Study Of Substation Equipment Fault Detection And Recognition Based On Infrared Thermal Image

In recent years,with the development of our country's social economy,the power industry has also made unprecedented progress,especially the development of smart grids has become a major demand for my country's power industry.The operation status monitoring and fault detection and identification of traditional substations mainly rely on the regular inspection and operation of the staff,which not only takes time and effort,but also increases the interference of relatively subjective judgment factors,and is accompanied by a certain degree of danger.In special cases It is even difficult to ensure fast and accurate detection and identification of substation equipment.With the rapid development of image processing and artificial intelligence technology,they have been widely used in various industries.This paper takes infrared images of substation equipment as the research object.Through image processing,feature extraction,classification and recognition technologies,it is expected to achieve the purpose of intelligent detection and identification of faults in substation equipment,thereby reducing the pressure of manual inspection and improving the power of substation equipment.Early warning capabilities for potential safety hazards to ensure safe and reliable operation of the power grid.The specific research contents of this article are as follows:(1)This paper analyzes the background,significance and research status of infrared image detection technology at home and abroad,discusses the application status of infrared image detection technology in fault detection of current substation equipment,and introduces the operation characteristics and structural characteristics of substation equipment.Several common types of thermal faults in power equipment.(2)In this paper,the preprocessing operation of infrared images of substation equipment is studied.For the shortcomings of infrared images,such as high noise,low contrast,and blurry images,the image denoising algorithms such as mean filtering,median filtering,and wavelet threshold are simulated.In contrast,a wavelet threshold denoising method based on median filtering is proposed;on this basis,in order to improve the contrast and clarity of the original infrared image,an image enhancement processing method based on the pulse coupled neural network PCCN model is proposed,The simulation of Matlab platform was used to verify the effectiveness of the method.Finally,in order to highlight the target image,Canny operator was used to segment the image to achieve a good segmentation effect.(3)This paper proposes an infrared image classification and recognition algorithm for substation equipment based on HOG feature extraction and support vector machine,so that the appearance and shape of local targets can be well described by the gradient or edge direction density distribution,thereby improving the substation equipment The accuracy of infrared image classification and recognition.(4)In order to avoid the misjudgment of human analysis and the impact of environmental interference,based on the above research,on the basis of the relative temperature difference method to classify the substation equipment faults,based on the MATLAB_GUI platform,the corresponding substation equipment fault detection and recognition system is equipped And,through the infrared image test of various types of high-voltage substation equipment,the accuracy of the system is verified.In this paper,the infrared image of the substation equipment is studied through image processing technology,and the infrared temperature measurement image can be used to improve the operation and maintenance efficiency of the equipment in the unattended substation,thus providing a theoretical basis for the intelligent detection and identification of actual substation equipment faults.Certain practical reference value.