Abstract:

Battery-management systems (BMS) must continuously update estimates of state-of-charge (SOC) in order to compute and calibrate estimates of state-of-health, state-of-energy, and state-of-power (state-offunction), and to prevent cell overcharge and undercharge conditions. There are many methods used to estimate SOC, with some having advantages over others. This paper reviews different SOC-estimation approaches for lithium-ion batteries and hopes to provide the reader with perspectives and insights based on experience working in the field. The physical significance of SOC was described, which can help distinguish between methods to estimate physical SOC versus engineering SOC. Different estimation approaches were discussed in some detail. The problem of defining a battery-pack SOC metric was presented and effcient methods to compute cell SOC for every cell in the pack were reviewed. Finally, some perspectives on the present state of the art and on needed future work in the area were presented.

Key words: lithium-ion-battery , battery-management systems (BMS) ,  state-of-charge (SOC) ,  review , modelbased estimation , data-based estimation