Substation Fire Monitoring Based on Data Fusion
Xinhai Li, Yongting Zhang, Qifeng Luo · 2024
This paper proposes a substation fire detection method based on data fusion technology to address the issues of high false alarm and false alarm rates in traditional substation fire alarm systems, as well as the inability to adopt different levels of fire alarm and fire protection measures according to the importance of different areas within the station. By developing intelligent fire detection algorithms, utilizing multi-sensor information fusion technology, and introducing D-S evidence theory for fusion reasoning, the goal of significantly improving the accuracy of intelligent fire detection has been successfully achieved. At the same time, a recognition framework was constructed and similarity functions were extracted to determine the temperature threshold during a fire, thus completing the overall monitoring research. The experimental data shows that using the D-S fusion algorithm for fire detection can quickly converge the detection time, improve the accuracy probability of detection, and enable the entire system to achieve intelligent monitoring tasks for substation fire protection.