Wide temperature multi-condition SOC estimation for Li-ion batteries based on F-FOMIAEKF

Abstract The state of charge (SOC) of the batteries is critical to obtain the operating parameters of the battery system and optimize the control. The mainstream research on SOC estimation is based on the integer order circuit model, and the fractional order impedance model is rarely studied at present. Combined with the shortcomings of the extended Kalman filter (EKF), in this paper, a new adaptive EKF algorithm with a forgetting factor is proposed based on the fractional 2-order model (F-FOMIAEKF), which can accurately estimate the SOC of lithium-ion batteries under wide temperature and multiple operating conditions. In the SOC estimation algorithm, because the new measurement data contributes to enhancing parameter estimation accuracy, the outdated historical data may accumulate errors, resulting in data saturation, so that the estimated value deviates greatly from the actual value. To address this notable issue, this paper integrates the multi-innovation theory within the EKF framework and introduces a forgetting factor to reduce the weight of historical data in the overall data. Simultaneously, the process noise and measurement noise are adaptively adjusted, resulting in the proposed F-FOMIAEKF. The results indicate that the method proposed in this paper can get rid of the limitation that the order of integer order calculus is an integer and can effectively improve the drawback of the low utilization rate of historical data in EKF. Meanwhile, it can accurately reflect the internal state of the power batteries and accurately estimate SOC under wide temperatures and multiple operating conditions.