A Lumped Disturbance Compensation Scheme for Unbiased State-of-Charge Estimation of Lithium-ion Batteries

Accurate state of charge (SOC) estimation of lithium-ion batteries is vital for reliable and safe operations. However, modeling uncertainty in the equivalent circuit model and sensor noise can deteriorate the estimation accuracy. Conventional compensation schemes have declined performance under time-varying disturbances and when the initial SOC is unknown. Aiming at this problem, this article aligns modeling uncertainty and voltage sensor bias as a lumped disturbance to be estimated. Lumped disturbance characteristics and SOC error analysis are derived analytically. A novel augmented modeling structure is proposed, incorporating disturbance dynamics in the form of higher order derivatives. When combined with a closed-loop recursive estimator featuring output error injection, the lumped disturbance can be effectively estimated for compensating the asymptotic bias of SOC. In addition, an event-triggered switching technique is developed to enable the lumped disturbance compensation scheme to be applied without knowing the initial SOC. Experimental results show that the proposed scheme effectively mitigates the lumped disturbance-introduced SOC error, achieving unbiased SOC estimation.