Lithium-Ion Battery Anomaly Detection Based on Big Data Platform of Large-Scale Energy Storage System

Battery anomaly detection is crucial for ensuring the operation safety of large-scale energy storage system (ESS). This paper presents a big data-based anomaly detection method for the voltage and direct current resistance (DCR) inconsistent faults. First, battery operation data is collected and logged in a big data platform of a large-scale energy storage system. Then, based on a simple Rint model, the DCR of every cell in the series-connected battery pack is calculated. Hence, the <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$3 \sigma$</tex>-criterion is applied to discriminate the faulty cells that exhibit anomaly in voltage and DCR. The proposed method is validated using one week of real-world operation data, and the detection results suggest that the proposed method is effective and reliable for ensuring the normal operation of ESS.