The Cloud-End Collaboration Battery Management System with Accurate State-of-Charge Estimation for Large-Scale Lithium-ion Battery System

The lithium-ion battery plays an increasingly im-portant role in energy storage and electric vehicles with the development of renewable energy. To ensure the safe operation of the lithium-ion battery system in application, the battery management system (BMS) is often equipped to monitor the battery cells in real-time. However, due to the inaccuracy and high cost, the traditional BMS is difficult to be adapted to the large-scale lithium-ion battery system, which will be increasingly common in the future. To overcome these two shortcomings, a cloud-end collaboration BMS framework is proposed in this study, which takes advantage of cloud computing and machine learning to monitor the numerous battery cells. In addition, based on this cloud-end collaboration BMS framework, the gated recurrent unit neural network and transfer learning algorithm are used to estimate the state-of-charge of different batteries. And the accuracy and flexibility of the method have been validated through experiments.

I. Introduction

By the end of 2021, the installed solar power and wind power have reached 305.987GW and 346.67GW, respectively. To handle the intermittent character of the above renewable power, the large-scale installation of energy storage system is indispensable [1]. Among the current energy storage tech-nologies, the lithium-ion battery which has the merits of high energy and power density, long cycle life, has been widely used in the stationary energy storage system and electric vehicle (EV) fields [2]. While, compared to other battery technologies, such as the lead-acid battery and the NIMH battery, the lithium-ion battery has a relatively higher safety risk, especially in overcharging or over-discharging situations. Hence, a battery management system (BMS) that can accurately monitor the battery state of charge (SOC) and conduct cell balancing, is essential to ensure the safe operation of the lithium-ion battery system in the application.