基于改进相关向量机的锂电池寿命预测方法

Lithium batteries have the advantages of light weight and safety, long cycle life, and good safety performance. As a widely-used energy storage power supply, lithium battery health management and life prediction are hot topics both at home and abroad. Lithium battery life assessment methods and prediction models were established. Battery decay models were established based on experimental historical data to evaluate the working status of the entire battery, and the equipment was maintained and replaced in time to ensure stable battery operation. In this paper, the kernel function of the relevance vector machine (RVM) was mainly improved, the performance of the relevance vector machine was optimized, the lithium battery life prediction bias was reduced, and the prediction accuracy was improved.