Comprehensive Early Warning Strategies Based on Consistency Deviation of Thermal- Electrical Characteristics for Energy Storage Grid

Abstract

The development of distributed energy storage systems is an important guarantee for renewable energy into the grid. Considering life, cost and safety factors, lithium iron phosphate (LiFePO₄) batteries have always been the dominant energy storage battery in energy storage systems. However, due to the relatively flat open circuit voltage platform of the LiFePO₄ batteries, the estimation of battery state of charge (SOC) and safety early warning of battery is difficult. To solve these problems, a multiple timescale comprehensive early warning strategy based on the consistency deviation of the electrical and thermal characteristics is established. The equivalent circuit model (ECM) and the unscented Kalman filter (UKF) method are used to estimate battery SOC. The established comprehensive early warning strategy is verified through fault trigger experiments of different time scale through different equivalent resistance. Experiment results show that comprehensive early warning strategy can realize early warning of different time scale failures of LiFePO₄ batteries under energy storage conditions. For more dangerous serious failures that cause safety valve broken, the safety early warning can be carried out 15 minutes in advance. This research can provide a reference for ensuring the safe and reliable operation of energy storage system.