Distributed Finite-Time SoC Estimation for Lithium-Ion Battery Packs

Accurate estimation of state of charge (SoC) for lithium-ion battery packs is crucial for ensuring system safety and optimizing energy management. Most existing SoC estimation methods are centralized, relying on a central node to unify computation, which tend to lead to a heavy communication burden and poor reliability in the event of node or communication failure. Therefore, this article proposes a SoC estimation method for lithium-ion battery packs based on a distributed consensus algorithm. The method consists of two functional modules, the first part of which employs the Luenberger observer for SoC estimation of each battery cell. The second part utilizes the signed gradient descent flow algorithm to realize the distributed estimation of the SoC of battery cells, and ultimately obtains the SoC of the battery pack. We demonstrate that the proposed method converges in a finite time. Experiments verify the effectiveness and robustness of the proposed method. The results show that, compared with the centralized methods, the proposed approach exhibits superior fault tolerance, maintaining estimation errors within 1% even under communication failures.