Immersive digital learning method and system for wind power hydraulic learning platform

Abstract

The invention provides an immersive digital learning method and system for a wind power hydraulic learning platform, and relates to the technical field of wind turbine generator training, and the method comprises the steps: constructing a fault model of a hydraulic system of a wind turbine generator; obtaining theoretical test data, and generating a first detection result based on the theoretical test data; constructing a normal model of the hydraulic system of the wind turbine generator; acquiring operation data of the trainee on the wind power hydraulic experiment table through a sensor, and performing practical operation test to generate a second detection result; performing hydraulic system learning evaluation according to the first detection result and the second detection result, and generating a learning effect evaluation report; and obtaining a learning effect evaluation report, performing report display through the wind power hydraulic system experiment table touch screen all-in-one machine, and performing teaching plan change and practical operation enhancement according to the evaluation report. According to the method and the device, the technical problems that in the prior art, a training method lacks pertinence and corresponding detection means, the training cost is high, and the risk is high can be solved.

Description

Technical Field

[0001] The present application relates to the technical field of wind turbine training, and in particular to an immersive digital learning method and system for a wind power hydraulic learning platform. Background Art

[0002] As an important component of wind turbines, the wind power hydraulic system's main function is to provide hydraulic driving force for the generator's variable pitch control device, safe pitch control device, yaw drive, brake device and parking brake device. At present, due to the continuous growth in demand for clean energy worldwide, the application scale of wind turbines continues to expand, which has led to a simultaneous increase in the use of wind power hydraulic systems. Wind power hydraulic systems are of great significance to the safe and stable operation of wind turbines. Therefore, relevant personnel in the wind power industry must fully understand the basic structure, working principle, normal data and causes of failures of the hydraulic system, so as to achieve real-time monitoring, repair and maintenance of the system to ensure the normal operation of the entire wind turbine. This urgently means that the industry needs to provide detailed training and learning for relevant operators.

[0003] Traditional training methods often rely on theoretical teaching and field operation. Theoretical teaching training mainly focuses on imparting knowledge points to the entire trainee body, which is not targeted and lacks corresponding testing methods, making it difficult to explore each trainee's mastery level and their own weaknesses. Field operation training requires the use of real wind turbine equipment, which undoubtedly increases the cost of training. In addition, during the field operation training process, due to the trainees' lack of proficiency in the operation of the equipment, it is easy to make mistakes, which may cause equipment failures, casualties and other safety accidents, further increasing the safety risks of field operation training.

[0004] In summary, traditional training methods lack pertinence and corresponding detection methods, and are costly and risky. Therefore, a method is needed to solve the above problems. Summary of the invention

[0005] The present invention provides an immersive digital learning method and system for a wind power hydraulic learning platform, which is used to solve the technical problems in the prior art that the training methods lack pertinence, lack corresponding detection means, have high training costs, and have high risks.

[0006] According to a first aspect of the present disclosure, an immersive digital learning method for a wind power hydraulic learning platform is provided, comprising:

[0007] According to the common faults and causes of the hydraulic system, a fault model of the hydraulic system of the wind turbine is constructed, and the fault model includes an oil quantity fault model, a temperature fault model, and a noise fault model; theoretical tests are carried out according to the fault model through the wind power hydraulic system test bench to obtain theoretical test data, and a first detection result is generated based on the theoretical test data; according to the technical requirements of the hydraulic system, a normal model of the hydraulic system of the wind turbine is constructed, and the normal model includes pressure index data, flow index data and temperature index data under the normal operation state of the hydraulic system; the trainees' operation data on the wind power hydraulic test bench are collected through sensors, and an interactive database is created, which is compared with the normal model, and a practical test is carried out to generate a second detection result; a hydraulic system learning evaluation is carried out according to the first and second test results, and a learning effect evaluation report is generated; the learning effect evaluation report is obtained, and the report is displayed through the wind power hydraulic system test bench touch screen all-in-one machine, and the teaching plan is changed and the practical operation is strengthened according to the evaluation report.

[0008] According to a second aspect of the present disclosure, an immersive digital learning system for a wind power hydraulic learning platform is provided, comprising:

[0009] A fault model construction module is used to construct a fault model of the hydraulic system of a wind turbine generator set according to common faults and causes of the hydraulic system, and the fault model includes an oil quantity fault model, a temperature fault model, and a noise fault model; a first detection result generation module is used to perform theoretical tests according to the fault model through a wind power hydraulic system test bench, obtain theoretical test data, and generate a first detection result based on the theoretical test data; a normal model construction module is used to construct a normal model of the hydraulic system of a wind turbine generator set according to the technical requirements of the hydraulic system, and the normal model includes pressure index data, flow index data, and temperature index data under the normal operation state of the hydraulic system; a second detection result generation module is used to collect trainees' operation data on the wind power hydraulic test bench through sensors, create an interactive database, compare it with the normal model, perform practical test, and generate a second detection result; a learning evaluation module is used to perform hydraulic system learning evaluation according to the first and second detection results, and generate a learning effect evaluation report; a report display and optimization module is used to obtain a learning effect evaluation report, display the report through the wind power hydraulic system test bench touch screen all-in-one machine, and change the teaching plan and strengthen the practical operation according to the evaluation report.

[0010] One or more technical solutions provided in this disclosure have at least the following technical effects or advantages:

[0011] According to the common faults and causes of the hydraulic system, a fault model of the hydraulic system of the wind turbine is constructed, and the fault model includes an oil quantity fault model, a temperature fault model, and a noise fault model; a theoretical test is performed according to the fault model through the wind power hydraulic system test bench to obtain theoretical test data, and a first test result is generated based on the theoretical test data; according to the technical requirements of the hydraulic system, a normal model of the hydraulic system of the wind turbine is constructed, and the normal model includes pressure index data, flow index data and temperature index data under the normal operation state of the hydraulic system; the trainees' operation data on the wind power hydraulic test bench are collected through sensors, an interactive database is created, and the data is compared with the normal model, and a practical test is performed to generate a second test result; the hydraulic system learning evaluation is performed according to the first test result and the second test result, and a learning effect evaluation report is generated; the learning effect evaluation report is obtained, and the report is displayed through the wind power hydraulic system test bench touch screen all-in-one machine, and the teaching plan is changed and the practical operation is strengthened according to the evaluation report. The technical problems that the training method lacks pertinence, lacks corresponding detection means, has high training costs, and has high risks in the prior art are solved, and the technical effects of increasing training personalization, adding detection means, and effectively reducing training costs and risks are achieved.

[0012] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] In order to more clearly illustrate the technical solutions in the present disclosure or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0014] Figure 1 A schematic diagram of a flow chart of an immersive digital learning method for a wind power hydraulic learning platform provided in an embodiment of the present application;

[0015] Figure 2 A structural schematic diagram of an immersive digital learning system for a wind power hydraulic learning platform provided in an embodiment of the present application.

[0016] Explanation of the accompanying drawings: fault model building module 11, first detection result generating module 12, normal model building module 13, second detection result generating module 14, learning evaluation module 15, report display and optimization module 16. DETAILED DESCRIPTION

[0017] The following is a description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0018] Embodiment 1

[0019] An immersive digital learning method for a wind power hydraulic learning platform provided by the embodiment of the present disclosure is referred to Figure 1 For illustration, the methods include:

[0020] S1: According to common faults and causes of hydraulic systems, a fault model of the hydraulic system of a wind turbine is constructed, wherein the fault model includes an oil quantity fault model, a temperature fault model, and a noise fault model;

[0021] Furthermore, a wind turbine hydraulic system fault model is constructed, including:

[0022] Obtain historical fault maintenance information of the hydraulic system of wind turbines, including system components, historical fault types, historical fault causes, and historical maintenance plans;

[0023] A graph database is used to construct a knowledge graph, where system components, historical fault types, historical fault causes, and historical maintenance plans are set as nodes, each node is given a unique identifier and the information of each node is described, and the logical relationship between the nodes is set as an edge. The nodes are connected into a knowledge graph through the edges, and the nodes and the edges are fixed elements in the knowledge graph;

[0024] The knowledge graph is classified according to the fault type to obtain the oil quantity fault model, temperature fault model, and noise fault model.

[0025] Specifically, the historical fault maintenance information of the hydraulic system of the wind turbine is obtained, including multiple historical data such as system components, historical fault types, historical fault causes, and historical maintenance plans. System components include power components, control components, actuators, auxiliary components, and hydraulic oil. Among them, the power components include hydraulic pumps, and common structural forms include gear pumps, vane pumps, and plunger pumps. The control components include pressure control valves, flow control valves, and direction control valves. The actuators include hydraulic motors, hydraulic cylinders, and hydraulic swing motors. Auxiliary components include oil tanks, oil filters, oil pipes and pipe joints, pressure gauges, oil level gauges, oil thermometers, etc. The historical fault types of the hydraulic system mainly involve three categories: oil leakage, heating of the hydraulic system, vibration, and noise. The historical fault causes and historical maintenance plans correspond to the fault types, mainly including component damage problems and maintenance operation steps corresponding to various fault types. For example, the oil leakage problem of the hydraulic system is mainly divided into internal leakage and external leakage. Internal leakage is a small amount of liquid leaking from the high-pressure chamber to the low-pressure chamber inside the hydraulic component. The corresponding maintenance plan is generally to reduce the wear of the components by debugging the original parts to alleviate the leakage. There are roughly two reasons for external leakage. One is that the pipe joint is loose or the sealing ring is damaged. This situation can be solved by tightening the joint or replacing the sealing ring. The other is that there is external leakage at the joint surface of the component, which is mainly caused by insufficient pre-tightening force of the fastening screws and wear of the sealing ring. At this time, the pre-tightening force should be increased or the sealing ring should be replaced.

[0026] Use Neo4j graph database technology to construct the acquired and integrated knowledge information into a knowledge graph. Set system components, historical fault types, historical fault causes, and historical maintenance plans as nodes in the knowledge graph, and assign each node a unique identifier to ensure the uniqueness and traceability of the data, and describe the information of each node, such as the specific description of the fault type, the specific analysis of the fault cause, and the specific operation of the maintenance plan. Set the logical relationship of each node as the edge of the knowledge graph, and link each node. The edge includes an association edge, a causal edge, and an execution edge. The association edge represents a correlation or attribute connection, such as the components involved in the fault type. The causal edge represents the causal relationship between nodes, such as the fault and the cause. The execution edge represents that a node performs a certain action and affects another node, such as the maintenance plan and the system component.

[0027] Based on the constructed overall knowledge graph and historical fault type classification, three types of fault models are extracted, including oil quantity fault model, temperature fault model, and noise fault model. Since the classification of hydraulic system faults is relatively clear, and the causes, maintenance plans, and components involved in each fault have a small cross-range, the overall knowledge graph can be divided into three types of knowledge graphs, which respectively reflect the knowledge involved in the three types of faults, laying the foundation for subsequent theoretical testing.

[0028] S2: performing a theoretical test according to the fault model through a wind power hydraulic system test bench to obtain theoretical test data, and generating a first detection result based on the theoretical test data;

[0029] Further, generating a first detection result includes:

[0030] Analyze the correctness of theoretical test data to obtain theoretical blind spot data classes, where the theoretical blind spot data classes are the knowledge weaknesses of users in various fault models;

[0031] Strengthen the theory according to the theoretical blind spot data category and generate a theoretical training plan;

[0032] Repeatedly train trainees according to the theoretical training plan to obtain a theoretical error probability table, wherein the theoretical error probability table is the probability of response errors occurring in different theories;

[0033] According to the theoretical error probability table, the theoretical training plan is traversed respectively to make error summary judgments, and the final error probability of the theoretical training is determined in combination with the fault model category;

[0034] The theoretical training final error probability is added to the first detection result.

[0035] Specifically, theoretical test questions are provided to trainees through the wind power hydraulic system test bench, and theoretical test data of the trainees are generated according to the results of the trainees' answers. The theoretical test questions correspond to three types of fault models, and are mainly used to detect whether the trainees are familiar with the causes of failures and maintenance methods of the hydraulic system. For example, how to repair the external leakage of hydraulic oil caused by the joint surface of the component, the two reasons for the heating of the hydraulic system, which components are prone to heat in the hydraulic system, the vibrations generated by the hydraulic system, the general categories of noise, the judgment of the flow pulsation noise of the hydraulic pump of the hydraulic system, cavitation noise, and casing vibration sound, etc.

[0036] Compare and analyze the collected theoretical test data with the knowledge graph covered by the fault model to obtain the theoretical knowledge in which the students answered incorrectly, generate theoretical blind spot data classes, and set up theoretical training plans based on the theoretical blind spot data classes. For example, if the students have weak knowledge on the diagnosis method of oil quantity failure, then the training plan needs to re-extract detailed information on the oil quantity failure as test questions for subsequent students to repeat practice.

[0037] According to the theoretical training plan, a theoretical error probability table is constructed to describe the probability of answering errors in different theories, including variables, knowledge points in the theoretical training plan and their associated parameters, and conditional relationships. According to historical data and the performance of trainees during the training process, the probability of correct and incorrect answers to knowledge points is calculated to obtain a theoretical error probability table; all theoretical knowledge points are traversed, possible error points and their probability of occurrence are analyzed, and the error probability of each knowledge point is evaluated using the theoretical error probability table. For example, if multiple knowledge blind spots in a theoretical task are related, the total failure probability is the combination of each independent probability. The final error probability of theoretical training is determined based on three types of failures, including the final error probability of the oil quantity failure model, the final error probability of the temperature failure model, and the final error probability of the noise failure model; the final error probability of theoretical training is used as a key indicator and added to the first test result, thereby improving the learning efficiency and training effect of trainees.

[0038] Furthermore, the final error probability of theoretical training is determined, including:

[0039] The theoretical error probability table is expressed as P(E i |C i ), where E i represents the event that the i-th theoretical knowledge is wrong, C i Indicates a given condition;

[0040] The theoretical training plan is represented as T = [S1, S2, ... S n ], where S k represents the kth theoretical knowledge;

[0041] Calculate the final error probability P(E any ), the formula is:

[0042]

[0043] Among them, E any Indicates the fault category;

[0044] Indicates the error event that occurred in the Kth theoretical knowledge;

[0045] Represents the kth theoretical knowledge S k Under the given condition C j The probability of an error occurring;

[0046] It indicates the probability that theoretical knowledge will not be wrong;

[0047] Represents the probability that all theoretical knowledge is correct.

[0048] S3: constructing a normal model of the hydraulic system of the wind turbine generator set according to the technical requirements of the hydraulic system, wherein the normal model includes pressure index data, flow index data and temperature index data under the normal operation state of the hydraulic system;

[0049] Furthermore, a normal model of the hydraulic system of the wind turbine is constructed, including:

[0050] According to the working principle and actual application requirements of the hydraulic system, the characteristics closely related to the normal operation state of the system are obtained as the input variables of the model, including pressure index data, flow index data and temperature index data;

[0051] The collected data are preprocessed to construct a normal model of the hydraulic system of the wind turbine generator set. The preprocessing includes data cleaning, outlier processing and data standardization.

[0052] Specifically, the health status data of the wind turbine hydraulic system during normal operation is collected. A system monitoring equipment group is deployed, including pressure sensors, flow sensors, and temperature sensors, to obtain the hydraulic system pressure data and timestamps, flow data and timestamps, and temperature data and timestamps in real time, forming a system normal data sequence, and forming system operation normal data and timestamps.

[0053] The three types of data in the normal operation data of the system are aligned on the time axis, and the data are preprocessed, including data cleaning, outlier processing and data standardization. The normal operation data sequence of the system is obtained, including pressure sequence, flow sequence and temperature sequence. The purpose of data cleaning is to remove erroneous data and noise data in the data. In the process of obtaining pressure, flow and temperature index data, erroneous data may be generated due to sensor failure, signal interference and other reasons. For example, the sensor suddenly appears abnormally high or low values. These data do not conform to the normal operation logic of the hydraulic system and need to be identified and eliminated. You can judge whether the data is abnormal by setting a reasonable threshold range, or you can use data smoothing technology to remove noise data. In addition to cleaning obviously erroneous data, you also need to process outliers. Outliers may be caused by short-term fluctuations in the system or special working conditions. For outliers, a variety of processing methods can be used. If it is caused by special working conditions, and the working condition is representative in the subsequent normal model construction, the outliers can be corrected; if the outliers are caused by unreliable factors such as measurement errors, they can be replaced with reasonable values ​​by interpolation or mean replacement. Since the dimensions of the three indicators, pressure, flow and temperature, are different, in order to facilitate subsequent model construction and analysis, the data needs to be standardized. The commonly used data standardization method is Z-score standardization, the formula is: Where x is the original data, μ is the mean of the data, and Δ is the standard deviation of the data. After standardization, the data of different indicators will be in a similar numerical range, which improves the accuracy and stability of model construction.

[0054] The normal model of the hydraulic system of the wind turbine is constructed according to the normal data sequence of the system operation. The normal model can objectively describe the fluctuation range of each indicator data when the hydraulic system of the wind turbine is in a healthy operating state. Therefore, abnormal fluctuations in pressure, flow and temperature are not only direct signals of the existence of faults, but also important bases for fault prediction, diagnosis and maintenance strategy formulation.

[0055] S4: Collect trainees’ operation data on the wind power hydraulic test bench through sensors, create an interactive database, compare it with the normal model, conduct practical tests, and generate the second test results;

[0056] Further, generating a second detection result includes:

[0057] Acquire the trainee's operation data on the wind power hydraulic test bench, wherein the operation data includes the trainee's adjustment of the opening of the hydraulic valve and the start or stop operation of the hydraulic pump;

[0058] Analyze and process the interactive database, compare the trainees' operation data with the normal model of the wind turbine hydraulic system, mark the wrong operation data, and obtain the operation error data class;

[0059] Carry out practical training according to the operational error data and generate a practical training plan;

[0060] Repeatedly train trainees according to the practical training plan to obtain a practical error probability table; the practical error probability table is used to describe the probabilities of different operation errors;

[0061] According to the practical error probability table, the practical training plan is traversed to perform error summary diagnosis, and the final error probability of the practical training is determined in combination with the normal model. The algorithm for obtaining the final error probability of the practical training is consistent with the final error probability of the theoretical training;

[0062] The final error probability of the practical training is added to the second detection result.

[0063] Specifically, interactive tests are carried out through the wind power hydraulic system test bench to obtain the trainees' operation data. The wind power hydraulic system test bench can restore the real operation scene by simulating the actual hydraulic system structure of the wind turbine set, including hydraulic station, yaw solenoid valve, yaw caliper, oil tank and other system components, providing students with immersive learning and operation practice.

[0064] Adjust the pressure, flow and temperature values ​​of the hydraulic system according to the trainee's operation data and add them to the interactive database. Call the normal model of the wind turbine hydraulic system to judge whether the data in the interactive database is within the required range of the model, extract the interactive data outside the range, trace back the operation behavior, mark the wrong operation data, and obtain the operation error data class. The operation error data class represents the typical problems encountered by users in actual operation, such as parameter adjustment errors.

[0065] Design practical training for the problems involved in the operational error data category, such as simulating alarm processing under specific working conditions, generating a practical training plan, including special training tasks for errors, and adding dynamic simulation scenarios.

[0066] According to the practical training plan, a practical error probability table is constructed to describe the probability of errors in different operations, including variables, operation points in the practical training plan and their associated parameters, and conditional relationships. According to historical data and the performance of trainees during the training process, the probability of correct and incorrect operation point adjustment is calculated to obtain a practical error probability table; all practical operation points are traversed, possible error points and their probability of occurrence are analyzed, and the probability of error occurrence of each operation point is evaluated using the practical error probability table. For example, if multiple error points of an operation task are associated, the total failure probability is the combination of each independent probability. The final error probability of practical training is determined according to three normal types, including the final error probability of pressure index data adjustment, the final error probability of flow index data adjustment, and the final error probability of temperature index data adjustment. The formula for obtaining the final error probability of practical training is the same as the final error probability of theoretical training; the final error probability of practical training is used as a key indicator and added to the second test result, thereby improving the learning efficiency and training effect of trainees.

[0067] S5: Performing a hydraulic system learning evaluation according to the first detection result and the second detection result, and generating a learning effect evaluation report;

[0068] Furthermore, a hydraulic system learning evaluation is performed according to the first detection result and the second detection result to generate a learning effect evaluation report, including:

[0069] Based on the first detection result, the mastery of theoretical knowledge is evaluated, and a first evaluation curve set is constructed, wherein the first evaluation curve set includes an oil quantity fault model evaluation curve, a temperature fault model evaluation curve, and a noise fault model evaluation curve;

[0070] Based on the second detection result, the mastery of practical knowledge is evaluated, and a second evaluation curve set is constructed, where the second evaluation curve set includes a pressure index evaluation curve, a flow index evaluation curve, and a temperature index evaluation curve;

[0071] Performing a theoretical assessment based on the first assessment curve set to generate a theoretical mastery score;

[0072] Performing a practical operation assessment based on the second assessment curve set to generate a practical operation proficiency score;

[0073] The multiple theoretical mastery scores and the multiple practical proficiency scores are comprehensively analyzed to generate the learning effect evaluation report.

[0074] Specifically, the theoretical learning of trainees is evaluated based on the first test results, and the learning effect of trainees in multiple theoretical trainings is analyzed, especially the degree of mastery of the three types of faults in the hydraulic system of wind turbines. The degree of mastery of the theoretical knowledge of trainees in each type of fault is quantitatively analyzed, and the first evaluation curve set is generated using the function fitting method, including the oil quantity fault model evaluation curve, the pressure fault model evaluation curve, and the temperature fault model evaluation curve. The horizontal axis of the curve coordinate axis is the learning time or learning stage, such as the initial, middle, and final stages, and the vertical axis is the knowledge mastery, such as the theoretical score at this stage.

[0075] Specifically, the oil quantity fault model evaluation curve is a quantitative analysis of the students' mastery of theoretical knowledge related to oil quantity faults. By obtaining the final probability of theoretical training in the first test result, the students' understanding of theoretical knowledge points such as the causes, detection methods, and solution strategies of oil quantity faults in the theoretical test is scored, and the theoretical score of this stage is obtained. The oil quantity fault model evaluation curve is constructed by connecting the score points of each stage. If the mastery score of the theoretical knowledge of oil quantity faults is low at the beginning of learning, it will gradually improve with the deepening of learning, and the curve will show an upward trend. Similarly, for the temperature fault model, the students' mastery of the theoretical knowledge of temperature faults is analyzed. For example, the depth of understanding of theoretical knowledge such as the harm of excessive temperature in the hydraulic system and the relationship with other system parameters. The fluctuation of the curve may reflect the difficulty or breakthrough of the learner in understanding the cause of temperature faults. For the noise fault model evaluation curve, the main consideration is the students' theoretical mastery of the theoretical root causes of noise generation and the relationship between noise and the operating status of the hydraulic system. If the learner has a new understanding of a certain theoretical point of noise fault in a certain period of time, the curve will change accordingly.

[0076] According to the second test results, the trainees' practical learning evaluation is carried out, and the learning effect of the trainees in multiple practical trainings is analyzed, especially the ability to adjust the three normal indicators of the hydraulic system of the wind turbine. The improvement of the trainees' operating skills is quantitatively analyzed, and the second evaluation curve set is generated using the function fitting method, including the pressure indicator evaluation curve, the flow indicator evaluation curve and the temperature indicator evaluation curve. The horizontal axis of the curve coordinate axis is the learning time or operation stage, such as the initial, middle and final stages, and the vertical axis is the proficiency of the operating skills, such as the practical score at this stage.

[0077] Specifically, the pressure index evaluation curve mainly reflects the trainees' ability to control and adjust the pressure index during the practical operation. By obtaining the final probability of the practical training in the second test result, the trainees' performance in the practical test is scored to obtain the practical score of this stage. The score can reflect the trainees' ability to adjust the pressure index when the hydraulic system is started, the operating load changes, and the system stops. By calculating the practical scores of each stage, the pressure index evaluation curve is constructed by connecting each score point. For the flow index evaluation curve, the main focus is on the trainees' ability to adjust the flow in practical operation. For example, the ability to reasonably control the flow according to different work requirements. The curve reflects the learner's learning process in flow control. From the initial unskilled to the gradual proficiency, the curve may change from large fluctuations to smoother. When constructing the temperature index evaluation curve, the trainees' ability to control the temperature of the hydraulic system in practical operation is considered. For example, during long-term operation, can the temperature be controlled within the normal range through reasonable operations, such as the use of heat dissipation devices? If the learner has poor temperature regulation ability at the beginning of practical operation, the curve will show an upward change as experience accumulates and skills improve.

[0078] Through comprehensive analysis of the first evaluation curve set and the second evaluation curve set, the weighted average method is used to generate the theoretical mastery score and the practical mastery score. Assuming that the weight of the oil quantity fault model evaluation curve is 0.3, the weight of the temperature fault model evaluation curve is 0.3, and the weight of the noise fault model evaluation curve is 0.4, the final score corresponding to each curve is multiplied by the weight and added together to obtain the theoretical mastery score. For example, the final score of the oil quantity fault model evaluation is 80 points, the final score of the temperature fault model evaluation is 75 points, and the final score of the noise fault model evaluation is 85 points. Then the theoretical mastery score = 80 × 0.3 + 75 × 0.3 + 85 × 0.4 = 81.5 points. Similarly, the practical mastery score is obtained.

[0079] Comprehensively analyze the theoretical mastery score and the practical mastery score to obtain the overall learning effect evaluation. For example, the weighted average score of the two can be calculated as the final learning effect score. Assuming that the weight of the theoretical mastery score is 0.5 and the weight of the practical mastery score is 0.5, the final learning effect score calculation formula is (81.5×0.5+80.2×0.5)=80.85 points.

[0080] According to the final score, the overall effect of the learner's learning of hydraulic systems can be evaluated. If the score is 90 points or above, it means that the learner has a good grasp of both theory and practice; if the score is between 70-89 points, it means that the learner has basically mastered the relevant knowledge and skills, but there is still room for improvement; if the score is below 70 points, the learner needs further training and guidance, and there may be major deficiencies in theory or practice.

[0081] Generate a learning effect evaluation report based on the student's final score, theoretical mastery score, practical mastery score, the first evaluation curve set and the second evaluation curve set. Generate a theoretical ability evaluation, practical ability evaluation and comprehensive ability evaluation of the student based on the theoretical mastery score, practical mastery score and final score. At the same time, perform a difference analysis and trend analysis between the first curve set and the second curve set to generate the student's learning effect at each stage, and then serialize the multi-stage learning effect according to the learning sequence, that is, organize the learning effect of each stage into serialized data according to the time sequence, and generate learning effect feedback parameters, including theoretical improvement range parameters, practical improvement range parameters and comprehensive improvement parameters, to learn the overall learning progress of the student.

[0082] S6: Obtain a learning effect evaluation report, display the report through the wind power hydraulic system experimental bench touch screen all-in-one machine, and make changes to the teaching plan and strengthen practical operations based on the evaluation report.

[0083] Specifically, by obtaining a personalized learning effect evaluation report for the trainee, we can understand the trainee's learning progress, current ability level and weak links, and formulate a personalized review plan for the trainee based on the evaluation report. The weaker theoretical knowledge is marked as the review focus. For the theoretical knowledge that is difficult to understand, more examples or animations can be added to the teaching plan. For example, when explaining the working principle of the hydraulic pump, an animation can be added to show the flow process of the hydraulic oil in the pump. For operational problems, the order or detail of the operating instructions can be adjusted. If the trainee does not operate properly in the pressure adjustment link, the operating instructions for this step can be refined in the teaching plan, emphasizing the key adjustment points and precautions. In terms of teaching methods, if it is found that the trainees do not have a deep understanding of the theoretical knowledge as a whole, it may be necessary to change from the traditional lecture-based teaching method to a group discussion or project-based learning method. For example, organize trainees to discuss in groups the importance of hydraulic systems in wind power equipment and the possible types of faults and solutions. According to the operational problems found in the evaluation report, arrange targeted practical exercises for the trainees. If students have problems with the operation of hydraulic valves, special hydraulic valve operation practice tasks can be set up on the experimental table to allow students to repeatedly practice the opening, closing, flow regulation and other operations of different types of hydraulic valves. Practical exercises of different difficulty levels can be set, from simple basic operations to complex combined operations, to gradually improve students' practical ability.

[0084] Embodiment 2

[0085] Based on the same inventive concept as the immersive digital learning method for the wind power hydraulic learning platform in the aforementioned embodiment, the present application also provides an immersive digital learning system for the wind power hydraulic learning platform, see the attached Figure 2 , the system comprising:

[0086] A fault model building module 11 is used to build a fault model of the hydraulic system of a wind turbine generator set according to common faults and causes of the hydraulic system, wherein the fault model includes an oil quantity fault model, a temperature fault model, and a noise fault model;

[0087] A first detection result generating module 12, the first detection result generating module 12 is used to perform a theoretical test according to the fault model through a wind power hydraulic system test bench, obtain theoretical test data, and generate a first detection result based on the theoretical test data;

[0088] A normal model building module 13, which is used to build a normal model of the hydraulic system of the wind turbine generator set according to the technical requirements of the hydraulic system, wherein the normal model includes pressure index data, flow index data and temperature index data under the normal operation state of the hydraulic system;

[0089] The second detection result generating module 14 is used to collect the trainee's operation data on the wind power hydraulic test bench through sensors, create an interactive database, compare it with the normal model, perform practical test, and generate a second detection result;

[0090] A learning evaluation module 15, which is used to perform a learning evaluation of the hydraulic system according to the first detection result and the second detection result, and generate a learning effect evaluation report;

[0091] The report display and optimization module 16 is used to obtain a learning effect evaluation report, display the report through the wind power hydraulic system experimental table touch screen integrated machine, and change the teaching plan and strengthen the practical operation according to the evaluation report.

[0092] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0093] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. An immersive digital learning method for a wind power hydraulic learning platform, characterized in that: The method comprises: According to common faults and causes of hydraulic systems, a fault model of the hydraulic system of a wind turbine is constructed, wherein the fault model includes an oil quantity fault model, a temperature fault model, and a noise fault model; Performing theoretical testing according to the fault model through a wind power hydraulic system test bench to obtain theoretical test data, and generating a first test result based on the theoretical test data; According to the technical requirements of the hydraulic system, a normal model of the hydraulic system of the wind turbine is constructed, wherein the normal model includes pressure index data, flow index data and temperature index data under the normal operation state of the hydraulic system; The sensors are used to collect the trainees’ operation data on the wind power hydraulic test bench, create an interactive database, compare it with the normal model, conduct practical tests, and generate the second test results; Performing a hydraulic system learning evaluation based on the first detection result and the second detection result, and generating a learning effect evaluation report; Obtain a learning effect evaluation report, display the report through the wind power hydraulic system experimental bench touch screen all-in-one machine, and make changes to the teaching plan and strengthen practical operations based on the evaluation report.

2. The immersive digital learning method for a wind power hydraulic learning platform according to claim 1, characterized in that: Construct a wind turbine hydraulic system fault model, including: Obtain historical fault maintenance information of the hydraulic system of wind turbines, including system components, historical fault types, historical fault causes, and historical maintenance plans; A graph database is used to construct a knowledge graph, where system components, historical fault types, historical fault causes, and historical maintenance plans are set as nodes, each node is given a unique identifier and the information of each node is described, and the logical relationship between the nodes is set as an edge. The nodes are connected into a knowledge graph through the edges, and the nodes and the edges are fixed elements in the knowledge graph; The knowledge graph is classified according to the fault type to obtain the oil quantity fault model, temperature fault model, and noise fault model.

3. The immersive digital learning method for a wind power hydraulic learning platform according to claim 1, characterized in that: Generating a first detection result, including: Analyze the correctness of theoretical test data to obtain theoretical blind spot data classes, where the theoretical blind spot data classes are the knowledge weaknesses of users in various fault models; Strengthen the theory according to the theoretical blind spot data category and generate a theoretical training plan; Repeatedly train trainees according to the theoretical training plan to obtain a theoretical error probability table, wherein the theoretical error probability table is the probability of response errors occurring in different theories; According to the theoretical error probability table, the theoretical training plan is traversed respectively to make error summary judgments, and the final error probability of the theoretical training is determined in combination with the fault model category; The theoretical training final error probability is added to the first detection result.

4. The method according to claim 3, characterized in that Determine the final error probability of theoretical training, including: The theoretical error probability table is expressed as P(E i |C i ), where E i represents the event that the i-th theoretical knowledge is wrong, C i Indicates a given condition; The theoretical training plan is represented by T = [S1, S2, ... S n ], where S k represents the kth theoretical knowledge; Calculate the final error probability P(E any ), the formula is: Among them, E any Indicates the fault category; Indicates the error event that occurred in the Kth theoretical knowledge; Represents the kth theoretical knowledge S k Under the given condition C j The probability of an error occurring; It indicates the probability that theoretical knowledge will not be wrong; Represents the probability that all theoretical knowledge is correct.

5. The immersive digital learning method for a wind power hydraulic learning platform according to claim 1, characterized in that: Construct a normal model of the wind turbine hydraulic system, including: According to the working principle and actual application requirements of the hydraulic system, the characteristics closely related to the normal operation state of the system are obtained as the input variables of the model, including pressure index data, flow index data and temperature index data; The collected data are preprocessed to construct a normal model of the hydraulic system of the wind turbine generator set. The preprocessing includes data cleaning, outlier processing and data standardization.

6. The immersive digital learning method for a wind power hydraulic learning platform according to claim 1, characterized in that: Generating a second detection result, including: Acquire the trainee's operation data on the wind power hydraulic test bench, wherein the operation data includes the trainee's adjustment of the opening of the hydraulic valve and the start or stop operation of the hydraulic pump; Analyze and process the interactive database, compare the trainees' operation data with the normal model of the wind turbine hydraulic system, mark the wrong operation data, and obtain the operation error data class; Carry out practical training according to the operational error data and generate a practical training plan; Repeatedly train trainees according to the practical training plan to obtain a practical error probability table, where the practical error probability table is used to describe the probabilities of different operation errors; According to the practical error probability table, the practical training plan is traversed to perform error summary diagnosis, and the final error probability of the practical training is determined in combination with the normal model. The algorithm for obtaining the final error probability of the practical training is consistent with the final error probability of the theoretical training; The final error probability of the practical training is added to the second detection result.

7. The immersive digital learning method for a wind power hydraulic learning platform according to claim 1, characterized in that: Performing a hydraulic system learning evaluation based on the first detection result and the second detection result to generate a learning effect evaluation report includes: Based on the first detection result, the mastery of theoretical knowledge is evaluated, and a first evaluation curve set is constructed, wherein the first evaluation curve set includes an oil quantity fault model evaluation curve, a temperature fault model evaluation curve, and a noise fault model evaluation curve; Based on the second detection result, the mastery of practical knowledge is evaluated, and a second evaluation curve set is constructed, where the second evaluation curve set includes a pressure index evaluation curve, a flow index evaluation curve, and a temperature index evaluation curve; Performing a theoretical assessment based on the first assessment curve set to generate a theoretical mastery score; Performing a practical operation assessment based on the second assessment curve set to generate a practical operation proficiency score; The multiple theoretical mastery scores and the multiple practical proficiency scores are comprehensively analyzed to generate the learning effect evaluation report.

8. An immersive digital learning system for a wind power hydraulic learning platform, characterized in that: An immersive digital learning method for a wind power hydraulic learning platform for implementing any one of claims 1 to 7, the system comprising: A fault model building module is used to build a wind turbine hydraulic system fault model based on common faults and causes of the hydraulic system, wherein the fault model includes an oil quantity fault model, a temperature fault model, and a noise fault model; A first detection result generating module is used to perform a theoretical test according to the fault model through a wind power hydraulic system test bench to obtain theoretical test data, and generate a first detection result based on the theoretical test data; A normal model building module is used to build a normal model of the hydraulic system of the wind turbine according to the technical requirements of the hydraulic system, wherein the normal model includes pressure index data, flow index data and temperature index data under the normal operation state of the hydraulic system; The second test result generation module is used to collect the trainees' operation data on the wind power hydraulic test bench through sensors, create an interactive database, compare it with the normal model, conduct practical test, and generate the second test result; A learning evaluation module, used to perform a learning evaluation of the hydraulic system according to the first detection result and the second detection result, and generate a learning effect evaluation report; The report display and optimization module is used to obtain the learning effect evaluation report, display the report through the wind power hydraulic system experimental bench touch screen all-in-one machine, and make teaching plan changes and practical operation enhancements based on the evaluation report.

Images