff, using h t Indicates the staff's mastery of each knowledge point; S103: Use the attention mechanism to aggregate the representations of different knowledge points, consider the dependencies between different knowledge points, and obtain the comprehensive knowledge state representation S of the staff member t ; ； in, represents the attention weight of knowledge point i at time node t; h t Indicates the staff's mastery of each knowledge point; Indicates the staff's mastery of knowledge point i at time node t; Represents the dependency matrix between knowledge point i and knowledge point j; Represents the attention weight of knowledge point j at time node t; Indicates the staff's mastery of knowledge point j at time node t; S104: A multi-layer perceptron is used to predict the staff's mastery of each knowledge point, and the difficulty of different knowledge points is reflected by the knowledge point difficulty vector. The specific calculation formula for the mastery of the knowledge point is: ； in, It represents the probability of staff mastering knowledge point i at time node t, which is used to optimize the subsequent personalized training strategy; Represents the learning behavior sequence and knowledge points that the staff has mastered before Under the condition of , predict whether the staff has mastered the knowledge point i at time node t; represents the set of knowledge points that the staff has mastered before time node t; σ represents the probability of converting the result of linear combination into the knowledge point mastered by the staff; is a weight vector, which indicates that at time point t, for knowledge point i, the model is used to measure the importance of the input feature; It represents the knowledge status of the staff member on knowledge point i at time point t; represents the state vector at time point t; Represents the knowledge point difficulty vector, which is used to reflect the difficulty of different knowledge points; is a constant used to adjust the output value so that when all input features are zero, the model can still predict a non-zero probability value; S105: By analyzing the staff's mastery of various knowledge points, identifying the staff's weak links in theoretical knowledge learning, automatically optimizing personalized training plans, and conducting targeted intensive training for the staff.

6. The coal mine intelligent mining occupation identification and training system according to claim 1 is characterized in that: The training and assessment module includes an operation practice unit. The operation practice unit combines with the intelligent platform of the comprehensive mining working face to provide the staff with a virtual environment practice operation. The virtual environment practice operation includes routine working condition simulation and abnormal working condition simulation. The routine working condition simulation guides the staff to conduct standardized operation training based on the actual work flow of the comprehensive mining working face; the abnormal working condition simulation randomly generates emergencies in the comprehensive mining working face to test the staff's emergency response and handling capabilities.

7. The coal mine intelligent mining occupation identification and training system according to claim 6 is characterized in that: The operation practice unit also conducts real-time analysis and diagnosis of the staff's operating behavior based on artificial intelligence algorithms. By capturing the staff's operations in the virtual environment, it judges the standardization and effectiveness of each operation action, and gives corresponding prompts and feedback to help the staff promptly discover and correct incorrect operating habits. At the same time, the operation practice unit can connect to multiple staff virtual reality terminals at the same time, allowing the staff to carry out team system operation training and practice information sharing, task coordination and emergency equipment capabilities in the comprehensive mining face.

8. The coal mine intelligent mining occupation identification and training system according to claim 6 is characterized in that: The training assessment module also includes an assessment unit, which records the actual operating ability of the staff throughout the process and uses the training results evaluation model to automatically evaluate and score the staff's operating behavior; The training outcome evaluation model is based on the key steps and requirements in the operating procedures of the fully mechanized mining face. It determines whether each operation of the staff is standardized and complies with the safe production process, and makes quantitative scores. Among them, the training outcome evaluation model uses weighted summation to achieve automatic evaluation and scoring. The specific calculation formula is: ； in, is the scoring result of the training outcome evaluation; S is the scoring item of the operation steps, which is used to reflect whether the staff operates in accordance with the requirements of the regulations; α is the step score weight, which is used to reflect the importance of the operation steps; D is the safety hazard deduction item, which is used to cultivate the safety awareness of the staff by deducting points for safety hazards; β is the safety hazard weight, which is used to reflect the importance of safety hazards; P is the personalized adjustment factor, which is used to dynamically adjust the scoring results according to the specific situation of the staff, so as to improve the pertinence of the training outcome evaluation.

9. The coal mine intelligent mining occupation identification and training system according to claim 1 is characterized in that: The training effect evaluation module adopts an incremental learning algorithm. As the theoretical learning and practical training data of staff members continue to accumulate, the theoretical knowledge recommendation model and personalized training strategy in the theoretical knowledge learning module are continuously optimized; The specific execution steps of the training effect evaluation module are: S201: continuously collect the training data generated by the staff in the theoretical learning module and the operation practice unit, and perform preprocessing operations on the training data to improve the data quality; S202: Filter out key features that affect the training effect from the preprocessed training data using statistical analysis and machine learning methods. The key features include the mastery of important knowledge points, the gap between the learning progress and the average level, and the error rate during the practical operation. S203: using an incremental learning algorithm, based on the original theoretical knowledge recommendation model and personalized training strategy, and using the newly collected training data to fine-tune and update the original theoretical knowledge recommendation model and personalized training strategy; S204: Based on the fine-tuned and updated theoretical knowledge recommendation model and personalized training strategy in step S203, the theoretical knowledge recommendation model is optimized, and the optimization includes adjusting the recommendation weights of knowledge points and optimizing the learning path; S205: improving the personalized training strategy based on the optimized theoretical knowledge recommendation model, and formulating a targeted personalized training plan based on the actual learning situation; S206: Continue to collect the latest training data of the staff and execute step S203 regularly.

10. The coal mine intelligent mining occupation identification and training system according to claim 1 is characterized in that: It also includes a training effect visualization module, which uses data visualization technology to intuitively present each staff member's theoretical knowledge mastery, practical skill level, and comprehensive ability improvement, providing a global analysis perspective for training managers and supporting comparative analysis with talent training data from other coal mining companies in the industry.

Images