Application of MMAE to the Fault Detection of Lithium-Ion Battery

With the advantage of high energy density, long cycle life and environmental friendliness, Lithium-ion battery has become a promising power source for hybrid and electric vehicles, which are liable to two kinds of failure, overcharge and overdischarge. Because of the capability of detecting multiple faults, Multiple Model Adaptive Estimation (MMAE) method was applied to a model-based fault detection of a lithium-ion battery with a two-order linear electrical model. Parameters that represent normal-mode and faulty-mode of the battery were obtained by a series of experiments, and three Kalman filters were designed thereafter. Finally, simulation verified the performance of the MMAE algorithm on fault detection of these two kinds of fault and it is shown that this technique is able to discern these faults rapidly and accurately.