Multi-objective optimization of thermal performance in battery system using genetic and particle swarm algorithm combined with fuzzy logics

A novel technique for multi-objective optimization of thermal management in battery system using hybrid Genetic algorithm and Fuzzy logic is developed. Secondly, Particle Swarm Optimization algorithm combined with Fuzzy logic is also proposed for the same. The combined algorithms and fitness function for fitness evaluation is written in-house C code. For the thermal performance fitness evaluation, realistic conjugate heat transfer condition at the battery and coolant interface is adopted. The objective functions are average Nusselt number, friction coefficient, and maximum temperature. Maximizing one causes proportional increase in another, hence to achieve a moderate condition of better Nusselt number with low pumping power cost and temperature within allowable limits, these algorithms assist in optimization. Five different independent operating parameters are selected for the Multi-objective optimization and brief results are presented. The Fuzzy logic membership functions adopted can be easily modified/selected by the user to suite the battery thermal problem at hand and to assign weight to each fitness function. The fitness function obtained using the proposed multi-objective optimization technique are closer and indicate safe temperature of battery with enhanced Nusselt number and minimum friction coefficient. The maximum multi-objective fitness obtained after normalization is 0.9.