A Rapid State-of-Charge Estimation Method for Retired Batteries Without Initial State Constraints

Many aging characteristics used for the battery state of health (SoH) estimation are sensitive to the state of charge (SoC). Therefore, to effectively assess retired batteries, the study on the rapid SoC estimation of retired batteries is urgent, especially when the battery initial working condition is restricted as equilibrium state in most existing research. This paper analyzes the characteristics of step voltage response of batteries in different initial states. A fast offline SoC estimation model without initial state constraints is proposed, which reduces the testing time to 1 minute. The genetic algorithm is introduced for initial SoC and parameter optimization. The experimental results show that the proposed method can accurately estimate SoC in various initial conditions such as discharge polarization, charge polarization and alternating charge-discharge polarization, which has significant improvement in applicability.