一种基于神经网络的pack级储能消防方法及系统
By using a neural network-based approach, representative battery packs were selected for monitoring. By utilizing a safety prediction neural network and a physical information model, the problems of high computing power consumption and lack of prediction in existing technologies for real-time monitoring were solved. This enabled early warning and prevention of fires, improving the efficiency and safety of the fire protection system.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- BEIJING FUSAIER SAFETY FIRE-FIGHTING EQUIP CO LTD
- Filing Date
- 2024-03-14
- Publication Date
- 2026-07-17
AI Technical Summary
In existing technologies, deep neural networks consume a lot of computing power to monitor each battery pack in real time, and current fire protection systems can only respond after a fire occurs, lacking the ability to predict and deploy targeted measures in advance.
A neural network-based approach is adopted to monitor representative battery packs by using temperature, voltage, and current information. A trained safety prediction neural network is used for prediction, combined with a physical information model to reduce computing power consumption and provide alarms and prevention before fires occur.
It enables early prediction and alarm before a fire, reduces wasted computing power, improves the prediction efficiency and safety of the fire protection system, and lowers hardware requirements.
Smart Images
Figure CN118403316B_ABST