The safety issue of the lithium-ion batteries is the key to their application and development. The management of lithium-ion batteries has been a hot topic of research for many years, which involves a number of scientific and engineering issues. This paper summarized the current research advances in lithium-ion battery management systems, covering. ••Typical architecture of the battery management system is presented.••Battery modeling and state estimation methods are reviewed.••Typical battery management strategies are presented and classified.••Future trends for each aspect are concluded and disclosed.Battery managementState estimationCharging strategiesFault diagnosisAC Alternating currentAI Artificial intelligenceBi-LSTM Bidirectional long short term memoryBMS Battery management systemBMTS Battery thermal management systemCC In electrochemical energy storage, the most mature solution is lithium-ion battery energy storage. The advantages of lithium-ion batteries are very obvious, such as high energy density and efficiency, fast response speed, etc,. With the reduction of manufacturing costs of the lithium-ion batteries, the demand for electrochemical energy storage is increasing,. Lithium-ion battery safety is one of the main reasons restricting the development of new energy vehicles and large-scale energy storage applications. In recent years, fires and spontaneous combustion incidents of the lithium-ion battery have occurred frequently, pushing the issue of energy storage risks into the limelight. The root cause is the abuse of lithium-ion batteries and the lack of effective monitoring and warning means. How to improve the safety and reliability of the battery system is the main task of the battery management system. Fig. 1 presents a typical architecture of the battery management system. This structure breaks through the traditional embedded hardware terminal resource limitation, based on the bidirectional real-time data communication network, and takes full advantage of the big data cloud platform information storage, computing capacity and other resources to upgrade the battery from traditional offline management to active online management. Through cloud-based online learning and digital twin model update, it overcomes the shortcomings of traditional BMS using fixed parameter models, thus realizing refined and pers.
