Electrical equipment operation training and examination system based on virtual technology

Abstract

The invention discloses an electrical equipment operation training examination system based on a virtual technology, and belongs to the field of intelligent examination systems, and the system comprises a virtual operation environment construction module which is used for collecting electrical equipment operation data to form an operation data set; the data preprocessing module is used for preprocessing the operation data set to obtain a preprocessed data set; the operation mode identification module is used for establishing an identification model of operation habits and behavior modes; the operation evaluation rule base module is used for constructing an evaluation rule and judging the normalization and effectiveness of the operation based on the evaluation rule; the virtual assessment scene library module is used for decomposing assessment content and performing skill assessment on an operator; and the risk assessment and fault diagnosis module is used for monitoring operation risks and identifying high-risk operations and potential faults.

Description

Technical Field

[0001] The invention belongs to the technical field of intelligent assessment systems, and in particular relates to an electrical equipment operation training and assessment system based on virtual technology. Background Art

[0002] In the training and assessment of electrical equipment operation, virtual technology faces the challenge of how to accurately assess the skill level of operators. Traditional assessment methods rely on manual observation and subjective judgment, and there are problems such as inconsistent assessment standards and insufficient objectivity of assessment results. The virtual system needs to be able to comprehensively collect the operator's operation data in the virtual environment, including operation steps, operation time, operation results, etc., and be able to analyze these data in real time to intelligently judge the standardization and effectiveness of the operation. However, there are differences in the operating habits and behavior patterns of different operators, and it is difficult for the virtual system to establish unified evaluation standards and rules. In addition, electrical equipment operation involves multiple links and scenarios, the operation steps are cumbersome, and the assessment content is complex and diverse. It is difficult for the virtual system to fully cover all assessment points. At the same time, electrical equipment operation has very high requirements for safety and reliability. The virtual system must be able to accurately judge the risks and hidden dangers of the operation, and give reasonable safety tips and improvement suggestions, which puts higher requirements on the intelligence level of the virtual system. Therefore, it is urgent to study a training and assessment method for electrical equipment operation based on virtual technology to solve technical problems such as inconsistent assessment standards, incomplete coverage of assessment content, and inaccurate assessment safety judgment, so as to achieve objective assessment and precise improvement of operating skills. Summary of the invention

[0003] In order to solve the above technical problems, the present invention proposes an electrical equipment operation training and assessment system based on virtual technology to solve the problems existing in the above prior art.

[0004] To achieve the above object, the present invention provides an electrical equipment operation training and assessment system based on virtual technology, comprising:

[0005] Virtual operating environment construction module, data preprocessing module, operating mode recognition module, operating evaluation rule base module, virtual assessment scenario base module and risk assessment and fault diagnosis module;

[0006] Wherein, the virtual operating environment construction module is used to collect electrical equipment operating data to form an operating data set;

[0007] The data preprocessing module is used to preprocess the operation data set to obtain a preprocessed data set;

[0008] The operation mode recognition module is used to establish a recognition model of operation habits and behavior patterns;

[0009] The operation evaluation rule base module is used to construct evaluation rules and judge the standardization and effectiveness of operations based on the evaluation rules;

[0010] The virtual assessment scenario library module is used to decompose the assessment content and conduct skill assessment on the operator;

[0011] The risk assessment and fault diagnosis module is used to monitor operational risks and identify high-risk operations and potential faults.

[0012] Optionally, the virtual operating environment construction module includes: an operating environment construction unit, an operating data acquisition unit and an original data self-checking unit;

[0013] Wherein, the operating environment construction unit is used to construct a three-dimensional model including electrical equipment and an operating interface;

[0014] The operation data collection unit is used to collect the operation data of the operator in the virtual operation environment; the operation data includes operation steps, operation time and operation results;

[0015] The original data self-checking unit is used to remove abnormal data from the operation data to obtain data in a unified format.

[0016] Optionally, the data preprocessing module includes: a data preprocessing unit and a data quality evaluation unit;

[0017] Wherein, the data preprocessing unit is used to perform denoising on the operation data set to obtain high-quality data;

[0018] The data quality evaluation unit is used to perform quality evaluation on the high-quality data to obtain an evaluation score. When the evaluation score is lower than a preset threshold, the high-quality data is re-preprocessed until the evaluation score meets the preset threshold to obtain a preprocessed data set.

[0019] Optionally, the operation mode recognition module includes: a model building unit, a model deployment unit and a model monitoring and maintenance unit;

[0020] The model building unit is used to establish a recognition model of operating habits and behavior patterns and to train and optimize the recognition model;

[0021] The model deployment unit is used to deploy the recognition model to the production environment to realize automatic recognition and classification of operation data;

[0022] The model monitoring and maintenance unit is used to continuously monitor the trained and optimized recognition model.

[0023] Optionally, the operation evaluation rule base module includes: a rule base unit, a rule matching and comparison unit, and an operation determination unit;

[0024] The rule base unit is used to construct an evaluation rule base for the standardization and effectiveness indicators of electrical equipment operation by combining standard procedures and expert knowledge of electrical equipment operation;

[0025] The rule matching and comparison unit is used to match and compare the preprocessing data set with the electrical equipment operation normativeness and effectiveness indicators in the evaluation rule base to obtain a reference space;

[0026] The operation determination unit is used to identify the operation data to be identified by using the identification model, and map the identification result to the reference space to obtain the operation determination result.

[0027] Optionally, the virtual assessment scenario library module includes: an assessment content acquisition and task decomposition unit, a scenario modeling unit, a scenario matching and task combination unit, an operation process tracking and recording unit and a comprehensive scoring unit;

[0028] The assessment content acquisition and task decomposition unit is used to acquire the assessment content and assessment requirements of electrical equipment operation, and decompose the assessment content into several typical operation tasks;

[0029] The scene modeling unit is used to construct virtual scenes for the typical operation tasks to obtain a plurality of virtual scenes;

[0030] The scene matching and task combination unit is used to combine the virtual scene with the typical operation task to form different assessment processes according to the assessment requirements;

[0031] The operation process tracking and recording unit is used to track and record the process of the operator executing different assessment processes;

[0032] The comprehensive scoring unit is used to comprehensively score the operator's operating skills using a comprehensive scoring algorithm.

[0033] Optionally, the risk assessment and fault diagnosis module includes: a risk assessment unit, a fault diagnosis unit and a safety prompt unit;

[0034] The risk assessment unit is used to analyze the real-time operation data using a support vector machine model to generate a risk index that represents the current operation risk level;

[0035] The fault diagnosis unit is used to match the real-time operation data with the fault mode in the rule base based on the risk index to obtain a fault diagnosis result;

[0036] The safety prompt unit is used to generate a safety warning prompt for the operator based on the fault diagnosis result and give corresponding suggested measures.

[0037] Compared with the prior art, the present invention has the following advantages and technical effects:

[0038] The present invention discloses an electrical equipment operation training and assessment system, which constructs a virtual operation environment, collects and preprocesses operation data, and establishes an operation habit recognition model. In combination with standard procedures and expert knowledge, an evaluation rule library is constructed to automatically judge the standardization and effectiveness of operations. Through task decomposition and scenario modeling, a comprehensive virtual assessment scenario library is created. At the same time, operational risks are monitored in real time, and high-risk operations and potential faults are identified. Finally, a multi-criteria decision-making method is used to conduct a comprehensive assessment of the operator's skills. The present invention realizes the intelligence and comprehensiveness of electrical equipment operation training and assessment, and improves the accuracy and efficiency of the assessment. Through the virtual environment and data analysis, the system can objectively evaluate the operator's skill level, identify problems and risks in the operation, provide strong support for training and assessment, and ultimately improve the safety and reliability of electrical equipment operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] The drawings constituting a part of the present application are used to provide a further understanding of the present application. The illustrative embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0040] Figure 1 This is a structural diagram of an electrical equipment operation training and assessment system based on virtual technology according to an embodiment of the present invention;

[0041] Figure 2 is a flow chart of an embodiment of the present invention. DETAILED DESCRIPTION

[0042] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0043] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0044] Embodiment 1

[0045] like Figure 1-Figure 2 As shown, this embodiment provides an electrical equipment operation training and assessment system based on virtual technology, including:

[0046] The virtual operating environment construction module is used to collect the electrical equipment operation data to form an operation data set. According to the requirements of the electrical equipment operation training assessment, a virtual operating environment is constructed to collect the operator's operation data in the virtual environment, including operation steps, operation time, operation results, etc., to form an operation data set.

[0047] The virtual operating environment construction module includes: an operating environment construction unit, an operating data acquisition unit and an original data self-checking unit; wherein the operating environment construction unit is used to construct a three-dimensional model and an operating interface containing electrical equipment; the operating data acquisition unit is used to collect the operating data of the operator in the virtual operating environment; the operating data includes operating steps, operating time and operating results; the original data self-checking unit is used to remove abnormal data from the operating data to obtain data in a unified format.

[0048] Furthermore, the electrical equipment operation training and assessment requirements are obtained, and according to the electrical equipment operation training and assessment requirements, a virtual operation environment is constructed, wherein the virtual operation environment includes a three-dimensional model and an operation interface of the electrical equipment; the operation data of the operator in the virtual operation environment is collected, wherein the operation data includes operation steps, operation time and operation results, to form an original operation data set; abnormal data is removed from the original operation data set, and the data format is unified to obtain a standardized operation data set.

[0049] Furthermore, as a specific implementation method of this embodiment, when obtaining the requirements for the assessment of electrical equipment operation training, the standards such as the "Electric Power Safety Work Regulations" and the "Electric Power Equipment Preventive Test Regulations" formulated by the State Grid Corporation can be referred to, and the key points of the assessment of electrical equipment operation can be extracted, such as the opening and closing operation of high-voltage switches, the oil level measurement of transformers, etc. According to these requirements, a modeling tool such as Unity 3D is used to build a realistic virtual electrical equipment operation environment, including three-dimensional models of equipment such as high-voltage switches and transformers, and corresponding operation interfaces, such as the closing button of the switch, etc. When the operator performs virtual operation, the system will automatically collect operation data, including the completion of the operation steps, the operation time of each step, and whether the final operation result is correct, etc., to form an original operation data set. The collected data is preprocessed, and abnormal data is filtered by setting a threshold, such as data with an operation time of more than 120 seconds, and all data are converted into a unified JSON format to obtain a standardized operation data set.

[0050] Furthermore, according to the electrical equipment operation specifications of the State Grid Corporation, an operation evaluation rule base is established, which includes the standard operation process of each operation step, the upper and lower limits of the normal operation time, and the judgment conditions of the operation results. For example, when the switch indicator light is on during the closing operation, it is judged that the operation is correct. The standardized operation data set is input into the evaluation model, and each operation step is evaluated using a rule-based reasoning algorithm. Taking the high-voltage switch closing operation as an example, if the operator's operation steps are consistent with the standard process, the operation time is between 20 and 60 seconds, and the switch status indicator light is on, then the step is judged to be correct and the score is 1 point, otherwise it is 0 points. The scores of all steps are weighted averaged, and the weights can be set according to the importance of each step, so as to obtain the operator's overall operation ability score, with a full score of 100 points. The operator's score is compared with the preset training goal, and the target achievement algorithm is used to calculate the operator's training effect. For example, if the training goal is to score more than 90 points, and the operator's score is 85 points, the achievement is 85 / 90=94%. Based on the degree of achievement, subsequent training improvement suggestions are output. For example, for steps where the operation time does not meet the requirements, the operator can be prompted to practice more and improve operational proficiency.

[0051] Data preprocessing module is used to preprocess the operation data set to obtain the preprocessed data set. For the operation data set, data preprocessing technology is used to perform denoising and other processing on the data to eliminate the noise in the data, improve the data quality, and provide high-quality data support for the subsequent operation standardization and effectiveness judgment.

[0052] The data preprocessing module includes: a data preprocessing unit and a data quality evaluation unit; wherein the data preprocessing unit is used to perform denoising on the operation data set to obtain high-quality data; the data quality evaluation unit is used to perform quality evaluation on the high-quality data to obtain an evaluation score. When the evaluation score is lower than a preset threshold, the high-quality data is re-preprocessed until the evaluation score meets the preset threshold to obtain a preprocessed data set.

[0053] Furthermore, the original data in the data set is obtained, and an exploratory analysis is performed on the original data to obtain the basic characteristics of the data, including the amount of data, data type, and data distribution. According to the results of the exploratory analysis, the specific plan for data preprocessing is determined. The quality of the normalized data is evaluated, and at least one data quality evaluation indicator is used. The data quality evaluation indicators include completeness, consistency, and accuracy. The quality score of the normalized data is calculated. If the quality score is lower than the preset threshold, the original data is re-preprocessed until the quality score meets the preset requirements. The trained judgment model is applied to the target business scenario, and the new operation data is intervened and optimized to provide data support for the efficient development of the target business.

[0054] Furthermore, as a specific implementation of this embodiment, after obtaining the original operation data set, the data is first subjected to exploratory analysis. Through statistical analysis, it is found that the original data set contains 10,000 operation records, covering the operation behaviors of 50 operators in a virtual environment. The data types mainly include three types: operation steps (strings), operation time (floating point numbers) and operation results (Boolean values). The operation time is distributed and analyzed, and histograms and box plots are drawn. It is found that the operation time presents a right-skewed distribution, with a median of 36 seconds and a small number of outliers. Based on the above analysis results, a data preprocessing scheme is determined, which mainly includes steps such as removing outliers, string encoding, and time normalization. The normalized data is quality assessed, using three indicators: integrity, consistency, and accuracy. The integrity assessment found that there are no missing values ​​in the normalized data set. The consistency assessment randomly divides the data set by cross-validation, trains the model on different subsets, and the average accuracy rate reaches 92%, indicating that the consistency of the data set is good. The accuracy assessment extracts 100 data by manual sampling inspection, and the domain experts make manual judgments, with an accuracy rate of 98%. The quality score calculated by combining the three indicators is 94, which is higher than the preset threshold of 95. Therefore, it is determined that the quality of the normalized data set meets the requirements and can enter the subsequent modeling stage. The trained operation ability judgment model is applied to the actual training scenario of the substation, and the operation data of the new trainees are analyzed. Through model prediction, it is found that some trainees spend too much time on key steps such as closing switches and fault handling, and there are irregular operations. In response to these problems, the system automatically generates improvement suggestions, such as strengthening the practice of closing switch operations, familiarizing with the operation process, etc., to help trainees improve their operation skills in a targeted manner. After a week of training, the average operation time of trainees was shortened by 20%, and the operation standardization was improved by 15%, which proves the practical application value of the system in substation training and provides strong support for improving training results.

[0055] The operation mode recognition module is used to establish a recognition model for operation habits and behavior patterns. Based on the preprocessed operation data set, pattern recognition algorithms such as support vector machines and decision trees are used to establish a recognition model for operation habits and behavior patterns. Through model training and optimization, the automatic recognition and classification of the operation habits and behavior patterns of different operators can be achieved.

[0056] The operation mode recognition module includes: a model building unit, a model deployment unit and a model monitoring and maintenance unit; among them, the model building unit is used to establish an identification model of operation habits and behavior patterns and to train and optimize the identification model; the model deployment unit is used to deploy the identification model to the production environment to realize automatic identification and classification of operation data; the model monitoring and maintenance unit is used to continuously monitor the trained and optimized identification model.

[0057] Furthermore, according to the characteristics of the preprocessed operation data set, a support vector machine algorithm is used to establish an initial operation habit and behavior pattern recognition model; the initial operation habit and behavior pattern recognition model is trained by a cross-validation method to obtain a trained operation habit and behavior pattern recognition model; in the process of training the initial operation habit and behavior pattern recognition model, a decision tree algorithm is introduced as a supplement, and the key features and rules in the operation habits and behavior patterns are captured through the decision tree algorithm to obtain an optimized operation habit and behavior pattern recognition model; the optimized operation habit and behavior pattern recognition model is deployed in a production environment to automatically identify and classify the real-time collected operation data; the performance of the optimized operation habit and behavior pattern recognition model in actual applications is continuously monitored, and the optimized operation habit and behavior pattern recognition model is regularly retrained and optimized.

[0058] Furthermore, as a specific implementation of this embodiment, according to the characteristics of the preprocessed data set, the support vector machine algorithm is adopted, the RBF kernel function is selected, the penalty coefficient C is set to 0, and the initial operation habit and behavior pattern recognition model is established. The initial model is trained by the 5-fold cross-validation method to obtain the trained model, and the average accuracy rate reaches more than 85%. In the training process, the decision tree algorithm is introduced, the CART algorithm is adopted, and the maximum depth of the tree is set to 5. The key features and rules of the operation habits and behavior patterns, such as operation frequency, operation duration, etc., are captured by the decision tree to optimize the support vector machine model to improve the interpretability and generalization ability of the model. The optimized model is deployed to the production environment, and automatic identification and classification are performed every 5 minutes for the real-time collected operation data, and the identification results are stored in the database, and the user portrait is updated in real time. The performance of the model in actual applications is continuously monitored, and the model is evaluated using evaluation indicators such as accuracy, recall rate and F1 value. When the model performance drops to a certain threshold, such as when the F1 value is lower than 8, the model retraining and optimization process is triggered, and the model is updated regularly to adapt to the changing business needs.

[0059] In the process of training the initial operation habit and behavior pattern recognition model, the decision tree algorithm is introduced as a supplement. The decision tree algorithm is used to capture the key features and rules in the operation habits and behavior patterns to obtain the optimized operation habit and behavior pattern recognition model.

[0060] Obtain the user's historical operation data and behavior data, perform data cleaning and feature extraction according to preset rules, and obtain a training sample set; use the support vector machine algorithm to train the training sample set to obtain an initial operation habit and behavior pattern recognition model; for the training results of the initial model, evaluate the accuracy and recall of the initial model through cross-validation. If the accuracy and recall of the initial model do not reach the preset threshold, the model optimization process is triggered; in the model optimization process, introduce the decision tree algorithm as a supplement, and construct the judgment nodes of key features and rules through the decision tree algorithm; fuse the judgment results of the decision tree algorithm with the output of the initial model, and dynamically adjust the weight coefficients of each feature in the initial model according to the fusion results to obtain the optimized model; use the optimized model to predict and identify new user operation and behavior data to determine whether they conform to the existing operation habits and behavior patterns; based on the results of multiple consecutive identifications, dynamically update the feature library of operation habits and behavior patterns, and regularly use the updated feature library to retrain the optimized model to achieve continuous optimization and upgrading of the model.

[0061] Furthermore, as a specific implementation method of this embodiment, first, obtain user operation records of the past three months from the user behavior log database, including clicks, browsing, purchases and other behaviors, totaling 10 million. Then, according to the preset data cleaning rules, such as removing duplicate records, filtering abnormal data, etc., 8 million valid data are obtained. Next, extract the key features of user operations, such as operation time, frequency, duration, etc., extract 20 features for each data, and form a training sample set. Use the support vector machine SVM algorithm, select the RBF kernel function, set the penalty factor C = 10, and train to obtain the initial model. Use 5-fold cross validation to evaluate the model performance. The current model has an accuracy of 85% and a recall of 80%. It does not reach the preset threshold of 90%, triggering model optimization. In the optimization process, the CART decision tree algorithm is introduced, and the Gini index is used as the splitting criterion to construct a decision tree and extract key features and rules, such as "If the operation frequency is high between 7pm and 10pm and the duration exceeds 30 minutes, it is determined to be a heavy user, etc." The judgment result of the decision tree is weighted and fused with the output of the SVM model, with a weight ratio of 6:4. The weight coefficients of each feature in the SVM model are dynamically adjusted, such as increasing the weight of the operation frequency feature, to obtain the optimized model. The optimized model is used to predict the operation data of new users. For example, if user A frequently browses the product page between 8pm and 9pm for 45 minutes, it is identified as the operation habit of a heavy user. In the recognition results for a consecutive week, the proportion of heavy users is stable at about 20%, and the feature is added to the pattern feature library. The model is retrained using the updated feature library every month. After 3 iterations, the accuracy and recall of the model are both increased to more than 95%, achieving continuous optimization of the model.

[0062] The operation evaluation rule base module is used to construct evaluation rules and judge the standardization and effectiveness of operations based on the evaluation rules. In combination with the standard procedures and expert knowledge of electrical equipment operation, an evaluation rule base for the standardization and effectiveness of operations is constructed, and the operation steps, operation time, operation results and other indicators are matched with the evaluation rules to automatically judge the standardization and effectiveness of operations and obtain the operation evaluation results. The operation evaluation rule base module includes: a rule base unit, a rule matching and comparison unit and an operation judgment unit; wherein the rule base unit is used to construct an evaluation rule base for the standardization and effectiveness indicators of electrical equipment operation in combination with the standard procedures and expert knowledge of electrical equipment operation; the rule matching and comparison unit is used to match and compare the preprocessed data set with the electrical equipment operation standardization and effectiveness indicators in the evaluation rule base to obtain a reference space; the operation judgment unit is used to use the recognition model to identify the operation data to be identified, and map the recognition results to the reference space to obtain the operation judgment results.

[0063] Furthermore, based on the standard procedures and expert knowledge of electrical equipment operation, an evaluation rule base for the standardization and effectiveness of operation is constructed, and the rule base contains standard steps, standard operation time ranges, expected operation results and other standardization and effectiveness indicators for various electrical equipment operations. Various parameters in the actual electrical equipment operation process are obtained, including operation steps, operation start and end time, equipment status after operation, etc., to form operation indicator data. The operation indicator data is matched and compared with the standardization and effectiveness indicators in the evaluation rule base, and a pattern recognition algorithm such as a decision tree or a support vector machine is used to determine whether each indicator meets the requirements in the rule base. If the operation indicator data meets all the standardization and effectiveness requirements in the rule base, the operation is determined to be a standard and effective operation; if there are indicators that do not meet the requirements, it is determined to be a non-standard or invalid operation. For operations that are determined to be non-standard or invalid, further analyze the specific indicator items that do not meet the requirements, determine the problems and causes of the operation, and generate an operation problem report. Based on the judgment results of the standardization and effectiveness of the operation, combined with the specific problem report, a comprehensive evaluation result of the operation is formed, indicating the standardization and effectiveness of the operation. The operation evaluation results are recorded and archived, the employees' operation level is assessed, and targeted operation training and improvement are carried out to improve the standardization and effectiveness of electrical equipment operation and ensure the safe and efficient operation of equipment.

[0064] Furthermore, as a specific implementation of this embodiment, a rule base for evaluating the operation norms and effectiveness of electrical equipment is constructed based on the standard procedures and expert knowledge of electrical equipment operation. The rule base defines the standard steps for various electrical equipment operations. For example, for the operation of the transformer, it is required to be carried out in the order of "disconnecting the low-voltage side switch-disconnecting the high-voltage side switch-pulling out the trolley-closing the grounding switch". The standard operation time range of each step is also clearly set. For example, the time interval from disconnecting the low-voltage side switch to disconnecting the high-voltage side switch should be controlled within 30 seconds. In addition, the expected operation results are also specified, such as the low-voltage side voltage should drop below 5V after the transformer is shut down. The system collects various parameters in the operation process of electrical equipment in real time, such as recording the action time of the circuit breaker through the sensor, and collecting the voltage data of the equipment through the voltmeter, and finally summarizes it to form a complete operation index data set. The data set is input into the operation norms and effectiveness judgment model based on the decision tree algorithm, and the model judges each operation record one by one. Taking a transformer operation as an example, by analyzing the edit distance between the sequence of operation steps and the standard requirements, it was found that there was a problem of missing steps; by comparing the actual execution time with the time range required in the rule base, it was found that the operation time of "closing the grounding switch" exceeded the standard time limit; by checking the low-voltage side voltage after the shutdown, it was judged that the result of the operation did not meet expectations. Based on the above three judgment results, the transformer operation was judged to be non-standard and invalid. The system automatically generates an operation problem report, pointing out the specific violations and possible causes, such as missing steps due to negligence, time limit exceeded due to unskilled operation, etc. Finally, the comprehensive evaluation results of the operation were recorded in the employee's operation file, and targeted retraining for transformer operation was arranged for him. By continuously carrying out the above process, the standardization and effectiveness of the electrical equipment operation of the power company have been significantly improved, the equipment failure rate has dropped by 20%, and the operating efficiency has increased by 15%.

[0065] Virtual assessment scenario library module is used to decompose the assessment content and conduct skill assessment on trainees. In view of the complexity and diversity of electrical equipment operation assessment, task decomposition and scenario modeling technology are used to decompose the assessment content into multiple typical operation tasks and scenarios, and a virtual assessment scenario library that fully covers the key points of the assessment is constructed. Through scenario matching and task combination, a comprehensive assessment of the operator's skills is achieved.

[0066] The virtual assessment scenario library module includes: an assessment content acquisition and task decomposition unit, a scenario modeling unit, a scenario matching and task combination unit, an operation process tracking and recording unit and a comprehensive scoring unit; wherein, the assessment content acquisition and task decomposition unit is used to obtain the electrical equipment operation assessment content and assessment requirements, and decompose the assessment content into several typical operation tasks; the scenario modeling unit is used to construct virtual scenes for several typical operation tasks to obtain several virtual scenes; the scenario matching and task combination unit is used to combine virtual scenes with typical operation tasks to form different assessment processes according to the assessment requirements; the operation process tracking and recording unit is used to track and record the process of the operator executing different assessment processes; the comprehensive scoring unit is used to use a comprehensive scoring algorithm to comprehensively score the operator's operating skills.

[0067] Furthermore, according to the requirements of the electrical equipment operation assessment, the assessment content is obtained, and the task is decomposed according to the assessment content to obtain multiple typical operation tasks; for each typical operation task, the scenario modeling technology is used to construct a virtual assessment scenario corresponding to the task, and the virtual assessment scenario should cover the assessment points of the task; the constructed multiple virtual assessment scenarios are organized into a virtual assessment scenario library, and the scenes in the virtual assessment scenario library are marked and indexed to achieve rapid retrieval and matching of the scenes; according to the assessment requirements, typical scenarios related to the assessment content are matched from the virtual assessment scenario library, and a complete assessment process is formed through scene combination; the assessment process is loaded in the virtual environment, and according to the task combination in the process, a virtual scene and operation interface corresponding to the assessment task are dynamically generated; the operator operates according to the assessment process in the virtual environment, and the system tracks and records the operation process, and judges whether the operation is standardized by comparing it with the preset operation specifications; according to the tracking record of the operation process and the standardization judgment result, a comprehensive scoring algorithm is used to objectively and comprehensively evaluate the operator's operating skills to obtain the assessment result.

[0068] Furthermore, as a specific implementation method of this embodiment, according to the State Grid Corporation's electrical equipment operation assessment specifications, the system automatically obtains the assessment outline, and through natural language processing technology, the assessment content is parsed into 10 typical operation tasks such as substation switching operation and distribution line maintenance operation. For each task, a realistic scene model including substations, distribution lines, etc. is constructed using three-dimensional modeling software such as 3ds Max, covering assessment details such as switching operation steps and maintenance operation points, forming 100 virtual assessment scenes corresponding to the tasks one by one. A relational database is used to manage the scene library, and multi-dimensional indexes such as scene name, task, and assessment points are used to achieve second-level retrieval of the scene. According to the assessment plan, the system automatically matches 50 typical scenes that are highly consistent with the assessment content from the scene library, and uses a directed graph algorithm to form a standard assessment process with a combination of 30% of substation scenes, 40% of line scenes, and 30% of other scenes. After the virtual simulation platform loads the assessment process, it calls the Unity 3D engine for dynamic rendering to generate realistic three-dimensional virtual assessment scenes and human-computer interaction interfaces according to the task type, scene environment and other parameters in the process. Operators wear VR equipment to carry out operations in an immersive manner. The system tracks and records the operation process in real time through motion capture, voice recognition and other technologies, uses the operation specifications in the expert knowledge base to match the operation process, and uses the fuzzy comprehensive evaluation method to judge the standardization of the operation. Standardized operations are scored 1 point, and non-standardized operations are deducted 5 points. The system performs weighted scoring on the key operation nodes tracked and recorded, with the weight of the operation result being 60% and the weight of the operation standardization being 40%. The hierarchical analysis method is used to calculate the assessment results of the operation skills. 85 points or more are excellent, 70-85 points are good, 60-70 points are passing, and below 60 points are failing.

[0069] The risk assessment and fault diagnosis module is used to monitor operational risks and identify high-risk operations and potential faults. Based on the safety and reliability requirements of electrical equipment operation, risk assessment and fault diagnosis technology are used to monitor and judge the risks and hidden dangers in the operation process in real time. By comparing with the preset risk threshold and fault mode, high-risk operations and potential faults are identified, and corresponding safety tips and improvement suggestions are given.

[0070] The risk assessment and fault diagnosis module includes: a risk assessment unit, a fault diagnosis unit and a safety prompt unit; wherein the risk assessment unit is used to analyze the real-time operation data using a support vector machine model to generate a risk index that represents the current operation risk level; the fault diagnosis unit is used to match the real-time operation data with the fault mode in the rule base based on the risk index to obtain a fault diagnosis result; the safety prompt unit is used to generate a safety warning prompt for the operator based on the fault diagnosis result and give corresponding recommended measures.

[0071] Furthermore, the real-time operation data of the electrical equipment is obtained, and the real-time operation data includes the equipment status and operation parameters; the acquired real-time operation data is analyzed using a pre-established risk assessment model to obtain a risk index that characterizes the risk level of the current operation; it is determined whether the risk index exceeds a preset threshold, and if so, it is determined that the current operation is a high-risk operation and a safety prompt is triggered; a rule-based fault diagnosis method is used to match the real-time operation data with a preset fault mode, and if the match is successful, it is determined that there is a potential fault and corresponding fault description information is generated; based on the risk index of the risk assessment and the fault description information of the fault diagnosis, safety prompts and improvement suggestions including high-risk operation identification results, potential fault identification results and corresponding response measures are automatically generated; the historical operation data and diagnosis results of the electrical equipment are obtained, and machine learning algorithms are used for training to obtain optimized risk assessment models and fault diagnosis rules; the real-time generated safety prompts and improvement suggestions are pushed to relevant operators through a human-machine interface or a mobile terminal to assist them in making safety decisions and optimizing operations; the operation data, risk events and fault conditions of the electrical equipment are continuously recorded and accumulated to form a complete equipment operation log to provide support for subsequent data analysis and model optimization.

[0072] Furthermore, as a specific implementation of this embodiment, the system acquires the running status data and operating parameters of the equipment in real time through the communication interface with the electrical equipment, such as voltage, current, power, temperature, etc., with a sampling frequency of 100Hz. For the acquired real-time data, the system adopts a risk assessment model based on a support vector machine (SVM) for analysis. The model can quantitatively evaluate the risk level of the current operation by training 1000 high-risk operation samples and 5000 normal operation samples in historical data, and generate a risk index between 0 and 100. When the risk index exceeds the preset threshold of 80, the system determines that the current operation is a high-risk operation, automatically triggers an audible and visual alarm, and pops up a safety prompt message on the human-machine interface. At the same time, the system adopts a fault diagnosis rule base based on expert knowledge, which contains 500 typical fault modes and their corresponding signs. By matching the real-time operation data with the fault mode in the rule base, if the matching degree exceeds 90%, it is determined that there is a potential fault, and the corresponding fault description information is generated, such as "abnormal vibration of the equipment, possible bearing wear", etc. The system integrates risk assessment results and fault diagnosis results, automatically generates safety tips and improvement suggestions, such as "The current operation risk is high, please reduce the load; abnormal equipment vibration is detected, it is recommended to check and maintain the bearings", etc., and pushes them to operators through the human-machine interface and mobile APP. In addition, the system continuously records equipment operation data and diagnosis results, automatically generates equipment operation logs every day, and uses incremental learning algorithms to optimize risk assessment models and fault diagnosis rules to improve the intelligence level and adaptability of the system.

[0073] The above are only preferred specific implementations of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. An electrical equipment operation training and assessment system based on virtual technology, characterized in that: include: Virtual operating environment construction module, data preprocessing module, operating mode recognition module, operating evaluation rule base module, virtual assessment scenario base module and risk assessment and fault diagnosis module; Wherein, the virtual operating environment construction module is used to collect electrical equipment operating data to form an operating data set; The data preprocessing module is used to preprocess the operation data set to obtain a preprocessed data set; The operation mode recognition module is used to establish a recognition model of operation habits and behavior patterns; The operation evaluation rule base module is used to construct evaluation rules and judge the standardization and effectiveness of operations based on the evaluation rules; The virtual assessment scenario library module is used to decompose the assessment content and conduct skill assessment on the operator; The risk assessment and fault diagnosis module is used to monitor operational risks and identify high-risk operations and potential faults.

2. The electrical equipment operation training and assessment system based on virtual technology according to claim 1 is characterized in that: The virtual operating environment construction module includes: an operating environment construction unit, an operating data acquisition unit and an original data self-checking unit; Wherein, the operating environment construction unit is used to construct a three-dimensional model including electrical equipment and an operating interface; The operation data collection unit is used to collect the operation data of the operator in the virtual operation environment; the operation data includes operation steps, operation time and operation results; The original data self-checking unit is used to remove abnormal data from the operation data to obtain data in a unified format.

3. The electrical equipment operation training and assessment system based on virtual technology according to claim 2 is characterized in that: The data preprocessing module includes: a data preprocessing unit and a data quality evaluation unit; Wherein, the data preprocessing unit is used to perform denoising on the operation data set to obtain high-quality data; The data quality evaluation unit is used to perform quality evaluation on the high-quality data to obtain an evaluation score. When the evaluation score is lower than a preset threshold, the high-quality data is re-preprocessed until the evaluation score meets the preset threshold to obtain a preprocessed data set.

4. The electrical equipment operation training and assessment system based on virtual technology according to claim 3 is characterized in that: The operation mode recognition module includes: a model building unit, a model deployment unit and a model monitoring and maintenance unit; The model building unit is used to establish a recognition model of operating habits and behavior patterns and to train and optimize the recognition model; The model deployment unit is used to deploy the recognition model to the production environment to realize automatic recognition and classification of operation data; The model monitoring and maintenance unit is used to continuously monitor the trained and optimized recognition model.

5. The electrical equipment operation training and assessment system based on virtual technology according to claim 4 is characterized in that: The operation evaluation rule base module includes: a rule base unit, a rule matching and comparison unit, and an operation determination unit; The rule base unit is used to construct an evaluation rule base for the standardization and effectiveness indicators of electrical equipment operation by combining standard procedures and expert knowledge of electrical equipment operation; The rule matching and comparison unit is used to match and compare the preprocessing data set with the electrical equipment operation standardization and effectiveness indicators in the evaluation rule base to obtain a reference space; The operation determination unit is used to identify the operation data to be identified by using the identification model, and map the identification result to the reference space to obtain the operation determination result.

6. The electrical equipment operation training and assessment system based on virtual technology according to claim 5 is characterized in that: The virtual assessment scenario library module includes: an assessment content acquisition and task decomposition unit, a scenario modeling unit, a scenario matching and task combination unit, an operation process tracking and recording unit and a comprehensive scoring unit; The assessment content acquisition and task decomposition unit is used to acquire the assessment content and assessment requirements of electrical equipment operation, and decompose the assessment content into several typical operation tasks; The scene modeling unit is used to construct virtual scenes for the typical operation tasks to obtain a plurality of virtual scenes; The scene matching and task combination unit is used to combine the virtual scene with the typical operation task to form different assessment processes according to the assessment requirements; The operation process tracking and recording unit is used to track and record the process of the operator executing different assessment processes; The comprehensive scoring unit is used to comprehensively score the operator's operating skills using a comprehensive scoring algorithm.

7. The electrical equipment operation training and assessment system based on virtual technology according to claim 6 is characterized in that: The risk assessment and fault diagnosis module includes: a risk assessment unit, a fault diagnosis unit and a safety prompt unit; The risk assessment unit is used to analyze the real-time operation data using a support vector machine model to generate a risk index that represents the current operation risk level; The fault diagnosis unit is used to match the real-time operation data with the fault mode in the rule base based on the risk index to obtain a fault diagnosis result; The safety prompt unit is used to generate a safety warning prompt for the operator based on the fault diagnosis result and give corresponding suggested measures.

Images