Research on Fire Warning System and Control Strategy of Energy Storage Power Station

Research on Fire Warning System and Control Strategy of Energy Storage Power Station

연구 분야: Infrastructure

학회: HP3C '23: Proceedings of the 2023 7th International Conference on High Performance Compilation, Computing and Communications

In recent years, fires in energy storage power stations occur frequently, causing immeasurable losses to people's lives and property. The existing fire warning system is not accurate in judging accidents and is prone to misjudgment. Based on the study of the mechanism and development process of the battery thermal runaway, this paper determines the fire characteristic parameters required for predicting the fire of the storage power station, and designs the fire warning system platform of the storage power station according to the characteristic parameters, realizing the real-time detection and uploading of various emergencies of the storage power station. In the process of research, the AdamW algorithm based back propagation neural network is proposed, which is improved on the basis of previous algorithms, and the simulation is verified in Python software. To reveal the experimental results can show that the improved algorithm is more effective, energy storage power station is more suitable for now.