This paper proposes a co-estimation scheme of state of charge (SOC), state of health (SOH), and state of function (SOF) for lithium-ion batteries in electric vehicles. The co-estimation denotes that the SOC, SOH and SOF are estimated simultaneously in real-time application. The model-based SOC estimation is fulfilled by the Extended Kalman Filter (EKF). The battery parameters related with the battery SOH and SOF are online identified using the Recursive Least Square (RLS) algorithm with a forgetting factor. The capacity and the maximum available output power are then estimated based on the identified parameters. The online update of the capacity and correlated parameters help improve the accuracy of the state estimation but with limited increase in the computation load, by making good use of the correlations among the states. The co-estimation scheme is validated in a real Battery Management System (BMS) with good real-time performance and convincible estimation accuracy.

The online estimation of battery state-of-health is an ever significant issue for the intelligent energy management of the autonomous electric vehicles. Machine-learning based approaches are promising for online state-of-health estimation. This paper proposes a machine-learning based algorithm for the online state-of-health estimation of Li-ion battery. A predictive diagnosis model used in the algorithm is established based on support vector machine (SVM). The support vectors, which reflects the intrinsic characteristics of the Li-ion battery, are determined from the charging data of fresh cells. Furthermore, the coefficients of the SVMs for cells at different state-of-health are identified once the support vectors are determined. The algorithm functions by comparing partial charging curves with the stored SVMs. Similarity factor is defined after comparison to quantify the state-of-health of the data under evaluation. The operation of the algorithm only requires partial charging curves, e.g. 15min charging curves, making fast on-board diagnosis of battery state-of-health into reality. The partial charging curves can be intercepted from a wide range of voltage section, thereby relieving the pain that there is little chance that the driver charges the battery pack from a predefined state-of-charge. Train, validation, and test are conducted for two commercial Li-ion batteries with Li(NiCoMn)1/3O2 cathode and graphite anode, indicating that the algorithm can estimate the battery state-of-health with less than 2% error for 80% of all the cases, and less than 3% error for 95% of all the cases.