Fractional-Order Observer-Based Current Sensor Fault Diagnosis for Lithium-Ion Battery Management System

The battery management system (BMS) is the cornerstone of the safe operation of electric vehicles (EVs), and its stable operation relies heavily on the accuracy of the battery current sensor. In this paper, a novel method for current sensor fault diagnosis in lithium-ion batteries is presented. First, a fractional-order (FO) battery model is proposed, and a hybrid particle swarm optimization algorithm with a cross-learning strategy is developed for model parameter identification. Second, a FO observer is designed for state of charge (SOC) estimation, and the stability of the observation error system is verified. Finally, by accurately estimating the change in SOC caused by a fault current and employing residual analysis, current sensor faults are identified. Experimental analysis, carried out under different operating conditions, reveals that the proposed method detects current sensor faults more accurately and quickly than the general open-circuit voltage method. Moreover, it can adapt to varying operating conditions and temperatures.