Research on internal short-circuit fault diagnosis methods for lithium-ion batteries based on WOA-VMD and PSO-SVM

With the continuous consumption of traditional fossil fuels, people have gradually started to realize the importance of protecting the environment. Therefore, new clean energy sources like wind power have been gradually receiving considerable attention in recent years, along with the rapid development of new energy vehicles as replacements for traditional cars. Lithium-ion batteries have emerged as essential energy storage equipment for clean energy systems and power sources of new energy vehicles. However, these batteries are prone to thermal runaway failure during their usage and pose safety concerns. To ensure the safe operation of energy storage equipment and new energy vehicles, this paper proposes fault diagnosis methods using whale optimization algorithm optimized variational mode decomposition (WOA-VMD) and particle swarm optimization support vector machine (PSO-SVM) for diagnosing internal short-circuit fault voltage signals of lithium batteries, as they cause runaway heating. First, the internal short-circuit fault voltage signal in the lithium battery is decomposed using VMD to obtain a series of natural mode components. Two parameters in the VMD algorithm that significantly impact the decomposition results include the number of decomposition layers K and penalty factor α. To achieve the best decomposition effect, WOA was used to determine the VMD optimal decomposition level K and penalty factor α. Further, the optimal parameter combination is found to be K = 10 and α = 1997. Subsequently, this optimal parameter combination was introduced into VMD decomposition to decompose the internal short-circuit fault signal of the lithium-ion battery to obtain 10 modal components. Thus, the sample entropy of each of the 10 modal components was calculated and used as the eigenvector. Finally, these eigenvectors were inputted into the SVM model, and then the PSO-optimized SVM model was used for fault diagnosis, providing the diagnostic results. The final results showed that the diagnostic accuracy of the direct SVM model remained stable at 66.667%, while the diagnostic accuracy of the PSO-optimized SVM model was stable at 96.667%. Compared with the direct SVM model, the PSO-SVM diagnostic model effectively identified the internal short-circuit fault of the lithium-ion battery in the SVM model after feature selection by the particle swarm optimization algorithm, thereby proving its effectiveness.