Electric power practical operation skill intelligent auxiliary scoring method based on AI

Abstract

The invention discloses an AI-based electric power practical operation skill intelligent auxiliary scoring method, and belongs to the technical field of electric power practical operation skill assessment, and the method comprises the steps: building a practical operation standard library by a platform side, and building a practical operation assessment model according to the practical operation standard library; the platform side constructs a user transmission channel, and a user transmits an electric power practical operation video and practical operation project information through the user transmission channel; receiving an electric power practical operation video and practical operation item information transmitted by the user in real time, analyzing the electric power practical operation video and the practical operation item information through a practical operation evaluation model to obtain practical operation score data, and sending the practical operation score data to the corresponding user; the labor cost and the training cost caused by employing a coach are reduced; meanwhile, the limitation of time and space is broken through, students can carry out practical operation training anytime and anywhere and obtain score feedback, the dependence on manual coaches is reduced, and a more convenient and efficient way is provided for skill improvement of employees in the power industry.

Description

Technical Field

[0001] The present invention belongs to the technical field of electric power practical operation skill assessment, and specifically is an AI-based intelligent auxiliary scoring method for electric power practical operation skills. Background Art

[0002] In the power industry, practical skills training is a core component in ensuring the safe operation of the power grid and improving the professional quality of practitioners. However, traditional practical training models for power plants rely heavily on on-site guidance and experience-based judgment from human instructors. Manual scoring is subject to high subjectivity and poor consistency. Different instructors may have different criteria for judging "operational standardization," such as tolerance for tool placement angles and operating sequences. This makes it difficult to quantify trainees' skill levels using unified standards. Furthermore, daily training for practitioners is difficult to achieve with real-time guidance from instructors, leaving them to rely solely on individual training. This is especially true for many employed practitioners, who often struggle to find instructors for practical guidance.

[0003] Based on this, in order to achieve intelligent scoring of practical power operation skills, the present invention provides an AI-based intelligent auxiliary scoring method for practical power operation skills. Summary of the Invention

[0004] In order to solve the problems existing in the above-mentioned solutions, the present invention provides an AI-based intelligent auxiliary scoring method for practical power operation skills.

[0005] The purpose of the present invention can be achieved through the following technical solutions:

[0006] The AI-based intelligent assisted scoring method for power practical skills includes:

[0007] Step 1: The platform establishes a practice standard library, which is used to store the practice standards and scoring criteria corresponding to various practice projects; and establishes a practice evaluation model based on the practice standard library;

[0008] Furthermore, the practical operation standard library is updated in real time, and the practical operation evaluation model is updated and adjusted after the practical operation standard library is updated.

[0009] Step 2: The platform builds a user transmission channel, through which users transmit power operation videos and operation project information;

[0010] Step 3: Receive the power practice video and practice project information transmitted by the user in real time, analyze the power practice video and practice project information through the practice evaluation model, obtain practice score data, and send the practice score data to the corresponding user.

[0011] Further, performing video quality analysis on the received power practice video and practice project information to obtain a video quality assessment result of the power practice video, wherein the video quality assessment result includes a qualified video quality and an unqualified video quality;

[0012] When the video quality assessment result shows that the video quality is qualified, no corresponding operation is performed;

[0013] When the video quality evaluation result is that the video quality is unqualified, video capture defects are determined, video feedback data is generated according to the video capture defects, and the video feedback data is sent to the user.

[0014] Furthermore, the method for performing video quality analysis on the power operation video and the operation project information includes:

[0015] Identify the practical project information, match the practical video requirements of the power practical video according to the practical project information, adjust the power practical video and the practical video requirements according to the practical project information, and obtain the practical analysis video and the practical analysis requirements;

[0016] A quality assessment model is established. The expression of the quality assessment model is:

[0017]

[0018] Where (s, k) is the input data, s is the practical analysis video, and k is the practical analysis requirement. s→k indicates that the practical analysis video meets the practical analysis requirement. The output data is the quality assessment value ZP(s, k), which is 1 or 0.

[0019] Analyze the practical operation analysis video and the practical operation analysis requirements through a quality assessment model to obtain a quality assessment value of the power practical operation video;

[0020] When the quality evaluation value is 1, the video quality evaluation result is that the video quality is qualified;

[0021] When the quality evaluation value is 0, the video quality evaluation result is that the video quality is unqualified.

[0022] Furthermore, the video feedback data includes video capture shortcomings and the impact of the shortcomings on the scores.

[0023] Furthermore, the video feedback data includes video capture shortcomings, the impact of shortcomings on scores, and suggestions for overcoming shortcomings.

[0024] Furthermore, the proposed methods to overcome the shortcomings include:

[0025] Generating a plurality of candidate methods according to the video acquisition shortcomings, wherein the candidate methods are solutions that have a probability of solving the video acquisition shortcomings;

[0026] Conduct simulation screening of the corresponding candidate methods based on the power operation video to obtain the screening results of the corresponding candidate methods, including unqualified and qualified screening results;

[0027] Prioritize the candidate methods that have passed the screening results to obtain a priority sequence, and generate suggestions for overcoming shortcomings based on the priority sequence.

[0028] Furthermore, simulated screening of candidate methods is conducted based on the power operation video, including:

[0029] Perform simulation supplement on the electric power operation video according to the selected method to obtain the electric power simulation video of the selected method;

[0030] Analyzing the power simulation video using a quality assessment model to obtain a quality assessment value of the selected method;

[0031] When the quality assessment value is 1, the screening result is qualified;

[0032] When the quality assessment value is 0, the screening result is unqualified.

[0033] Furthermore, when users are conducting practical electricity operations, they can transmit the practical electricity operation videos and practical operation project information of the users in real time through user transmission channels, conduct real-time analysis of the practical electricity operation videos and practical operation project information, obtain practical operation scoring data, conduct risk warning analysis based on the practical operation scoring data, obtain risk warning results, and send the practical operation scoring data and risk warning results to users.

[0034] An AI-based intelligent assisted scoring system for practical power operation skills, including both the platform and user sides;

[0035] The platform includes a practical standard library, a practical evaluation module, and a quality assessment module;

[0036] The practical standard library is used to store the practical standards and scoring standards corresponding to various practical projects;

[0037] The practical evaluation module is used to perform practical evaluation, receive the power practical operation video and practical operation project information transmitted by the user end in real time, analyze the power practical operation video and practical operation project information through the preset practical evaluation model, obtain practical operation score data; and send the practical operation score data to the corresponding user;

[0038] The quality assessment module is used to perform video quality analysis on the received power operation video and the practical operation project information, obtain the video quality assessment result of the power operation video, and perform corresponding feedback processing according to the video quality assessment result;

[0039] The user terminal includes a video module and a practical risk module;

[0040] The video module is used for users to send power practice videos and practice project information to the platform, and receive corresponding practice score data in real time;

[0041] The practical operation risk module is used to provide real-time risk warning when users perform practical electricity operations.

[0042] Compared with the prior art, the present invention has the following beneficial effects:

[0043] The AI-based intelligent assisted scoring method does not need to rely on a large number of manual coaches, reducing the labor and training costs generated by hiring coaches; at the same time, it breaks the limitations of time and space, and trainees can conduct practical training and obtain scoring feedback anytime and anywhere, reducing dependence on manual coaches, and providing a more convenient and efficient way for power industry practitioners to improve their skills; and users can score and analyze some process steps according to their needs, which facilitates users to conduct practical training and evaluation for unskilled steps, avoiding the need for users to repeatedly train skilled practical steps in order to achieve AI scoring, thereby improving practical efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0045] Figure 1 This is a principle block diagram of the present invention. DETAILED DESCRIPTION

[0046] The technical solutions of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0047] like Figure 1 As shown in the figure, the AI-based intelligent assisted scoring method for practical power operation skills includes:

[0048] Step 1: The platform establishes a practical standard library for scoring electric power practical operation, and stores the practical standards and scoring standards corresponding to various electric power practical operation projects through the practical standard library; the practical standard library can be set up according to the current electric power practical operation examination standards, evaluation standards, industry specifications, industry requirements, etc.; based on the practical standard library, the platform establishes a corresponding practical evaluation model based on AI technology, and the practical evaluation model analyzes the electric power practical operation video of the corresponding electric power practical operation project to obtain corresponding practical scoring data, including practical scores and corresponding unmet practical standards, and can also include step scores for each step, comprehensive scores, standard practical operation tutorials and other related data.

[0049] In one embodiment, because there is currently an intelligent model for intelligent evaluation of power operation videos, the platform can establish it based on the existing AI algorithm and optimize and adjust the existing intelligent model.

[0050] Step 2: Build a user transmission channel, such as through mini-programs, apps, web pages, etc. to transmit subsequent data to the user end used by the user. The user transmits the power practice video and practice project information through the user transmission channel; the practice project information is the relevant information filled in by the user for evaluating what practice project, whether to further limit the practice process, etc. For example, if the normal practice project has three process steps ABC, the user can only score and analyze the two process steps BC according to needs; it is convenient for users to conduct practical training and evaluation for unskilled steps, avoiding the need for users to repeatedly train skilled practice steps in order to achieve AI scoring, thereby improving practice efficiency.

[0051] For example, when a practitioner wants to evaluate his or her electricity practice, he or she can record the electricity practice video by himself or herself, attach corresponding practice project information, and transmit the electricity practice video and practice project information through the user terminal preset by the platform.

[0052] Step 3: Receive the power practice video and practice project information transmitted by the user in real time, analyze the power practice video and practice project information through the practice evaluation model, obtain the corresponding practice score data, and send the practice score data to the corresponding user.

[0053] In one embodiment, because the electric power practice video is recorded and collected by the user, the collected video quality is likely to be poor, which affects the evaluation results. Therefore, a video quality analysis is performed on the received electric power practice video and the practice project information to determine whether the electric power practice video meets the practice video requirements and obtain a corresponding video quality evaluation result. The video quality evaluation result includes a qualified video quality and an unqualified video quality.

[0054] When the video quality assessment result shows that the video quality is qualified, no corresponding operation is performed;

[0055] When the video quality evaluation result is that the video quality is unqualified, video capture defects are determined, video feedback data is generated according to the video capture defects, and the video feedback data is sent to the user.

[0056] In one embodiment, determining whether a practical power operation video meets practical operation video requirements can be done based on existing methods, such as presetting practical operation video requirements for corresponding practical operation projects and analyzing the practical power operation video and practical operation project information based on the practical operation video requirements. For example, a corresponding quality assessment model is established based on a CNN network, a DNN network, etc., and a corresponding training set is manually established for training. The training set includes input data and output data. The input data is the practical power operation video, practical operation project information, and practical operation video requirements, and the output data is the video quality assessment result. The judgment is made using the quality assessment model after successful training.

[0057] In one embodiment, a method for determining whether a power operation video meets the requirements of a power operation video includes:

[0058] Identify the practical project information, match the practical video requirements of the corresponding power practical video according to the practical project information, and adjust the power practical video and practical video requirements according to the practical project information, that is, remove data outside the steps corresponding to the practical project information, that is, cut and extract unnecessary requirements and videos to improve subsequent judgment efficiency; or do not make adjustments; mark the adjusted power practical video and practical video requirements as practical analysis video and practical analysis requirements respectively;

[0059] A quality assessment model is established. The expression of the quality assessment model is:

[0060]

[0061] Where (s, k) is the input data, s is the practical analysis video, and k is the practical analysis requirement. s→k indicates that the practical analysis video meets the practical analysis requirement. The corresponding training set is set using historical power practical videos for training. The output data is the quality assessment value ZP(s, k), which is 1 or 0.

[0062] Analyze the practical analysis video and the practical analysis requirements through the quality assessment model to obtain the quality assessment value of the corresponding power practical operation video;

[0063] When the quality evaluation value is 1, the video quality evaluation result is that the video quality is qualified;

[0064] When the quality evaluation value is 0, the video quality evaluation result is that the video quality is unqualified.

[0065] When it is determined that the video quality is unqualified, the video acquisition defects are determined based on the reasons for the unqualifiedness, mainly defects related to operation occlusion and clarity.

[0066] In one embodiment, the video feedback data includes video acquisition shortcomings.

[0067] In one embodiment, the video feedback data includes video capture shortcomings and the impact of the shortcomings on the scores. The impact of the shortcomings on the scores of the power practice is the impact of the video capture shortcomings on the scores of the power practice, which is determined based on the scoring criteria and the scoring situation. For example, the impact of various video capture shortcomings on the scores of the power practice is pre-counted, summarized and sorted, and then matched.

[0068] In one embodiment, the video feedback data includes video capture shortcomings, impact of the shortcomings scores, and suggestions for overcoming the shortcomings. The suggestions for overcoming the shortcomings are methods that the user can use to overcome the video capture shortcomings, which are generated based on the video capture shortcomings.

[0069] In one embodiment, the suggestion for overcoming the shortcomings can be intelligently generated based on existing shortcomings analysis methods and overcoming approaches.

[0070] In one embodiment, a method for determining a suggestion for overcoming the disadvantages includes:

[0071] Establish corresponding overcoming directions for various video acquisition shortcomings, such as increasing video acquisition points, replacing acquisition equipment with higher acquisition quality, etc.; determine optional overcoming directions based on video acquisition shortcomings, and generate several possible solutions to the video acquisition shortcomings based on the overcoming directions. The solutions are mainly generated based on historical solutions or the platform's preset database for storing solutions to various video acquisition shortcomings; specifically, existing methods can be used for generation; such as establishing intelligent models based on existing common AI technologies, deep learning algorithms, etc., and generating several alternative methods through analysis using the intelligent models.

[0072] Based on the electric power practical operation video, the corresponding candidate methods are simulated and screened to determine the suggestions for overcoming the shortcomings; that is, simulation is performed according to the candidate methods to determine whether the shortcomings of the video acquisition can be overcome, such as whether all the obscured practical steps can be collected, whether the clarity meets the requirements, etc.; the candidate methods that cannot be overcome are eliminated, and the remaining candidate methods are sorted according to priority to obtain a priority sequence, and suggestions for overcoming the shortcomings are generated according to the priority sequence, that is, they are generated according to the template of the suggestion for overcoming the shortcomings, which simply includes only the priority sequence, and can also be supplemented with data such as the advantages and disadvantages, and costs of each candidate method.

[0073] In one embodiment, simulation screening and priority sorting of corresponding candidate methods are performed based on the power operation video, and existing technologies can be used for screening evaluation and priority calculation, such as priority sorting from a cost perspective.

[0074] In one embodiment, simulation screening of candidate methods based on the power operation video includes:

[0075] The power operation video is simulated and supplemented according to the selected method to obtain the corresponding power simulation video. The simulation can be performed according to existing prediction, inference, artificial intelligence and other related technologies to determine the power operation video collected according to the selected method;

[0076] Analyze the power simulation video through the quality assessment model to obtain the corresponding quality assessment value;

[0077] When the quality assessment value is 1, the screening result is qualified;

[0078] When the quality assessment value is 0, the screening result is unqualified.

[0079] In one embodiment, evaluating the priorities of the remaining candidate methods includes:

[0080] Evaluate the implementation cost of each candidate method and determine the priority in order of implementation cost from smallest to largest.

[0081] In one embodiment, in order to address the safety risk concerns of some users during actual operation, a real-time warning function can be set up, that is, real-time video monitoring is performed when the user is performing actual electricity operation, and the power operation video is transmitted to the platform in real time. The actual operation evaluation model performs real-time analysis to determine whether there is an operation error and the risk situation of the corresponding error, and then issues a risk warning to the corresponding user; auxiliary warnings can also be combined with temperature monitoring, infrared monitoring, etc.

[0082] That is, when users are conducting practical electricity operations, they can transmit the user's ongoing practical electricity operation video and practical operation project information in real time through the user transmission channel, conduct real-time analysis of the practical electricity operation video and practical operation project information, obtain practical operation scoring data, conduct risk warning analysis based on the practical operation scoring data, obtain risk warning results, and send the practical operation scoring data and risk warning results to the user.

[0083] In one embodiment, risk warning is performed based on the actual operation scoring data, and analysis can be performed according to existing warning analysis methods.

[0084] An AI-based intelligent assisted scoring system for practical power operation skills, including both the platform and user sides;

[0085] The platform includes a practical standard library, a practical evaluation module, and a quality assessment module;

[0086] The practical operation standard library is used to store the practical operation standards and scoring standards corresponding to various practical operation projects.

[0087] The practical evaluation module is used to perform practical evaluation, receive the power practical operation video and practical operation project information transmitted by the user end in real time, analyze the power practical operation video and practical operation project information through a preset practical evaluation model, and obtain practical operation scoring data. The practical evaluation model is established by the platform based on the practical operation standard library; and the practical operation scoring data is sent to the corresponding user.

[0088] The quality assessment module is used to perform video quality analysis on the received power practice video and practice project information, obtain a video quality assessment result of the power practice video, and perform corresponding feedback processing according to the video quality assessment result.

[0089] The user terminal includes a video module and a practical risk module;

[0090] The video module is used for users to send power practice videos and practice project information to the platform, and receive corresponding practice score data in real time.

[0091] The practical operation risk module is used to provide real-time risk warning when users perform practical electricity operations.

[0092] The above formulas are all calculated by removing dimensions and taking their numerical values. The formula is a formula that is closest to the actual situation obtained by collecting a large amount of data and performing software simulation. The preset parameters and preset thresholds in the formula are set by technicians in this field according to actual conditions or obtained by simulating a large amount of data.

[0093] The above embodiments are only used to illustrate the technical method of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. The AI-based intelligent auxiliary scoring method for electric power practical skills is characterized by: include: The platform establishes a practice standard library, which is used to store practice standards and scoring criteria corresponding to various practice projects; and establishes a practice evaluation model based on the practice standard library; The platform builds a user transmission channel, through which users transmit power practice videos and practical project information; The power practice video and the practice project information transmitted by the user are received in real time, the power practice video and the practice project information are analyzed by the practice evaluation model to obtain practice score data, and the practice score data is sent to the corresponding user.

2. The AI-based intelligent auxiliary scoring method for electric power practical skills according to claim 1 is characterized in that: The practical operation standard library is updated in real time, and the practical operation evaluation model is updated and adjusted after the practical operation standard library is updated.

3. The AI-based intelligent assisted scoring method for electric power practical skills according to claim 1 is characterized in that: Performing video quality analysis on the received power operation video and the operation project information to obtain a video quality assessment result of the power operation video, wherein the video quality assessment result includes a qualified video quality and an unqualified video quality; When the video quality assessment result shows that the video quality is qualified, no corresponding operation is performed; When the video quality evaluation result is that the video quality is unqualified, video capture defects are determined, video feedback data is generated according to the video capture defects, and the video feedback data is sent to the user.

4. The AI-based intelligent auxiliary scoring method for electric power practical skills according to claim 3 is characterized in that: Methods for analyzing the video quality of power operation videos and operation project information include: Identify the practical project information, match the practical video requirements of the power practical video according to the practical project information, adjust the power practical video and the practical video requirements according to the practical project information, and obtain the practical analysis video and the practical analysis requirements; A quality assessment model is established. The expression of the quality assessment model is: Where (s, k) is the input data, s is the practical analysis video, and k is the practical analysis requirement. s→k indicates that the practical analysis video meets the practical analysis requirement. The output data is the quality assessment value ZP(s, k), which is 1 or 0. Analyze the practical operation analysis video and the practical operation analysis requirements through a quality assessment model to obtain a quality assessment value of the power practical operation video; When the quality evaluation value is 1, the video quality evaluation result is that the video quality is qualified; When the quality evaluation value is 0, the video quality evaluation result is that the video quality is unqualified.

5. The AI-based intelligent auxiliary scoring method for electric power practical skills according to claim 3 is characterized in that: Video feedback data includes video capture shortcomings and the impact of shortcomings on ratings.

6. The AI-based intelligent auxiliary scoring method for electric power practical skills according to claim 3 is characterized in that: Video feedback data includes video capture shortcomings, the impact of shortcomings on scoring, and suggestions for overcoming shortcomings.

7. The AI-based intelligent auxiliary scoring method for electric power practical skills according to claim 6 is characterized in that: Suggested methods for overcoming shortcomings include: Generating a plurality of candidate methods according to the video acquisition shortcomings, wherein the candidate methods are solutions that have a probability of solving the video acquisition shortcomings; Conduct simulation screening of the corresponding candidate methods based on the power operation video to obtain the screening results of the corresponding candidate methods, including unqualified and qualified screening results; Prioritize the candidate methods that have passed the screening results to obtain a priority sequence, and generate suggestions for overcoming shortcomings based on the priority sequence.

8. The AI-based intelligent auxiliary scoring method for electric power practical skills according to claim 7 is characterized in that: Simulate screening of candidate methods based on practical power operation videos, including: Perform simulation supplement on the electric power operation video according to the selected method to obtain the electric power simulation video of the selected method; Analyze the power simulation video using a preset quality assessment model to obtain a quality assessment value of the selected method; When the quality assessment value is 1, the screening result is qualified; When the quality assessment value is 0, the screening result is unqualified.

9. The AI-based intelligent auxiliary scoring method for electric power practical skills according to claim 1 is characterized in that: When users are conducting practical electricity operations, they can transmit the practical electricity operation videos and practical operation project information of the users in real time through user transmission channels, conduct real-time analysis of the practical electricity operation videos and practical operation project information, obtain practical operation scoring data, conduct risk warning analysis based on the practical operation scoring data, obtain risk warning results, and send the practical operation scoring data and risk warning results to users.

10. The AI-based intelligent auxiliary scoring system for electric power practical skills is characterized by: Executing the AI-based intelligent assisted scoring method for electric power practical skills as described in any one of claims 1 to 9, including a platform end and a user end; The platform includes a practical standard library, a practical evaluation module, and a quality assessment module; The practical standard library is used to store the practical standards and scoring standards corresponding to various practical projects; The practical evaluation module is used to perform practical evaluation, receive the power practical operation video and practical operation project information transmitted by the user end in real time, analyze the power practical operation video and practical operation project information through the preset practical evaluation model, obtain practical operation score data; and send the practical operation score data to the corresponding user; The quality assessment module is used to perform video quality analysis on the received power operation video and the practical operation project information, obtain the video quality assessment result of the power operation video, and perform corresponding feedback processing according to the video quality assessment result; The user terminal includes a video module and a practical risk module; The video module is used for users to send power practice videos and practice project information to the platform, and receive corresponding practice score data in real time; The practical operation risk module is used to provide real-time risk warning when users perform practical electricity operations.

Images