Research On Intelligent Evaluation Method Based On Insulators Surface Contamination Online Detecting Result

With the rapid developing of modern economy and the continuous rising of power system voltage level, the influent of flashover failure to the national economy is more and more serious. Although currently used measures of flashover defending use a lot of human and material resources, the results are not satisfying. As previous research shows that the on-line monitoring system, which the contaminating condition of high voltage insulators is usually estimated by detecting the root mean square (r.m.s.) of surface leakage current, is limited because the judge depends on the leakage current, which is affected not only by the surface contamination of insulators but also by the environmental factors such as temperature, humidity, and so on. The relationship between the current and those factors mentioned above is nonlinear and very complicated. Therefore, this paper makes use of the intelligent control method to evaluate surface contamination states of the insulators. Firstly, some basic theories and on-line detecting measures existed of insulator contamination flashover are studied and analyzed, and the drawbacks are pointed out as well. The influence of environmental factors on the exterior electrical characteristics of insulators with different contamination conditions was analyzed depended on laboratory simulation experiments and field data. The leakage current r.m.s value, peak value, amplitude and times of the pulses and surrounding temperature and humidity are chosen as five input variables. Applying fuzzy reasoning method to the insulator contamination on-line monitoring is proposed.Furthermore, an improved contamination valuation method based on fuzzy neural network is introduced. Experiments in the laboratory and the field show the feasibility of this method. It provides a new way to solve the practicability matter oninsulators surface contamination on-line detecting system.