Abstract:

Lithium plating is one of the main contributors to internal short-circuit failures in energy storage lithium-ion batteries. The online monitoring of the lithium plating process is a key direction in research for ensuring battery safety. However, existing methods are limited to monitoring reversible lithium plating in engineering applications, and machine learning-based online monitoring algorithms have difficulty with model training due to the lack of lithium plating data from actual batteries. To address these problems, this work proposes a method for monitoring irreversible lithium plating using unsupervised domain adaptive transfer learning. The source and target domain data for the irreversible lithium plating monitoring model were generated through simulations using an electrochemical-thermal-aging model and low-temperature lithium plating aging experiments. Features related to lithium plating were extracted from the discharge curves, and an irreversible lithium plating monitoring model based on a multilayer perceptron was developed within an unsupervised unsupervised domain adaptive transfer learning framework. This allowed for the transfer of the irreversible lithium plating monitoring model from the source domain data to the target domain data. The results showed that on the simulated data, the algorithm achieved an accuracy rate above 99% in detecting the occurrence of irreversible lithium plating. On the experimental data, the qualitative judgements by the algorithm about the irreversible lithium plating were consistent with the actual conditions. This demonstrates that the proposed method can effectively monitor irreversible lithium plating in batteries and offers a new approach to the online monitoring of irreversible lithium plating in energy storage lithium-ion batteries.

Key words: lithium-ion batteries, lithium plating monitoring, pseudo two-dimensional model, multilayer perceptron, transfer learning