Towards co-estimation of lithium-ion battery state of charge and state of temperature using a thermal-coupled extended single-particle model

Abstract

The state of charge (SOC) and state of temperature (SOT) are two crucial battery states that are often individually estimated in lithium-ion batteries (LIBs). The cross-interferences between the battery SOC and SOT are not extensively considered, which may pose significant challenges for the simultaneous estimation of these two states. To this end, a composite SOC and SOT co-estimation framework is proposed by employing a thermal-coupled extended single-particle model (TESPM) and the square-root adaptive unscented Kalman filtering algorithm (SR-AUKF). First, a battery TESPM is developed, and a multi-objective stepwise parameter identification scheme is presented to parameterize the LIBs. Then, the experimental validation results indicate that the maximum voltage root mean square error (RMSE) of the proposed TESPM is 42.42 mV and the maximum SOT RMSE is 0.36K. Next, following the reduced-order TESPM and its governing state-space equations, the co-estimation framework of the battery SOC and SOT is proposed based on the SR-AUKF. In which, the square-root filtering is merged with adaptive unscented Kalman filtering to prevent the divergence of the filtering results. Lastly, extensive simulations and test investigations are conducted to confirm the effectiveness of the proposed co-estimation framework across a wide temperature range and under various operating conditions.