State of charge (SOC) is very important parameter for monitoring the battery charge and discharge operation and estimating the drive distance of electric vehicle. Especially, with the cycle number increasing, the precision estimation of SOC for battery management system is still not well resolved. Therefore, in this study, aim at accurate sampling of voltage, current and temperature signals based on LTC6803-3 chip, the paper proposed a support vector machine (SVM) optimized by particle swarm optimization (PSO) to improve SOC estimation accuracy. The results demonstrate that the proposed PSO-SVM model has good forecasting performance.

In a solar electric vehicle, the optimal sizing of hybrid power system can be considered as a multi-objective optimization problem. The two conflicting goals are to maximize the Loss of Peak Power Probability (LPPP) and minimize the system cost. And the former is related to the reliability of the system while the latter relates to whether production prototype so the two optimization objectives are important. An improved particle swarm algorithm was presented to optimal size the hybrid power system. Here the mutation operator of genetic algorithm was introduced and the acceleration factor could change with time. The optimization results show that: the improved particle swarm algorithm can well solve the hybrid power system for multi-objective optimization problems.