Design And Implementation Of Intelligent Detection System For Electrical Equipment

With the rapid development of the power industry,the country's reliability requirements for power grid systems are getting higher and higher.Online monitoring and safety warnings have become important functional requirements of the grid system.However,in order to find problems in time to eliminate hidden dangers,it takes a lot of human objects,so the intelligentization and automation of the grid is gradually put on the agenda.Infrared monitoring and diagnosis technology is a very effective online monitoring method.It can not only find defects through online detection,but also can be combined with other methods to locate faults,which brings great convenience to maintenance.In recent years,deep learning technology has advanced by leaps and bounds,and more and more recognition tasks in the image field have achieved good performance in deep learning solutions.The long-term operation of the grid system has accumulated a large amount of data,which also provides the possibility of exploration for deep learning applications.This paper designs and develops an infrared image anomaly detection system based on deep learning,which can accurately identify and judge the presence or absence of abnormalities in several common electrical equipments,such as knives,capacitors,and voltage transformers.The system functions mainly include three parts:infrared diagnosis,infrared image management,and inspection task arrangement.Among them,the VGG network of the convolutional neural network is applied in the main infrared diagnosis part for identification.The image management section uses well-designed MySQL database data as a data support.The inspection task arrangement section arranges the maintenance workers to troubleshoot according to the test results.Based on the actual needs of the power grid,this paper introduces the technology,system architecture design,algorithm implementation and system implementation of the system in detail.