Fuzzy logic approach for the prediction of the lifetime of epoxy insulators

Epoxy resins are insulators, which are employed as adhesives, coatings, sealants, moldings and potting compounds to provide void-free insulation for components in several electrical and electronic applications. Prediction methodology for the lifetimes of epoxy resins insulators under service conditions is of great importance and requires reliable laboratory work and experimental techniques. The objective of this work is to apply fuzzy logic algorithm for the prediction of lifetime of epoxy insulators. The prediction is achieved by measuring the surface leakage current under accelerated aging in laboratory. The base case of the epoxy insulators in this study is no filler and dry condition. Fillers are then added with different concentration rates to the insulators and the effect on lifetime is examined under various weather conditions and compared to the base case. The obtained simulation results using two fuzzy logic membership functions are compared with the experimental measured data The impact of the characteristics of modeling membership function (trapezoidal or triangular) and the number of functions partitions on the error rate is illustrated using Matlab fuzzy logic Toolbox.