Predicting the insulation status of High voltage rotating machines Using intelligent techniques

Abstract

High voltage rotating machines are widely used in the power newlineStations sugar mills steel mills cement mills etc Insulation is an essential newlinecomponent for proper operation of electrical machines without any fault and newlinethe condition of insulation of an electrical machine mainly determines its life newlineperiod The economic aspects of its periodic maintenance like the loss in newlineproduction due to breakdown and the investment cost due to frequent newlinereplacement with a new machine is huge The insulation condition of the newlinemachine can be assessed by measuring the various parameters like insulation newlineresistance polarization index leakage current capacitance dissipation factor newlinepartial discharge magnitude and surge voltage withstanding strength To newlineassess these parameters of the stator winding insulation used in high voltage newlinerotating machines a number of measurements have been made on actual newlinestator coils of machines For various test voltages the leakage current newlinecapacitance and dissipation factor are measured and correlated as a function newlineof test voltages Back Propagation Neural BPN Network algorithm with bipolar newlinesigmoidal activation function Fuzzy logic approach BPN with slope newlineparameter and Proportional Integral and Derivative PID controller concept newlineand Adaptive Neuro Fuzzy Inference System ANFIS newline newline