Research On SOC （State Of Charge）Estimation Method Of LiFePO₄Battery

LiFeO4battery has significant advantages in safety, environmental protection, cycle life and sources of material, and is recognized as the best choice for storage battery and power battery. The accurate estimation of SOC (State of Charge) plays an important role in research, development and marketing of battery. This thesis conducted in-depth research of LiFeO4battery SOC estimation.The main contents of this thesis are follows:(1) An improved genetic algorithm was proposed for parameter optimization problem of regression modeling. Crossover probability and mutation probability could be adjusted with regression error and evolution time, so the improved genetic algorithm converged faster than standard genetic algorithm.(2) A battery testing platform was built. LiFeO4battery experiments were completed under different temperatures and different discharging rate.(3) Voltage, discharging rate, temperature, capacity and other characteristics of LiFeO4battery were analyzed based on experimental data.(4) Considering the characteristics of LiFeO4battery and comparing to mainstream SOC estimation method, the improved genetic algorithm is applied to the least squares support sector model (LSSVR) of SOC estimation.(5) The improved LSSVR SOC estimation model of LiFeO4battery was built, trained and tested with experimental data. The results were compared and analyzed. The results showed that the model obtained a satisfactory accuracy, and the rationality and feasibility of the improved LSSVR SOC estimation model were proved.(6) At last, the main contents of all sections of this thesis were reviewed. Some problems of the research process were discussed, as well as the future research directions.