Modelling and Prediction of Reactive Power at Railway Stations Using Adaptive Neuro Fuzzy Inference Systems

Electricity has become an important concern in today’s society. This is due to the fact that electric grid now a greater number of non-linear components. The AC-powered locomotive one these aim this paper was model and predict reactive power produced by AC locomotive. presents study on modelling prediction locomotives. Reactive flow significant impact network voltage levels efficiency. research conducted using intelligent techniques—more precisely, adaptive neuro fuzzy inference system (ANFIS). Several approaches ANFIS structure were used research. Of these, we mention ANFIS-grid partition, subtractive clustering c-means (FCM) clustering. Thus; for predicting power, trained, then tested. For training ANFIS, experimental data obtained from measurements performed train supply sub-station used. taken over period time when locomotives far away station, close at respectively. currents voltages substation, respectively active, reactive, distorted powers, measured acquisition board. With with performed, variation made. analysed results comparing between several types architectures. values RMSE, RMS compared structures ANFIS.