Volume 17: Technology Innovation to Accelerate Energy Transitions

Reinforced Temperature Prognosis of Energy Storage System Based on Two-Node Electrothermal Model and Integrated Long and Short-term Memory Network Marui Li, Chaoyu Dong*, Qian Xiao, Xiaodan Yu, Zhe Wang, Hongjie Jia

https://doi.org/10.46855/energy-proceedings-8639

Abstract

As an important part of the energy system, energy storage system, especially with the increasing popularity of renewable energy, has become more and more important. Subsequently, the problems of supervision and control of the energy storage system have become increasingly prominent. Temperature regulation is an important part. Many methods have been proposed to predict the temperature of the battery energy storage system. At present, it is mainly divided into the method based on electrothermal model and the method based on data-driven. In this paper, firstly, a two-node electrothermal model is established. Then an integrated network with dual inputs and dual long and short-term memory networks is established. Finally, the adaptive boosting algorithm is used to modify the prediction results of the surface temperature of the battery energy storage system. The experimental results show that the proposed coupling model is effective and progressive.

Keywords energy storage system, electrothermal model, long and short memory system, adaptive boosting

Copyright ©
Energy Proceedings