Diagnosing inconsistencies in battery energy storage systems: A framework integrating electrical, thermal, and aging perspectives

Battery Energy Storage Systems (BESSs) are vital for grid stability and renewable energy integration. However, the inconsistencies across cells and packs can impair the performance and safety. Existing diagnostic approaches primarily address the voltage imbalance, offering only a partial view of the BESS reliability. This paper proposes a unified inconsistency diagnosis framework that simultaneously evaluates electrical, thermal, and aging inconsistencies. A low-rank subspace projection method enables reliable voltage inconsistency detection under low-resolution data and varying operational profiles. To capture thermal imbalance, the Thermal Consistency Coefficient (TCC) is introduced as a physics-based metric that quantifies pack-level thermal inconsistency using sparse sensor data. For aging assessment, an enhanced Least Squares (LS) method is developed to robustly estimate the health status of the battery pack under fluctuating loads. These perspectives are integrated through an entropy-weighted fusion scheme, yielding an objective and unified inconsistency score. Validation on a battery cluster within a 1.5MWh in-service BESS demonstrates the framework’s ability to identify inconsistent cells and quantify pack-level inconsistencies across voltage, thermal, and aging perspectives.