Fault Diagnosis of Series Batteries based on GWO-SVM

Nowadays, the fault diagnosis of battery internal resistance and contact resistance in EV series batteries is based on experience and has no clear diagnostic criteria. To solve this problem, the grey wolf swarm optimization support vector machine (GWO-SVM) is proposed. Acquiring the data in the published papers about the changes of resistance under the internal resistance and contact resistance fault. The sample data of normal condition and battery internal resistance fault, contact resistance fault, battery internal resistance and contact resistance mixed fault are simulated in Simscape and imported into GWO-SVM and other artificial intelligence diagnosis models. The results show that the proposed GWO-SVM has higher diagnostic accuracy than other diagnostic models in fault diagnosis of tandem battery packs, maintaining a higher accuracy rate in the case of mixed faults. This method is capable of distinguishing various types of faults effectively.