Estimating Model on Torque for Alternating Current Electric Dynamometer Based on the Support Vector Machine Combined with the Fuzzy System

Abstract

In order to solve the problems such as high cost and complicated measurement system existing in the process of direct torque measurement for an alternating current electric dynamometer, the indirect measurement of support vector machine combined with the fuzzy system was adopted. Fuzzy rules were obtained by training the support vector machine and at the same time the genetic algorithm was brought to optimize parameters of the support vector machine and thus could obtain the most optimum fuzzy rule base. Moreover, torque of the alternating current electric dynamometer was estimated. The result shows that the precision of estimation with the support vector machine combined with the fuzzy system is high. So the direct torque measurement could be replaced by the indirect one and the support vector machine combined with the fuzzy system could be a method of cost reduction at present.