Lithium Plating Diagnosis of Lithium-ion Batteries Based on Clustering with Multidimensional Features

As one of the primary energy storage devices today, lithium-ion batteries (LIBs) play a crucial role across various industries. However, lithium plating on the anode of LIBs significantly impacts their lifespan and safety, posing a substantial challenge to the further development of this technology. To address this issue, this paper proposes a lithium plating diagnostic method for LIBs based on multidimensional feature extraction and clustering analysis. By extracting high-precision battery model features and incremental capacity curve features, and developing a density-based clustering algorithm optimized by particle swarm optimization, a method for diagnosing lithium plating faults in LIBs is introduced. The accuracy of this method is then validated by capacity degradation rates and post-mortem analysis. The results indicate that lithium plating diagnosis based on multidimensional features is more accurate than diagnosis based on single-dimensional features, with a 10% improvement in lithium plating detection rate.