Multi-fault diagnosis strategy based on a non-redundant interleaved measurement circuit and improved fuzzy entropy for the battery system

Accurately diagnosing and distinguishing the multiple faults of the battery system in electric vehicles (EVs) is critical for ensuring the safety of drivers and vehicles. However, using conventional diagnosis methods may lead to misdiagnosis when distinguishing faults. Therefore, this paper proposes a multi-fault diagnosis strategy based on a non-redundant interleaved measurement circuit (NRIMC) and improved fuzzy entropy (IFuzzyEn) for the battery system. The strategy first uses the NRIMC to generate a specific voltage sequence for the multi-fault diagnosis and then employs IFuzzyEn to calculate the anomalies of the voltage sequence. Finally, the fault location and type are determined based on the number of abnormal voltage sequences, while the degrees of faults are judged based on the magnitude of the abnormal values. The study involves battery pack level (BPL), battery module level (BML), and battery cell level (BCL). The multiple faults considered include sensor faults, connection faults at different degrees, short-circuit faults at different degrees, and faults under different discharge rates. The efficiency and accuracy of the proposed method are verified through theoretical and experimental analysis. Overall, this study contributes to developing a reliable multi-fault diagnosis strategy for EVs, which has significant implications for ensuring their safety and reliability.