Research On Automatic Monitoring Technology Of Power Equipment Combined With Image Processing And Deep Learning

In recent years,our country’s electricity consumption has increased every year.Automatic monitoring of electricity consumption with electrical equipment is an extremely important link in the power supply system.The safe operation of the power supply system is related to the safety of life and property.This has caused the number of power equipment to monitor the status of electricity consumption to increase continuously,resulting in shortages of inspection workers and personnel on duty,increased operating costs,lower labor efficiency,and other problems.Moreover,power equipment is generally built in areas with a more complex environment,and the staff is also required to conduct inspections not safe.Moreover,with the development of computer software and hardware,deep learning technology has gradually matured.Researchers have used deep learning technology in various fields,including the field of automatic monitoring of power equipment,so that workers do not need to manually copy the status of equipment on site,and can be used directly.The front-end equipment automatically monitors the status of the equipment.However,the use of deep learning methods cannot simultaneously identify various types of objects on power equipment and determine their status.Moreover,when the shooting position of the inspection device is deviated or the environment is bad,the recognition of the object category will be affected,thus the object The state discrimination produces a large error.This paper analyzes the problems encountered in the automatic monitoring technology of power equipment based on the above problems.According to the characteristics of the power equipment scene,a method combining image processing and deep learning is proposed.This method is used to analyze and process the detection target.First,use the deep learning target detection method to identify the category of the object.For this part,this paper proposes a multi-target detection model.This model is based on the original YOLOv3 model and optimizes the FPN feature fusion layer and loss function part,and The K-means clustering method is used to modify the a priori box size of the convolutional neural network to fit the size of the object in the power equipment scene;Secondly,according to different categories,different image processing methods are used to determine the status.For this part,this article uses a multi-category comprehensive discrimination method to determine the status of each object.This method is to integrate various image processing methods according to different categories after target detection.In the case of different statuses on the indicator light,knife switch,and instrument panel,it is impossible to quickly judge the changed status.Finally,the state discrimination algorithm is integrated with the deep learning target detection network to form an automatic monitoring technology for power equipment that combines image processing and deep learning.The experiments in this paper conduct performance verification on the public data set and the self-labeled data set,and set up a comparative experiment to verify the effect of the improved network structure above the design on the model recognition performance.The multi-target detection model designed in this paper is obtained through experiments,which improves the detection accuracy of small objects such as indicator lights and switch switches with inconspicuous features,and the predicted object coordinates become more accurate.According to the accurate coordinates,the status will be improved.The discrimination part is more accurate,so as to ensure the accuracy of network detection;Secondly,the multi-category comprehensive discrimination method makes the judgment of the operating state of the power equipment more accurate.Finally,through the use of automatic power equipment monitoring technology that combines image processing and deep learning,not only the problem of not being able to identify various types of objects on power equipment and distinguish their status at the same time is solved,but also the monitoring efficiency is greatly improved,saving a lot of labor.Labor,reduce the frequency of accidents.