Fuzzy Reasoning Application In Insulator Detecting System

The fuzzy reasoning alarming model application in contamination on-line detecting system for substation is discussed in this paper.Some basic concepts of insulator contamination are introduced firstly. Existing methods of avoiding contamination flashover and methods of detecting contamination degree of insulators are discussed in details. These methods are analyzed and disadvantages of them are pointed out.The nature and character of leakage current are analyzed. A high precised and broad bandwidth sensor is developed. The design of on-line contamination detecting system is introduced. The relationships between leakage current, environment humidity, environment temperature and level of pollution are studied, then, the problem of fixed value alarming model is analyzed. It is proposed that in order to increase the reliability of prediction, the nonlinear prediction model, which depends on the nonlinear relationships between above factors, is indispensable. Because the nonlinear relationships can hardly be obtained, a nonlinear alarming model based on fuzzy reasoning method is developed tosimulate the complex relationship in this paper.Sets of reasoning rules are obtained via three methods, which is summarizing the experiences of expert, analyzing the date of local supervising, studying the date of simulate experiments in lab. The rules are revised by simulation on computer.The results of theoretical analyzing and simulation proved that the fuzzy reasoning alarming method could predict the pollution degree of insulator more accurately. A new method is given to solve the problem of alarming reliability of contamination on-line detecting system.