Fault diagnosis method for Li-ion battery based on voltage curve
Li-ion battery is widely used in electric vehicles,battery inconsistency faults and internal short-circuit faults are serious safety hazards for electric vehicles.The absolute mean difference method is used to amplify the voltage characteristics of faulty cells in a battery pack.Then,a nonlocal mean filtering algorithm is used to filter it to obtain a smooth fault voltage characteristic.Finally,the similarity distance between the eigenvalue curves is calculated by the dynamic time warping(DTW)algorithm to realize automatic localization of faulty batteries.Validation is carried out by using operating vehicle voltage data.The diagnostic method has a good detection effect on two types of faulty vehicles:one is the inconsistent faulty vehicle where the DTW distance of the faulty battery deviates from other batteries for a long period of time,the other is the faulty vehicle where the DTW distance of the faulty battery deviates from other batteries for a short period of time with an internal short-circuit.A comparison with the cosine similarity verifies that the proposed method is more accurate.
Li-ion batteryinconsistency faultinternal short-circuit faultfault diagnosisreal vehicle datadynamic time warping(DTW)algorithm
万方数据
维普
