Ruoyin Zheng

Battery management systems depend on precise estimation of the battery’s state of charge. The accuracy and adaptability of estimation are limited by current approaches. An adaptive extended Kalman filter technique based on the moss growth optimization algorithm with wind intensity optimization is proposed to overcome these problems. The weight factor based on individual fitness and the number of iterations is combined to dynamically modify wind intensity to improve the moss growth optimization algorithm, which is then utilized to optimize the process noise covariance matrix in the Kalman filter. At the same time, the algorithm’s accuracy is enhanced by adaptively updating the measurement noise covariance matrix using the smoothing factor, which also prevents the filter from diverging and allows the algorithm to automatically update the noise parameters to better respond to changes in battery state and the influence of the external environment. Pulse current discharge experiments at various temperatures are used to confirm the efficacy of the improved algorithm proposed in this paper. The results demonstrate the superiority of the WOMGO-AEKF algorithm in terms of SOC estimate accuracy, since the root mean square error are much lower under various temperature circumstances than those of the conventional EKF algorithm.