Research On Aging State Of Transformer Based On Analysis Of Particles In Oil

In power systems,transformers play an important role in the conversion,transmission,and distribution of electrical energy.It is the core equipment of the power system,and its failure may cause large-scale power outages and other accidents,causing serious economic losses.According to statistics,the main reason for the transformer failure is the insulation failure caused by the aging of the internal oil-paper composite insulation,which causes the operation process to fail.Therefore,regular evaluation of the aging status of the transformer insulation system can not only monitor the operation status of the transformer,but also provide necessary and reliable basis for formulating corresponding maintenance strategies,effectively preventing or reducing accidents.This article starts from the perspective of the aging characteristics of oil-paper insulation systems.Based on the detection and analysis of particulate matter in transformer oil,an in-depth study of the relationship between particulate matter in oil and the aging state of transformers is conducted.The main research contents are as follows:(1)The aging status of 10 transformers randomly selected in Shandong province whose operating life is more than 10 years is investigated,the aging data of the main components and structural materials of these transformers are collected and analyzed,and it is confirmed that the aging of oil-paper insulation system is the main cause of transformer aging.And on this basis,the aging mechanism of the oil-paper insulation system of transformers and the theoretical basis for the aging state research are studied in depth.(2)According to the detection standards and requirements of particulate matter in oil,appropriate hardware equipment such as camera,lens and light source were selected,and a dual-path image acquisition based transformer oil particulate matter detection system was designed and built.The detection system uses the dual lens structure to obtain the particle size and shape information from two perpendicular angles,which eliminates the effect of particle overlap on the measurement results.(3)Considering that after many years of actual operation of the transformer,the particles in transformer oil are not only fiber particles,but also mixed with iron,copper,silicon dioxide and other particles,which will interfere with the diagnosis of the aging status of the oil paper insulation.Therefore,in order to avoid such problems,the HOG+SVM algorithm was applied to the detection of particulate matter in transformer oil,which improves the utilization of parameter information such as particle size and shape,thereby achieving accurate identification of fiber particles.(4)In order to study the relationship between the particulate matter in transformer oil and the aging state of insulation,the accelerated aging simulation experiment was carried out,and the particulate matter detection system of transformer oil based on dual-path image acquisition was built to detect the particulate matter generated by aging.The methods of multivariate statistical analysis and reliability analysis were used to analyze the typicality and dispersion of the measured data,and the differences and influencing factors of measured fiber particle parameters in characterizing transformer insulation aging were analyzed by referring to relevant standards.