Estimating the DP Value of the Paper Insulation of Oil-Filled Power Transformers Using an ANFIS Algorithm

Abstract

The condition of the transformer insulation has an impact on the transformer’s performance during the operation. The aging of the oil-impregnated cellulose insulation and the associated loss of mechanical strength are the important factors that limit the life of expectancy of a transformer. To determine the condition of the oil-impregnated cellulose insulation, the Degree of Polymerization (DP) parameter is commonly used. An Adaptive Neuro-Fuzzy Inference System (ANFIS) has been developed to predict the DP Value by the chemical characteristics and dissolved gas parameters (acidity, interfacial tension, CO, CO2, breakdown voltage, and water content of the oil). This paper generates some algorithms which are based on the input space partitioning method to generate rules (grid partition or subtractive clustering) and data is normalized or not. The estimation result has been observed and evaluated to provide that the ANFIS algorithm is suitable to estimate insulation condition on field operating transformers.