Coal seam fully-mechanized mining hydraulic support robot digital twinborn teaching system

Abstract

The invention discloses a coal seam fully-mechanized mining hydraulic support robot digital twinborn teaching system, and relates to the technical field of coal industry teaching, and the system comprises the following components: a model construction module, a virtual simulation and operation module, a fault simulation and diagnosis module, a data interaction and remote monitoring module, and a teaching management and evaluation module. According to the invention, a comprehensive and systematized scheme is provided for the teaching of the coal seam fully-mechanized mining hydraulic support robot through integrating a plurality of modules of model construction, virtual simulation and operation, fault simulation and diagnosis, data interaction and remote monitoring, and teaching management and evaluation; students can perform operation practice and troubleshooting training of equipment in a virtual environment without contacting actual equipment, so that teaching cost is reduced, safety is improved, the system can feed back operation results and evaluation scores of the students in real time, teachers can know learning conditions of the students in time, targeted teaching guidance is performed, and teaching efficiency is improved. And the teaching effect and the learning efficiency are obviously improved.

Description

Technical Field

[0001] The present invention relates to the field of coal industry teaching technology, and in particular to a digital twin teaching system for a coal seam fully-mechanized mining hydraulic support robot. Background Art

[0002] With the rapid development of the coal industry, coal seam fully-mechanized mining technology has increasingly become the key to improving coal mine production efficiency and safety. As the core equipment in the coal seam fully-mechanized mining process, the importance of hydraulic support robots is self-evident. However, the operation of hydraulic support robots is complex and requires high skills in troubleshooting and repair. This makes it particularly important to provide efficient and safe training for relevant technical personnel.

[0003] There are many shortcomings in the traditional teaching and training of coal seam fully mechanized mining hydraulic support robots. First, the training method is single and mostly relies on on-site field operations, which makes the training cost high and poses safety hazards. Once the operation is improper, it may cause equipment damage or even casualties. Secondly, the training effect is limited. Due to the limited opportunities to actually operate the equipment, it is difficult for trainees to fully master the various operating skills and troubleshooting methods of the equipment. In addition, the traditional training method lacks real-time feedback and evaluation mechanisms, and it is difficult for teachers to understand the trainees' learning situation in a timely manner, resulting in poor teaching results. Finally, the traditional training method is difficult to simulate various complex working conditions and fault scenarios, which limits the trainees' skill improvement and ability to deal with complex situations.

[0004] In summary, the traditional teaching and training methods of coal seam fully-mechanized mining hydraulic support robots have the problems of high cost, high risk and limited effect. In order to overcome these shortcomings, it is particularly important to develop a digital twin teaching system for coal seam fully-mechanized mining hydraulic support robots. Summary of the Invention

[0005] The purpose of this invention is to make up for the shortcomings of the existing technology and provide a digital twin teaching system for coal seam fully mechanized mining hydraulic support robots. It can provide an immersive and interactive training experience by simulating various complex working conditions and fault scenarios, helping students to fully master the equipment's operating skills and troubleshooting methods in a virtual environment. At the same time, the system has a real-time feedback and evaluation mechanism, which helps teachers to understand students' learning situation in a timely manner and provide targeted teaching guidance.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: a digital twin teaching system for a fully-mechanized hydraulic support robot in a coal seam, the system comprising the following components: a model building module, a virtual simulation and operation module, a fault simulation and diagnosis module, a data interaction and remote monitoring module, and a teaching management and evaluation module;

[0007] The model building module uses SolidWorks 3D modeling software to create an accurate 3D digital model based on the actual drawings and structural parameters of the hydraulic support robot. The modeling process covers the construction of an exterior model, an internal structure model, and detailed component models to ensure that the model truly reflects the physical characteristics and motion patterns of the equipment. At the same time, a parametric modeling method is used to set the geometric and physical parameters of the model as variable parameters, facilitating rapid adjustment and optimization of the model according to different needs. Models of different specifications can be generated by changing parameters such as support height, column diameter, and component size.

[0008] The virtual simulation and operation module: builds a virtual simulation operation environment based on the digital twin model, designs an intuitive operation interface, provides start, stop, move, lift, and telescope operation buttons, and the system provides real-time feedback on operation results to help students understand the operation process and the working principle of the equipment. By setting equipment initialization, troubleshooting, and comprehensive application tasks, students' operation skills are improved in a task-driven manner. In addition, with the help of virtual simulation technology, an immersive interactive training experience is provided. Students can operate through a mouse, keyboard, or VR device. The system responds and feedbacks the results in real time, and provides task guidance and error prompts to assist students in learning.

[0009] The fault simulation and diagnosis module: designs corresponding fault simulation algorithms based on the types and characteristics of mechanical, electrical, and hydraulic faults that may occur in hydraulic support robots. At the same time, it formulates fault diagnosis strategies, provides virtual diagnostic tools, diagnostic prompts, and repair guidance, guides students in troubleshooting and repairing faults, and improves students' fault handling capabilities.

[0010] The data interaction and remote monitoring module enables data interaction between the digital twin model and the actual hydraulic support robot. It collects operating status data through sensors installed on the actual equipment, transmits the data using communication technology, processes and analyzes the data, and updates the digital twin model status in real time. Students can remotely monitor the robot's operating status through the system, perform remote operation and debugging, send control commands, and adjust and optimize equipment parameters.

[0011] The teaching management and evaluation module provides teaching management functions, including student information management, course management, teaching resource management and teaching progress management. At the same time, it has a teaching evaluation function, which comprehensively evaluates students' learning outcomes through theoretical knowledge tests, practical operation assessments, and learning attitude evaluations, generates a comprehensive evaluation report, and provides a basis for teachers to improve their teaching.

[0012] Furthermore, when the model building module uses SolidWorks 3D modeling software to build the model, an adaptive parameter fusion algorithm is used to determine the model parameters. The algorithm first quantifies the physical characteristics and working environment factors of the hydraulic support robot and constructs an initial parameter set. ,in Indicates the Initial parameters, at the same time, the environmental impact factor is introduced , Indicates the The degree of influence of various environmental factors on model parameters. During the modeling process, the algorithm dynamically adjusts the parameters according to the environmental data collected in real time. Specifically, the real-time parameters are calculated using the following formula: :

[0013]

[0014] in It's environmental factors Parameters The weight coefficient, weight coefficient The determination adopts the historical data learning method. By analyzing a large amount of historical operation data, the decision tree algorithm in machine learning is used to train with the accuracy of parameter adjustment as the objective function to obtain the weight of each environmental factor on different parameters. In addition, in order to ensure the adaptability of the model under different working conditions, the working condition factor is introduced. , further modify the real-time parameters:

[0015]

[0016] Operating Condition Factor Based on the real-time collected equipment operation status data, the fuzzy logic algorithm is used to determine that the equipment operation status is divided into different fuzzy subsets, and the value of the working condition factor is determined according to the preset fuzzy rules. The model constructed in this way can more accurately reflect the actual situation of the hydraulic support robot in different environments and working conditions, providing a more reliable foundation for subsequent virtual simulation and fault simulation.

[0017] Furthermore, in the virtual simulation and operation module, a more realistic operation feedback and task evaluation are achieved, and an operation evaluation algorithm based on behavior feature matching is adopted. The algorithm first decomposes the standard operation process of the hydraulic support robot, extracts the key behavior features of each operation step, and constructs a standard behavior feature library. ,in Indicates the When students perform virtual operations, the system collects students' operation behavior data in real time and constructs students' operation behavior feature vectors. ,The accuracy of the student’s operation is evaluated by calculating the similarity between the student’s operation behavior feature vector and each vector in the standard behavior feature library. The similarity calculation adopts the improved cosine similarity formula:

[0018]

[0019] in and They are the standard behavioral feature vectors and student operation behavior feature vector No. elements, It is an adjustment coefficient, which is determined by optimizing experimental data. At the same time, the time factor of the operation is taken into account and the time weight is introduced. , weight the evaluation results of each operation step, time weight Determined based on the ratio of standard operation time to students’ actual operation time:

[0020]

[0021] The final operation evaluation score is:

[0022]

[0023] In this way, the system can evaluate students' operational skills more comprehensively and accurately, and provide students with more targeted feedback and guidance.

[0024] Furthermore, the fault simulation and diagnosis module adopts a fault diagnosis algorithm based on multi-source information fusion, which integrates the equipment status data, historical fault data and expert experience knowledge. First, the status data of the equipment is feature extracted to construct the status feature vector. ,in Indicates the At the same time, historical fault data is analyzed to build a fault mode library ,in Indicates the Failure modes, introducing expert experience knowledge, and building a knowledge rule base ,in Indicates the Knowledge rules, when diagnosing faults, first calculate the state feature vector The degree of match with each failure mode in the failure mode library , using the Euclidean distance formula:

[0025]

[0026] in is the state eigenvector No. The first feature A quantity, Is the failure mode No. components, and then, according to the knowledge rule base Correct the matching degree for each knowledge rule , if the rule conditions are met, the corresponding matching degree is adjusted: ,in It is the adjustment coefficient determined according to the knowledge rule. Finally, the fault type is determined according to the corrected matching degree, and the fault mode with the smallest matching degree is selected as the diagnosis result. At the same time, the reliability of the diagnosis is improved and the confidence level is introduced. , the confidence is calculated based on the distribution of matching degrees and the degree of satisfaction of knowledge rules:

[0027]

[0028] Through this multi-source information fusion fault diagnosis algorithm, the fault of the hydraulic support robot can be diagnosed more accurately and the efficiency of fault handling can be improved.

[0029] Furthermore, in the data interaction and remote monitoring module, in order to ensure the real-time and reliability of data transmission, a data transmission algorithm based on dynamic priority scheduling is adopted. This algorithm assigns dynamic priorities to different types of data according to the importance and timeliness of the data. First, the data is divided into key data , important data and general data Three categories, for key data, the real-time location, posture and pressure data of the device, its priority Determine based on the rate of change of data and the degree of impact on the safe operation of the equipment:

[0030]

[0031] in is the change in key data, is the current value of the key data, It is the impact index of key data on the safe operation of equipment and is the weight coefficient, which is determined by experiments and experience. Determine based on the data update cycle and the impact on device performance:

[0032]

[0033] in It is the update cycle of important data. It is the impact index of important data on equipment performance. and is the weight coefficient for general data, its priority Relatively low, determined by data storage requirements and transmission frequency:

[0034]

[0035] in is the storage size of general data, is the general data transmission frequency, and It is a weight coefficient. During the data transmission process, scheduling is performed according to the priority of the data, and high-priority data is transmitted first to ensure that key data can be transmitted to the digital twin model in a timely and accurate manner, providing support for real-time monitoring and decision-making of the system.

[0036] Furthermore, the teaching management and evaluation module adopts a comprehensive evaluation algorithm based on learning trajectory analysis. This algorithm records the learning trajectory of students in the process of theoretical teaching, practical teaching and extended learning, including learning time, learning content, and operation records, and constructs a student learning trajectory vector. ,in Indicates the First, cluster analysis is performed on the learning trajectory vectors to divide students into different learning groups. The K-means clustering algorithm is used to determine the cluster centers with learning efficiency and learning effect as the objective function. , each cluster center represents a typical learning mode, then calculate the distance between each student's learning trajectory vector and each cluster center , using the Manhattan distance formula:

[0037]

[0038] in is the student learning trajectory vector No. The first feature A quantity, is the cluster center No. Components, based on the distance, determine the learning group to which the student belongs. The learning group corresponding to the cluster center with the smallest distance is selected as the group to which the student belongs. For each learning group, a corresponding evaluation model is established. The evaluation model comprehensively considers the students' theoretical knowledge mastery, practical operation skills, learning attitude and innovation ability factors. Through the multiple linear regression model, with the student's final score as the dependent variable and each evaluation factor as the independent variable, the weight of each evaluation factor is determined:

[0039]

[0040] in It is evaluation factors, is its weight. The comprehensive evaluation algorithm based on learning trajectory analysis can comprehensively evaluate students' learning situation and provide a basis for teaching management and personalized teaching.

[0041] Furthermore, in the virtual simulation operation environment, in order to enhance the immersion and interactivity of students, an interactive optimization algorithm based on physiological feedback is adopted. This algorithm collects students' physiological data and monitors students' emotional state and attention level in real time. First, a physiological feature vector is constructed. ,in Indicates the The system uses the support vector machine algorithm in machine learning to classify physiological feature vectors and divide students' emotional states into different categories: excitement, calmness, and tension. At the same time, it evaluates students' attention levels and dynamically adjusts the parameters of the virtual simulation operation environment according to their emotional states and attention levels. When students are nervous, the difficulty of the operation tasks is reduced and prompt information is added. When students are not paying attention, the color and sound of the operation interface are changed to attract their attention. Through this interactive optimization algorithm based on physiological feedback, the teaching environment can be adjusted according to the students' real-time status, thereby improving their learning experience and learning effects.

[0042] Furthermore, in the fault simulation and diagnosis module, the authenticity and diversity of fault simulation are improved by adopting a fault simulation algorithm based on random disturbance and scenario combination. The algorithm first models the normal and operating state of the hydraulic support robot to obtain the normal state model. , then, introduce the random perturbation factor , randomly perturb the parameters of the normal state model. The value range of the random perturbation factor is determined according to the actual operation of the equipment and the possibility of failure. The fault state model is generated by the following formula

[0043]

[0044] At the same time, in order to simulate more complex fault scenarios, a scenario combination method is used to combine different fault types and fault degrees to build a fault scenario library. ,in Indicates the A fault scenario is randomly selected from the fault scenario library during fault simulation. The virtual model of the hydraulic support robot is adjusted according to the requirements of the scenario to generate the corresponding fault state. At the same time, in order to ensure the repeatability and traceability of the fault simulation, the parameters and generation process of each fault scenario are recorded. Through this fault simulation algorithm based on random perturbations and scenario combinations, more realistic and complex fault scenarios can be generated, thereby improving students' fault handling capabilities and ability to deal with complex situations.

[0045] Furthermore, in the data interaction and remote monitoring module, in order to achieve predictive analysis of the equipment operation status, a prediction algorithm based on the fusion of time series analysis and machine learning is adopted. The algorithm first performs time series analysis on the collected equipment status data, and uses the autoregressive integral moving average model to model and predict the data. Assume that the time series of the equipment status data is , the expression of the ARIMA model is:

[0046]

[0047] in is the autoregressive order, is the difference order, is the moving average order, is the lag operator, and are model parameters, It is a white noise sequence. The parameters of the ARIMA model are determined by training historical data. 、 and , use the trained ARIMA model to make short-term predictions on the equipment status data and get the predicted value In order to improve the accuracy of prediction, a machine learning algorithm is introduced. The predicted value of the ARIMA model and the original data are used as the input of the LSTM network to make a one-step prediction of the device status. The output of the LSTM network is the final predicted value. At the same time, in order to evaluate the accuracy of the prediction, the mean square error (MSE) and mean absolute percentage error (MAPE) are introduced as evaluation indicators:

[0048]

[0049]

[0050] By continuously adjusting the parameters of the ARIMA model and LSTM network, the MSE and MAPE are minimized, thereby improving the accuracy of the prediction. This prediction algorithm based on the fusion of time series analysis and machine learning can predict the operating status of the equipment in advance and provide decision support for equipment maintenance and management.

[0051] Compared with the existing technology, this coal seam fully-mechanized mining hydraulic support robot digital twin teaching system has the following beneficial effects:

[0052] 1. The system provides a comprehensive and systematic solution for the teaching of coal seam fully-mechanized mining hydraulic support robots by integrating multiple modules, including model construction, virtual simulation and operation, fault simulation and diagnosis, data interaction and remote monitoring, as well as teaching management and evaluation. Students can practice equipment operation and troubleshooting in a virtual environment without having to touch the actual equipment, thereby reducing teaching costs and improving teaching safety. At the same time, the system can provide real-time feedback on students' operation results and evaluation scores, which helps teachers understand students' learning situation in a timely manner and provide targeted teaching guidance, significantly improving teaching effectiveness and learning efficiency.

[0053] 2. The system adopts advanced digital twin technology and a variety of intelligent algorithms, such as adaptive parameter fusion algorithm, operation evaluation algorithm based on behavioral feature matching, and fault diagnosis algorithm based on multi-source information fusion, to achieve accurate simulation and real-time monitoring of the operating status of the hydraulic support robot. This not only helps students to deeply understand the working principle and operation process of the equipment, but also helps them master the skills of troubleshooting and repair. Through learning and training in this system, students can better adapt to the actual working environment and lay a solid foundation for their future career development.

[0054] Other advantages, objects and features of the present invention will be described in part in the following description and, in part, will be apparent to those skilled in the art based on an examination of the following or may be learned from the practice of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] To more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. Those skilled in the art can also derive other drawings based on these drawings without inventive effort.

[0056] Figure 1 This is a function realization flow chart of a digital twin teaching system for a fully-mechanized hydraulic support robot in coal seam mining;

[0057] Figure 2This is a flow chart of the overall architecture of a digital twin teaching system for a coal seam fully mechanized mining hydraulic support robot. DETAILED DESCRIPTION

[0058] In order to further illustrate the technical means and effects adopted by the present invention to achieve the predetermined purpose of the invention, the specific implementation methods, structures, features and effects of the present invention are described in detail below in conjunction with the accompanying drawings and preferred embodiments.

[0059] Example 1

[0060] This embodiment describes a coal mining equipment practical training course at a coal vocational and technical school, in which students need to learn how to operate a hydraulic support robot. Because actual equipment operation involves certain risks and is costly, the school uses a digital twin teaching system for coal seam fully-mechanized mining hydraulic support robots for virtual simulation operation training.

[0061] Using SolidWorks 3D modeling software, we can create an accurate 3D digital model based on the actual drawings and structural parameters of the hydraulic support robot. When building the model, we use the adaptive parameter fusion algorithm to determine the model parameters. For example, under a specific working condition, the initial parameter set is known. , environmental impact factors , according to the real-time collected environmental data (such as temperature, humidity), through the formula Calculate real-time parameters , assuming the current temperature For a key structural parameter The weight coefficient ,humidity Its weight coefficient , , (After quantization, it is 2.5), (After quantization, it is 0.6), then , and then introduce the operating condition factor , through the fuzzy logic algorithm to determine its value according to the real-time collected equipment operation status data, and further correct the real-time parameters .

[0062] Students use VR devices to operate in a virtual simulation environment. The system provides operation buttons such as start, stop, move, lift, and retract. For example, when performing a bracket lifting operation, students operate the VR handle to simulate the operation action. The system provides real-time feedback on the operation results, allowing students to intuitively see the lifting and lowering changes of the bracket. The system uses an operation evaluation algorithm based on behavioral feature matching to evaluate student operations. First, a standard behavioral feature library is constructed. When students are operating, the operation behavior data is collected in real time to construct the student operation behavior feature vector , through the improved cosine similarity formula Calculate the similarity, assuming , for a certain operation step, ,but At the same time, the operation time factor is taken into consideration and time weight is introduced. , final operation evaluation score .

[0063] Teachers use this module to manage student information and course content. After the training course, they will comprehensively evaluate students' learning outcomes through theoretical knowledge tests, practical operation assessments, and learning attitude evaluations. For example, theoretical knowledge tests account for 30% of the total score, practical operation assessments account for 50%, and learning attitude evaluations account for 20%. The final score of each student is calculated comprehensively to generate a comprehensive evaluation report.

[0064] Example 2

[0065] This embodiment describes an internal training center at a coal mining enterprise where newly hired technicians need to learn fault diagnosis and repair skills for hydraulic support robots. The training center uses the digital twin teaching system to conduct fault simulation and diagnosis teaching.

[0066] The fault simulation algorithm based on random disturbance and scenario combination is used to first model the normal operation state of the hydraulic support robot and obtain the normal state model. , introducing a random perturbation factor , through the formula Generate a fault state model, for example, randomly perturb the pressure parameters of a key component, assuming that the normal pressure value is , random perturbation factor , then the pressure value in the fault state becomes At the same time, build a fault scenario library ,During fault simulation, a fault scenario is randomly selected from the fault scenario library, such as selecting scenario ,This scenario requires the simultaneous occurrence of electrical faults and hydraulic leakage faults. According to this requirement, the virtual model is adjusted to generate the corresponding fault state, and the parameters and generation process of the fault scenario are recorded. During fault diagnosis, a fault diagnosis algorithm based on multi-source information fusion is used to extract features from the equipment state data and construct a state feature vector. , analyze historical fault data and build a fault mode library , introduce expert experience knowledge and build a knowledge rule base , through the Euclidean distance formula Calculate the state eigenvector The matching degree of each failure mode in the failure mode library, assuming , ,but , and then according to the knowledge rule base Correct the matching degree, for knowledge rules that meet the rule conditions , adjust the matching degree Finally, the fault type is determined based on the corrected matching degree, and the fault mode with the smallest matching degree is selected as the diagnosis result. At the same time, the confidence level is introduced. Assess the reliability of diagnostic results.

[0067] The training teacher uses this module to record the students' operation records, learning time and other learning trajectory data during the fault diagnosis learning process, and uses a comprehensive evaluation algorithm based on learning trajectory analysis to construct the student learning trajectory vector. , cluster analysis is performed on the learning trajectory vector using the K-means clustering algorithm, the students are divided into different learning groups, and the distance between each student's learning trajectory vector and each cluster center is calculated ,According to the distance, the learning group to which the trainee belongs is determined. For each learning group, a corresponding evaluation model is established. The weight of each evaluation factor is determined through the multiple linear regression model to evaluate the trainee's performance in fault diagnosis learning.

[0068] The above description is merely a preferred embodiment of the present invention and does not constitute any form of limitation to the present invention. Although the present invention has been disclosed as above in terms of a preferred embodiment, it is not intended to limit the present invention. Any person skilled in the art can, without departing from the scope of the technical solution of the present invention, make some changes or modifications to equivalent embodiments using the technical contents disclosed above. However, any brief modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention are still within the scope of the technical solution of the present invention.

Claims

1. A digital twin teaching system for coal seam fully mechanized mining hydraulic support robot, characterized in that: The system includes the following components: model building module, virtual simulation and operation module, fault simulation and diagnosis module, data interaction and remote monitoring module, and teaching management and evaluation module; The model building module uses SolidWorks 3D modeling software to create a 3D digital model based on the actual drawings and structural parameters of the hydraulic support robot. The modeling process covers the construction of the appearance model, the internal structure model and the detailed component models. At the same time, the parametric modeling method is used to set the geometric parameters and physical parameters of the model as variable parameters. The virtual simulation and operation module: builds a virtual simulation operation environment based on the digital twin model, designs an intuitive operation interface, provides start, stop, move, lift, and telescope operation buttons, and the system provides real-time feedback on operation results to help students understand the operation process and the working principle of the equipment. By setting equipment initialization, troubleshooting, and comprehensive application tasks, students' operation skills are improved in a task-driven manner. In addition, with the help of virtual simulation technology, an immersive interactive training experience is provided. Students can operate through a mouse, keyboard, or VR device. The system responds and feedbacks the results in real time, and provides task guidance and error prompts to assist students in learning. The fault simulation and diagnosis module: designs corresponding fault simulation algorithms based on the types and characteristics of mechanical, electrical, and hydraulic faults that occur in hydraulic support robots. At the same time, it formulates fault diagnosis strategies, provides virtual diagnostic tools, diagnostic prompts, and repair guidance, and guides students to troubleshoot and repair faults. The data interaction and remote monitoring module realizes data interaction between the digital twin model and the actual hydraulic support robot, collects operating status data through sensors installed on the actual equipment, transmits the data using communication technology, processes and analyzes the data, and updates the digital twin model status in real time; The teaching management and evaluation module provides teaching management functions and, at the same time, has teaching evaluation functions. It comprehensively evaluates students' learning outcomes through theoretical knowledge tests, practical operation assessments, and learning attitude evaluations, and generates a comprehensive evaluation report.

2. A digital twin teaching system for coal seam fully mechanized mining hydraulic support robots according to claim 1, characterized in that: When the model building module uses SolidWorks 3D modeling software to build the model, the adaptive parameter fusion algorithm is used to determine the model parameters. The algorithm first quantifies the physical characteristics and working environment factors of the hydraulic support robot and constructs the initial parameter set. ,in Indicates the Initial parameters, at the same time, the environmental impact factor is introduced , Indicates the The degree of influence of various environmental factors on model parameters. During the modeling process, the algorithm dynamically adjusts the parameters according to the environmental data collected in real time. Specifically, the real-time parameters are calculated using the following formula: : in It's environmental factors Parameters In addition, the adaptability of the model under different working conditions introduces the working condition factor , further modify the real-time parameters: Operating Condition Factor According to the real-time collected equipment operation status data, the fuzzy logic algorithm divides the equipment operation status into different fuzzy subsets, and determines the value of the operating condition factor according to the preset fuzzy rules.

3. The digital twin teaching system for coal seam fully mechanized mining hydraulic support robot according to claim 1 is characterized in that: In the virtual simulation and operation module, a more realistic operation feedback and task evaluation are achieved by adopting an operation evaluation algorithm based on behavior feature matching. The algorithm first decomposes the standard operation process of the hydraulic support robot, extracts the key behavior features of each operation step, and constructs a standard behavior feature library. ,in Indicates the When students perform virtual operations, the system collects students' operation behavior data in real time and constructs students' operation behavior feature vectors. ,The accuracy of the student’s operation is evaluated by calculating the similarity between the student’s operation behavior feature vector and each vector in the standard behavior feature library. The similarity calculation adopts the improved cosine similarity formula: in and They are the standard behavioral feature vectors and student operation behavior feature vector No. elements, It is an adjustment coefficient. At the same time, considering the time factor of the operation, the time weight is introduced , weight the evaluation results of each operation step: The final operation evaluation score is: .

4. The digital twin teaching system for coal seam fully mechanized mining hydraulic support robot according to claim 1 is characterized in that: In the fault simulation and diagnosis module, a fault diagnosis algorithm based on multi-source information fusion is adopted. First, the state data of the equipment is feature extracted to construct the state feature vector ,in Indicates the At the same time, historical fault data is analyzed to build a fault mode library ,in Indicates the Failure modes, introducing expert experience knowledge, and building a knowledge rule base ,in Indicates the Knowledge rules, when diagnosing faults, first calculate the state feature vector The degree of match with each failure mode in the failure mode library , using the Euclidean distance formula: in is the state eigenvector No. The first feature A quantity, Is the failure mode No. components, and then, according to the knowledge rule base Correct the matching degree for each knowledge rule , if the rule conditions are met, the corresponding matching degree is adjusted: ,in It is the adjustment coefficient determined according to the knowledge rule. Finally, the fault type is determined according to the corrected matching degree, and the fault mode with the smallest matching degree is selected as the diagnosis result. At the same time, the confidence level is introduced. , the confidence is calculated based on the distribution of matching degrees and the degree of satisfaction of knowledge rules: .

5. The digital twin teaching system for coal seam fully mechanized mining hydraulic support robot according to claim 1 is characterized in that: In the data interaction and remote monitoring module, a data transmission algorithm based on dynamic priority scheduling is adopted. This algorithm assigns dynamic priorities to different types of data according to the importance and timeliness of the data. First, the data is divided into key data , important data and general data Three categories, for key data, the real-time location, posture and pressure data of the device, its priority Determine based on the rate of change of data and the degree of impact on the safe operation of the equipment: in is the change in key data, is the current value of the key data, It is the impact index of key data on the safe operation of equipment and is the weight coefficient, its priority Determine based on the data update cycle and the impact on device performance: in It is the update cycle of important data. It is the impact index of important data on equipment performance. and is the weight coefficient for general data, its priority Relatively low: in is the storage size of general data, is the general data transmission frequency, and It is the weight coefficient. During data transmission, scheduling is performed according to the priority of the data, and data with high priority is transmitted first.

6. The digital twin teaching system for coal seam fully mechanized mining hydraulic support robot according to claim 1 is characterized in that: The teaching management and evaluation module adopts a comprehensive evaluation algorithm based on learning trajectory analysis. This algorithm records students' learning trajectories in the process of theoretical teaching, practical teaching and extended learning, including learning time, learning content, and operation records, and constructs a student learning trajectory vector. ,in Indicates the First, cluster analysis is performed on the learning trajectory vectors to divide students into different learning groups. K-means clustering algorithm is used to determine the cluster centers. , each cluster center represents a typical learning mode, then calculate the distance between each student's learning trajectory vector and each cluster center , using the Manhattan distance formula: in is the student learning trajectory vector No. The first feature A quantity, is the cluster center No. Components, determine the learning group to which the student belongs based on the distance, and select the learning group corresponding to the cluster center with the smallest distance as the group to which the student belongs. For each learning group, establish a corresponding evaluation model, and use the multiple linear regression model with the student's final score as the dependent variable and each evaluation factor as the independent variable to determine the weight of each evaluation factor: in It is evaluation factors, is its weight.

7. The digital twin teaching system for coal seam fully mechanized mining hydraulic support robot according to claim 1 is characterized in that: In the virtual simulation operation environment, an interactive optimization algorithm based on physiological feedback is adopted. This algorithm collects students' physiological data and monitors students' emotional state and attention level in real time. First, a physiological feature vector is constructed. ,in Indicates the The system uses the support vector machine algorithm in machine learning to classify physiological feature vectors and divide students' emotional states into different categories: excitement, calmness, and tension. At the same time, it evaluates students' attention levels and dynamically adjusts the parameters of the virtual simulation operation environment according to their emotional states and attention levels. When students are nervous, the difficulty of the operation tasks is reduced and prompt information is added. When students are not paying attention, the color and sound of the operation interface are changed to attract students' attention.

8. The digital twin teaching system for coal seam fully mechanized mining hydraulic support robot according to claim 1 is characterized in that: In the fault simulation and diagnosis module, a fault simulation algorithm based on random disturbance and scenario combination is adopted. The algorithm first models the normal and operating state of the hydraulic support robot to obtain the normal state model. , then, introduce the random perturbation factor , randomly perturb the parameters of the normal state model, and generate the fault state model by the following formula : At the same time, in order to simulate more complex fault scenarios, a scenario combination method is used to combine different fault types and fault degrees to build a fault scenario library. ,in Indicates the In the fault simulation, a fault scenario is randomly selected from the fault scenario library, and the virtual model of the hydraulic support robot is adjusted according to the requirements of the scenario to generate the corresponding fault state. At the same time, the parameters and generation process of each fault scenario are recorded.

9. The digital twin teaching system for coal seam fully mechanized mining hydraulic support robot according to claim 1 is characterized in that: In the data interaction and remote monitoring module, a prediction algorithm based on the fusion of time series analysis and machine learning is adopted. The algorithm first performs time series analysis on the collected equipment status data and uses the autoregressive integral moving average model to model and predict the data. Assume that the time series of the equipment status data is , the expression of the ARIMA model is: in is the autoregressive order, is the difference order, is the moving average order, is the lag operator, and are model parameters, It is a white noise sequence. The trained ARIMA model is used to make short-term predictions on the equipment status data to obtain the predicted value. , introduces machine learning algorithms, takes the predicted value of the ARIMA model and the original data as the input of the LSTM network, and makes a one-step prediction of the device status. The output of the LSTM network is the final predicted value At the same time, the mean square error (MSE) and mean absolute percentage error (MAPE) are introduced as evaluation indicators: By continuously adjusting the parameters of the ARIMA model and LSTM network, the MSE and MAPE are minimized.

Images