ANFIS Based Contamination Level Prediction of Porcelain Insulator Under Various Pollution Conditions

作者：Krishnamoorthi K^*; Chandrasekar S

来源：International Review of Electrical Engineering, 2011, 6(2): 1038-1045.

• Porcelain Insulators
• Contamination Flashover
• Leakage Currents
• ANFIS Model
• Prediction Characteristics
• Contamination Prediction

This paper deals with analysis of leakage current of porcelain insulator in order to develop a better diagnostic tool to identify the pollution severity of outdoor insulators. In this work laboratory based Pollution performance tests are carried out on porcelain insulator under AC voltage at different pollution levels and relative humidity conditions with sodium chloride as a contaminant. The pollution severity of outdoor insulators could be identified from the DWT STD-MRA (Standard Deviation Multi Resolution Analysis) Distortion Ratio pattern analysis of leakage current signals. Four parameters have been selected and used as the inputs for an Adaptive Neuro-Fuzzy Inference System(ANFIS) model, such as leakage current, distortion ratio, relative humidity and temperature. The model is appropriate to predict the equivalent salt deposit densities (ESDD) with an error difference of less than 0.0050 mg/cm(2) if the training data and the testing data are selected at the security stage. This reported result in the optimal prediction input parameters provide an invaluable tool for maintenance planning of a contamination flashover.

• 出版日期2011-4