.png){kind=icon}
.png){kind=icon}
Return
An intelligent indoor fire localization system combining dynamic clustering algorithm and particle swarm optimization algorithm
.png){kind=icon}
PRE
AI
DOI：10.1016/j.jobe.2024.111180.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
With the development of urbanization, indoor fire rescue has become an important issue. In fire rescue, the fire traceability technology is critical. However, due to the diversity and hidden nature of indoor fires, there are few reliable fire localization methods. This study develops an indoor fire localization system based on Unity engine. Real-time data transmission technology as well as artificial intelligence algorithms are utilized. Utilizing only a few sensors, the system can achieve the functions of fire localization and temperature field prediction. This paper proposes a fire localization algorithm that can be effectively applied to this system: a multi-particle swarm optimization algorithm based on the concept of dynamic clustering. The algorithm overcomes the defects of the particle swarm optimization algorithm, and can eliminate outliers and error values, which improves the reliability and stability of fire localization. The effectiveness of the fire localization algorithm and the intelligent fire localization system is verified through two sets of model-scale experiments and numerical simulation. The results show that the system can effectively provide visualization of indoor fire localization, which can be used to support rescue and evacuation for fire safety.
Keywords:
Indoor fire
Particle swarm optimization
Dynamic clustering
Fire source location
Journal
IF:
7.4
Papers:
1.6W
Citations:
6.6W