Prediction of Performance Degradation in Lithiumion Battery Energy Storage Systems Based on Machine Learning

As a key supporting technology, the performance degradation prediction of lithium battery energy storage system (ESS) has become a research hotspot. The goal of this article is to explore a prediction method of lithium battery ESS performance degradation based on machine learning (ML) to improve the operation and maintenance efficiency and reliability of the system. In this article, firstly, the mechanism of performance degradation of lithium battery is analyzed, and the key factors affecting battery life are selected as characteristics. Then, a performance degradation prediction model based on ML is constructed. The model can capture the long-term dependence in time series data and is suitable for dynamic prediction of lithium battery performance degradation. With the help of the actual operation data of the supporting energy storage project of the fourth wind farm in Honggeer, Wulanchabu wind power base, an empirical study is carried out. As a result, it is not difficult to find that the proposed prediction model has high prediction accuracy, and the prediction error is mostly controlled in a small range. Economic analysis shows that the maintenance decision and battery replacement decision based on predictive model can significantly reduce the operation and maintenance cost and improve the economic benefits of the system.