Modeling and prediction for discharge lifetime of battery systems using hybrid evolutionary algorithms

A hybrid evolutionary modeling algorithm (HEMA) is proposed to build the di scharge lifetime models with multiple impact factors for battery systems as well as make predictions. The main idea of the HEMA is to embed a genetic algorithm (GA) into genetic programming (GP), where GP is employed to optim ize the structure of a model, while a GA is employed to optimize its parame ters. The experimental results on lithium-ion batteries show that the HEMA works effectively, automatically and quickly in modeling the discharge life time of battery systems. The algorithm has some advantages compared with mo st existing modeling methods and can be applied widely to solving the autom atic modeling problems in many fields. (C) 2001 Elsevier Science Ltd. All r ights reserved.