A new defuzzification method for fuzzy control of power converters

In this paper, a new defuzzification method is proposed which can provide improved performance in fuzzy control for DC-DC converters. A comparative study of different defuzzification methods adopted in fuzzy logic control (FLC), such as center of area (COA), center of sums (COS), height method (HM), middle of maxims (MOM), center of largest area (COLA), and first of maxims (FM), for application to DC-DC buck-converters is presented. The distinction among the characteristics which lead to varying performance is outlined. A new method called height weighted second maxims (HWSM) is proposed and its performance is assessed. The paper also presents simulation results of the performance of the closed-loop converters from the standpoint of start-up transient, bad regulation and line regulation. The simulations show that COA, COS, and HM defuzzification methods have better dynamic performance and less steady state error. The new HWSM defuzzification method provides further improvement