Securing BESS Modules: Anomaly Detection in Cell Voltages Through Statistical Analysis

Ensuring safety in Battery Energy Storage Systems (BESS) is paramount to prevent failures and enhance reliability. This paper introduces a novel algorithm for detecting abnormal behaviour by analyzing the voltage distribution of cells within a module. The method uses z-scores to calculate the relative percentage of anomalies in each cell and standardized moments to capture voltage deviations across the entire module. From these metrics, an anomaly score is derived for each cell. The algorithm is validated using data from a 4-cell in-series (4S) module model, demonstrating its effectiveness. Additionally, a sensitivity analysis of the configurable parameters highlights their impact on the anomaly score. This innovative algorithm is crucial for BESS safety, as it ensures early detection of potential issues, thereby preventing failures and enhancing the reliability and longevity of the system.