| The performance of the insulator directly affects the safe and reliable operation of the transmission line,but the insulator may be partially deteriorated due to the manufacturing process or poor working environment,which affects the reliability of power supply.It is very important to find out and replace the deteriorated insulator in time to ensure the safe and reliable operation of the transmission line,but the efficiency of traditional manual line inspection is low,which can not find the deteriorated insulator in time,resulting in the decline of power supply quality in the power sector.Therefore,it is very important for the safe and reliable operation of power grid to analyze the distribution characteristics of insulator electric field and put forward the insulator fault detection strategy.Based on the project background of Huainan-Huxi UHV transmission line,this paper sums up the types,functions and structural characteristics of the insulators,selects XWP-70 line insulator as the research object,solves the electric field distribution based on the finite element method,summarizes the electric field distribution characteristics of the insulator string,further analyzes the common fault types and hazards of the insulator,and aims at the deteriorated insulator This kind of fault with high probability of occurrence and high degree of harm has been studied in depth.The deteriorated insulator model has been established in ANSYS,and its influence on the electric field distortion effect of insulator has been explored through finite element calculation.On this basis,the transmission system model has been established based on ATP-EMTP,and the leakage current waveform information corresponding to the deteriorated insulator has been calculated in the time domain,which has been extracted In order to provide information for building insulator fault diagnosis model,10 feature quantities including root mean square,kurtosis,etc.are included.After the redundancy of feature quantity is determined by correlation factor analysis,the above feature quantity is further extracted by principal component analysis to realize the dimensionality reduction of original database,so as to provide effective feature quantity for fault diagnosis model.In addition,the effective feature quantity is determined by correlation factor analysis.After the data set is divided into training set and test set,the probabilistic neural network is established by MATLAB as the fault diagnosis model.Based on the training set and test set,the fault diagnosis model is established and verified.The results show that under normal working conditions,the electric field distribution of insulator string is concentrated on the conductor side of insulator,and the electric fieldin the middle part of insulator string is nearly evenly distributed.Improving the uniformity of the electric field on the conductor side of insulator string is conducive to the improvement of the overall electrical performance of insulator string;after the deterioration of insulation,the electrical insulation performance is reduced,resulting in the distortion of the electric field near the insulator string,which is embodied in the potential of the deteriorated part and the electric field of the insulator string The field strength value is close to zero,the electric field value nearby increases,and the distortion effect decreases with the increase of the distance from the deteriorated insulator to the conductor side.Through the extraction of the leakage current,it is found that the waveform characteristics vary with the different deterioration positions of the insulator.The fault diagnosis model can be established by taking the leakage current of the insulator as the breakthrough point.The model is based on the probabilistic neural network.By learning the effective characteristics of the leakage current,the intelligent fault location recognition can be realized.After verification,the accuracy of the judgment can reach 88.89%. |
PDF Full Text Request