SOC estimation based on online parameter identification and AUKF 2020-01-1183
The SOC plays an important role in vehicle energy management, power battery capacity utilization, battery charge and discharge protection. Battery model accuracy and noise variance will greatly affect the result of SOC estimation algorithm. In order to solve this problems, this paper builds Second-order equivalent circuit model, and applys the recursive least squares algorithm to identify the battery parameters online, and consequently the adaptive unscented Kalman method is proposed to estimation SOC. In order to verify the performance of the proposed algorithm, this paper uses experimental data of lithium battery to build a simulation model and test environment. The results show that compared with the current three algorithms, the proposed method in this paper has high estimation accuracy and minimum root mean square error.