Battery Cell SOH Deviation Analysis for Capacity Fault Detection

The patent transforms the diagnosis approach by changing from direct capacity measurement to SOH-based deviation analysis. It introduces multiple parameters including SOH values, moving average values, and deviation values to accurately identify abnormal battery cells. The diagnosis accuracy is improved by calculating SOH deviations from moving averages rather than using simple threshold comparisons.

The patent introduces moving average values as an intermediary between raw SOH measurements and final diagnosis results. This intermediary layer smooths out measurement fluctuations and provides a more stable baseline for detecting true abnormalities, thereby improving diagnosis reliability without requiring complex measurement equipment.

The patent performs preliminary calculations of moving average values during normal operation, preparing the baseline data needed for rapid abnormality detection. When abnormalities need to be detected, the system can immediately compare current SOH values against pre-calculated moving averages, enabling timely detection without requiring intensive real-time computation or continuous monitoring.

The patent implements selective monitoring by focusing computational resources on calculating deviations for battery cells that show potential abnormalities. Rather than continuously analyzing all battery cells with equal intensity, the system applies partial monitoring to normal cells and excessive (intensive) analysis only when deviation thresholds are approached, optimizing energy consumption while maintaining detection reliability.