Application Of Infrared Image Processing In Fault Diagnosis Of Overhead Line

In recent years,with the rapid development and improvement of the power industry,the power grid has also developed rapidly,driving the continuous improvement of the operating voltage level of the power grid system and the gradual expansion of the network scale.Due to the increase in the operating voltage level of the power grid and the ultra-long distance transmission of electricity,the country has stricter requirements on the safety and reliability of power grid operation.Traditional power fault diagnosis and maintenance methods are time-consuming and laborious,and even require power outage detection,which brings great inconvenience to people’s lives.In recent years,infrared thermal imaging technology has developed rapidly and gradually matured in the power industry.Infrared imaging is to diagnose the state of electrical equipment when it is running based on the thermal energy distribution of the equipment.It has the characteristics of remote shooting without direct contact,and it can also obtain equipment information without taking the shutdown or closing operations.To a large extent weaken the risk of operation.This article proposes the application of infrared image processing to the fault diagnosis of overhead lines.First,it introduces the basic principles of infrared thermal imaging and image segmentation,as well as the composition and common faults of overhead lines.Secondly,perform image gray-scale and denoise filtering on the infrared image,and then propose an improved K-means algorithm,based on the iterative idea of the traditional K-means algorithm,use the gray-scale histogram to extract the gray-scale frequency distribution information and preset the cluster value k,And use the concept of quantile instead of the method of randomly selecting the initial cluster centers in the traditional algorithm,and then perform segmentation simulation on the preprocessed image.At the same time,an optimization criterion is designed to verify the preset clustering value of this subject.Through comparative analysis of multiple cases,it is proved that the improved K-means algorithm is not only more accurate and efficient.Finally,the improved K-means algorithm is applied to the actual fault diagnosis case of overhead lines,and the final result of the fault diagnosis is obtained.Through the diagnosis example analysis of the clamps,insulators and fittings on overhead lines,it is proved that the improved K-means algorithm has achieved good application effects in the fault diagnosis of overhead lines.The result of the case analysis shows that the diagnosis result is accurate,the algorithm is simple to run,reliable,and practical.Figure[35] Reference[81]...