大数据驱动的动力电池健康状态估计方法综述

State of health estimation of power batteries is one of the key algorithms of the battery management systems, which is of great significance for improving power battery energy utilization efficiency, reducing thermal runaway risk, as well as power battery maintenance and residual value evaluation. Comparative analysis has been done on experimental-based, model-based and data-driven methods, and data-driven methods are elaborated from three aspects: dataset construction, health indicators extraction, model establishment. The big data collection methods and data preprocessing methods are summarized. The health indicators extraction methods are compared by their pros and cons and applicable scenarios. The basic principles of different health state estimation models are discussed. The conclusion that model fusion is the direction of future technology development is proposed. Finally, facing the future application scenarios of big data in electric vehicles, the current issue and prospective are depicted.