Predictive Maintenance of VRLA Batteries in UPS towards Reliable Data Centers July 2020 Conference: IFAC World Congress 2020At: Berlin, Germany Project: Artificial Intelligence for Cyber Physical Systems Authors: Jing-Xian Tang Tsinghua University Jin-Hong Du Carnegie Mellon University Lin Yiting Sun Yat-Sen University Qing-Shan Jia Tsinghua University Abstract: The reliability of data centers can be severely aected when battery failure occurs in the Uninterruptible Power Supply (UPS). Thus it has become a central issue for the industry to discover failure-impending batteries in UPS. In this paper, we consider this important problem and present a data-driven method for predictive battery maintenance. The major contributions are as follows.First, we develop a changepoint detection technique for ecient data labeling. Second, new features are designed to fully utilize the dataset. Third, we build a predictive classication model which can discriminate between healthy and failure-impending batteries. Our method has been built and evaluated on 209,912,615 records from Tencent data center involving nearly 300 batteries monitored over 2 years. The experiment on test set shows that our method is able to predict battery replacement with 98% accuracy and averagely 15 days in advance, which outperforms the previous maintenance policy by more than 8%.
VIEW FULL TEXT