Operating State Diagnosis Strategy Of Transmission Line Insulators Based On FPN

The pollution flashover of transmission line insulators is one of the major hidden threats to the power grids.Due to the difference of regions,the impact of the environment and industrial pollutants,insulators flashover accidents is still occur frequently.Therefore,the analysis of the pollution flashover mechanism and discharge model still needs to be studied.After analyzing existing techniques,a smart diagnosis system of insulator operating status is presented which is based on artificial intelligence,fuzzy inference system,Petri network and neural network theories.The inference framework can determine its pollution status and flash probability by monitoring relevant feature quantities.Because of the relationship between flashover accident and leakage current on the surface of the insulator,this paper improves a leakage current monitoring strategy with remote,fault location,and flashover protection using Rogowski coils.The strategy can measure the characteristics of the leakage current,and then use the fuzzy inference rules of Fuzzy Petri Nets(FPN)to describe the complex relationship between the electrical and non-electrical parameters.In the complex reasoning node,this paper introduces an optimized neural network learning strategy to improve the weights of each layer adaptively.So that the diagnostic system can update existing knowledges to predict the pollution degree and the flashover status according to actual conditions.Finally,this paper uses the actual experimental data to test examples.The simulation experiments show that the proposed intelligent diagnosis method can effectively monitor the pollution degree of insulators and evaluate the probability of flashover to determine whether there is a need for cleaning jobs.The insulator diagnosing and monitoring system designed in this paper can effectively improve the failure patrol efficiency of the grid equipment,reduce the number of poles for maintenance personnel and make the inspection more targeted.It is true that it can save time and labor costs,so it has important engineering application value and theoretical research significance.