Intelligent inspection image recognition method for power transmission and transformation equipment based on artificial intelligence

Abstract

The invention relates to an intelligent inspection image recognition method for power transmission and transformation equipment based on artificial intelligence, and the method achieves the monitoring comprehensiveness and accuracy of the power transmission and transformation equipment by comprehensively analyzing various indexes through multi-latitude monitoring in a manner of fusing satellite remote sensing positioning, unmanned aerial vehicle inspection visual recognition and operation data external detection equipment. Meanwhile, an intelligent inspection terminal and an edge computing technology are utilized, so that real-time data processing and comprehensive analysis are realized, and the accuracy of monitoring data is improved; moreover, by preprocessing and enhancing the image data set, smoothing and sharpening optimization of the image data can be realized, and the contrast and definition of the image are enhanced, so that the defect identification efficiency of the power transmission and distribution equipment is improved; the enhanced image data set is segmented and classified according to the type of equipment, so that the rapidity and accuracy of subsequent defect identification can be improved; the method has the advantages of being accurate, comprehensive, economical, efficient and stable and reliable in operation.

Description

Technical Field

[0001] The present invention belongs to the technical field of electric power inspection, and in particular relates to an artificial intelligence-based intelligent inspection image recognition method for power transmission and transformation equipment. Background Art

[0002] The electric power industry is a basic pillar industry for national energy security and national economic development. Among them, the efficient and reliable operation of power transmission and transformation equipment has become the cornerstone for ensuring the safety of the power grid and reliable power supply, and is of vital importance to promoting the high-quality development of the power system. During the operation of power transmission and transformation equipment, it is necessary to inspect it through inspection equipment, and collect data and images, so that operators can promptly discover and solve faults. However, the existing power transmission and transformation equipment has a relatively single inspection mode, and has shortcomings such as low defect recognition rate and weak safety management and control capabilities. It is difficult to adapt to the new requirements of high-quality development of the power grid and modern enterprise management. Therefore, in order to solve the above problems, it is necessary to develop an accurate, comprehensive, economical and efficient, stable and reliable intelligent inspection image recognition method for power transmission and transformation equipment based on artificial intelligence. Summary of the invention

[0003] The purpose of the present invention is to overcome the shortcomings of the prior art and to provide an accurate, comprehensive, economical and efficient, stable and reliable intelligent inspection image recognition method for power transmission and transformation equipment based on artificial intelligence.

[0004] The object of the present invention is achieved by: a method for intelligent inspection image recognition of power transmission and transformation equipment based on artificial intelligence, comprising the following steps: Step 1: Through the "air-space-ground" integrated collaborative inspection mode, integrating satellite remote sensing positioning, drone inspection visual recognition and real-time monitoring of external detection equipment for operation data, using multi-dimensional monitoring methods, the external status of power transmission and transformation equipment, line partial discharge, temperature and grounding status are detected, and the inspection images and monitoring data are transmitted to the background server in real time; Step 2: While storing and backing up the inspection images and monitoring data through the storage module inside the backend server, an image data set is established using the inspection images and monitoring data; Step 3: The image data set is smoothed and sharpened through the image preprocessing module, and the power transmission and transformation equipment and foreign objects in the channel environment are marked. Then, the image data set is preprocessed based on the saliency detection algorithm combining global and local features. After that, the image data set is enhanced through the image enhancement method to obtain the enhanced image data set. Step 4: Construct an image feature extraction network, and extract feature data values ​​based on the enhanced image data set to form an original training sample set. Then, segment the image data set according to the type of power transmission and transformation equipment to obtain a classified image data set. Step 5: Based on the type of equipment, determine the possible defects of the equipment to be identified. Then, according to the equipment defect prediction results, formulate multiple defect thresholds to build a defect recognition model. Finally, use the preset defect recognition model to perform defect recognition on the classified image data set. Step 6: Analyze and summarize the fault defect data of the power transmission and transformation equipment identified by the defect identification model through the intelligent analysis module, generate an analysis report, and propose at least two solutions. Then, use the information transmission module to transmit the analysis report and solutions to the operator's visualization terminal in real time. The operator arranges and issues maintenance instructions to the maintenance personnel through the visualization terminal.

[0005] Furthermore, the defect recognition model in step 5 is obtained by training the improved YOLOv5m network.

[0006] Furthermore, a quality evaluation unit is provided in the image preprocessing module in step 3, and the quality evaluation unit can be used to perform quality evaluation on the image data set to obtain the quality score of each frame of the power transmission equipment inspection image, and screen out the power transmission equipment inspection images whose quality score is greater than a preset score threshold.

[0007] Furthermore, in step 6, a fault grading unit is provided inside the intelligent analysis module, and the fault grading unit can be used to grade the fault defects of the power transmission and transformation equipment according to their severity.

[0008] Furthermore, in step 3, the image dataset is enhanced by an image enhancement method, including enhancing the contrast and clarity of the image and removing noise.

[0009] Beneficial effects of the present invention: The present invention adopts an integrated "air-space-ground" collaborative inspection mode, integrates satellite remote sensing positioning, drone inspection visual recognition and external detection equipment for operation data, and comprehensively analyzes various indicators through multi-dimensional monitoring, thereby achieving comprehensiveness and accuracy in monitoring power transmission and transformation equipment; at the same time, by utilizing intelligent inspection terminals and edge computing technology, data is processed and comprehensively analyzed in real time, which optimizes the timeliness of data acquisition, reduces labor costs, improves the accuracy of monitoring data, and provides reliable support for the safe and stable operation of power transmission and transformation systems; and, by preprocessing and enhancing the image data set, image data smoothing and sharpening optimization can be achieved, and while marking power transmission and transformation equipment and foreign objects in the channel environment, the image contrast and clarity are enhanced, as well as image noise is removed. By adopting this structure, the clarity of image data can be increased, thereby improving the efficiency of defect identification of power transmission and distribution equipment; by segmenting and classifying the enhanced image data set according to the type of equipment, the rapidity and accuracy of subsequent defect identification can be increased; in general, the present invention has the advantages of being accurate and comprehensive, economical and efficient, and stable and reliable in operation. DETAILED DESCRIPTION

[0010] The present invention is further described below.

[0011] Embodiment, a method for intelligent inspection image recognition of power transmission and transformation equipment based on artificial intelligence, comprising the following steps: Step 1: Through the "air-space-ground" integrated collaborative inspection mode, integrating satellite remote sensing positioning, drone inspection visual recognition and real-time monitoring of external detection equipment for operation data, using multi-dimensional monitoring methods, the external state of power transmission and transformation equipment, line partial discharge, temperature and grounding state are detected, and the inspection images and monitoring data are transmitted to the background server in real time. This method can improve the comprehensiveness of the inspection of transmission lines and substation equipment, while increasing the accuracy and timeliness of inspection data acquisition; Step 2: While storing and backing up the inspection images and monitoring data through the storage module inside the backend server, an image data set is established using the inspection images and monitoring data; Step 3: The image data set is smoothed and sharpened by the image preprocessing module, and the power transmission and transformation equipment and foreign objects in the channel environment are marked. In this process, the quality evaluation unit set in the image preprocessing module can be used to evaluate the quality of the image data set, so as to obtain the quality score of each frame of the power transmission equipment inspection image, and then the power transmission equipment inspection images with a quality score greater than the preset score threshold are screened out; then, the image data set is preprocessed based on a saliency detection algorithm that combines global and local features, and then the image data set is enhanced by an image enhancement method, that is, the contrast and clarity of the image are enhanced and the noise is removed, so as to obtain an enhanced image data set; Step 4: Construct an image feature extraction network, and extract feature data values ​​based on the enhanced image data set to form an original training sample set. Then, segment the image data set according to the type of power transmission and transformation equipment to obtain a classified image data set. Step 5: Based on the type of equipment, determine the possible defects of the equipment to be identified. Then, according to the equipment defect prediction results, formulate multiple defect thresholds, and obtain the defect recognition model by training the improved YOLOv5m network. Finally, use the preset defect recognition model to perform defect recognition on the classified image data set. Step 6: Analyze and summarize the fault defect data of the power transmission and transformation equipment identified by the defect identification model through the intelligent analysis module, generate an analysis report, and propose at least two solutions. At the same time, the fault classification unit inside the intelligent analysis module can be used to classify the fault defects of the power transmission and transformation equipment according to their severity; then, the analysis report and solution are transmitted to the operator's visualization terminal in real time by the information transmission module, and the operator arranges and issues maintenance instructions to the maintenance personnel through the visualization terminal.

[0012] The present invention innovates the intelligent inspection mode of power transmission equipment. Through multi-dimensional detection of the status of power transmission and transformation equipment, combined with satellite remote sensing technology and drone image recognition technology, the "air-space-ground" integrated collaborative inspection mode is used, integrating satellite remote sensing positioning, drone inspection visual recognition and external detection equipment for operation data, and comprehensively analyzing various indicators through multi-dimensional monitoring, thus achieving the comprehensiveness and accuracy of power transmission and transformation equipment monitoring; at the same time, by using intelligent inspection terminals and edge computing technology, data is processed and comprehensively analyzed in real time, which optimizes the timeliness of data acquisition, reduces labor costs, and improves the accuracy of monitoring data, providing a reference for power transmission and transformation systems. It provides reliable support for the safe and stable operation of the system; and, by preprocessing and enhancing the image data set, it can achieve smoothing and sharpening optimization of the image data, and while marking the power transmission and transformation equipment and foreign objects in the channel environment, it can enhance the contrast and clarity of the image, as well as remove image noise. By adopting this structure, the clarity of the image data can be increased, thereby improving the efficiency of defect identification of the power transmission and distribution equipment; thereafter, by segmenting and classifying the enhanced image data set according to the type of equipment, the rapidity and accuracy of subsequent defect identification can be increased; in general, the present invention has the advantages of being accurate and comprehensive, economical and efficient, and operating stably and reliably.

[0013] The above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit the same. Although the present invention has been described in detail with reference to the above embodiments, a person skilled in the art should understand that the specific implementation modes of the present invention can still be modified or replaced by equivalents, and any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be included in the scope of the claims of the present invention.

Claims

1. An artificial intelligence-based intelligent inspection image recognition method for power transmission and transformation equipment, characterized in that: The following steps are involved: Step 1: Through the "air-space-ground" integrated collaborative inspection mode, integrating satellite remote sensing positioning, drone inspection visual recognition and real-time monitoring of external detection equipment for operation data, using multi-dimensional monitoring methods to detect the external status of power transmission and transformation equipment, line partial discharge, temperature and grounding status, and transmit the inspection images and monitoring data to the background server in real time; Step 2: While storing and backing up the inspection images and monitoring data through the storage module inside the backend server, an image data set is established using the inspection images and monitoring data; Step 3: The image data set is smoothed and sharpened through the image preprocessing module, and the power transmission and transformation equipment and foreign objects in the channel environment are marked. Then, the image data set is preprocessed based on the saliency detection algorithm combining global and local features. After that, the image data set is enhanced through the image enhancement method to obtain the enhanced image data set. Step 4: Construct an image feature extraction network, and extract feature data values ​​based on the enhanced image data set to form an original training sample set. Then, segment the image data set according to the type of power transmission and transformation equipment to obtain a classified image data set. Step 5: Based on the type of equipment, determine the possible defects of the equipment to be identified. Then, according to the equipment defect prediction results, formulate multiple defect thresholds to build a defect recognition model. Finally, use the preset defect recognition model to perform defect recognition on the classified image data set. Step 6: Analyze and summarize the fault defect data of the power transmission and transformation equipment identified by the defect identification model through the intelligent analysis module, generate an analysis report, and propose at least two solutions. Then, use the information transmission module to transmit the analysis report and solutions to the operator's visualization terminal in real time. The operator arranges and issues maintenance instructions to the maintenance personnel through the visualization terminal.

2. The method for intelligent inspection image recognition of power transmission and transformation equipment based on artificial intelligence as claimed in claim 1, characterized in that: The defect recognition model in step 5 is obtained by training the improved YOLOv5m network.

3. The method for intelligent inspection image recognition of power transmission and transformation equipment based on artificial intelligence as claimed in claim 1, characterized in that: In the step 3, a quality evaluation unit is provided in the image preprocessing module, and the quality evaluation unit can be used to evaluate the quality of the image data set to obtain the quality score of each frame of the power transmission equipment inspection image, and screen out the power transmission equipment inspection images whose quality score is greater than a preset score threshold.

4. The method for intelligent inspection image recognition of power transmission and transformation equipment based on artificial intelligence as claimed in claim 1, characterized in that: In step 6, a fault classification unit is provided inside the intelligent analysis module, and the fault classification unit can be used to classify the fault defects of the power transmission and transformation equipment according to the severity of the fault defects.

5. The method for intelligent inspection image recognition of power transmission and transformation equipment based on artificial intelligence as claimed in claim 1, characterized in that: In step 3, the image data set is enhanced by an image enhancement method, including enhancing the contrast and clarity of the image and removing noise.