Genetic optimization of a fuzzy system for charging batteries

A large variety of nickel-cadmium (Ni-Cd) batteries have been developed to meet a wide range of user needs, ranging from low-current-level uses like emergency power sources for semiconductor memories to very high-power applications such as motor-operated cordless drills. This paper presents a genetic algorithm approach to optimize a fuzzy rule-based system for charging such high-power Ni-Cd batteries. For the optimization of the fuzzy system, a special objective function is developed which is based on the entropy of a fuzzy system. The resulting fuzzy system is able to charge high-power Ni-Cd batteries in about 10 min with a current of 6 A.