Journal
INTERNATIONAL JOURNAL OF THERMAL SCIENCES
Volume 171, Issue -, Pages -Publisher
ELSEVIER FRANCE-EDITIONS SCIENTIFIQUES MEDICALES ELSEVIER
DOI: 10.1016/j.ijthermalsci.2021.107247
Keywords
Underground pipe gallery; Artificial intelligence algorithm; Ant colony optimization; Fire source location; 3D space; Temperature field
Categories
Funding
- National Program on Key R&D Project of China [2020YFB2103503]
- National Natural Science Foundation of China [52008104]
- Program of Chang Jiang Scholars of Ministry of Education
- National Science Found for Distinguished Young Scholars of China [51625803]
Ask authors/readers for more resources
An artificial intelligence algorithm using temperature data was developed to detect fire source in underground pipe galleries, achieving satisfactory results. With the requirement of only a few temperature sensors, the algorithm shows potential for wide application.
Since there exists few effective fire detection technologies for underground pipe gallery application, a temperature data-driven bio-inspired artificial intelligence algorithm is developed to detect fire source in 3D space of the underground pipe gallery, in which a simple physical model is used. In the developed algorithm, Ant colony optimization (ACO) is the first time to be used to determine tunnel fire source, and the new and special pheromone evaporation method and heuristic factor are developed for fitting the concerned problem here. Three fire experiments are used to support the ability of the algorithm. Satisfactory results can always be obtained, which shows that the developed algorithm can be used to estimate the tunnel fire source as well as temperature prediction. In addition, since only temperature data at several sensors is necessary in the developed algorithm, it has a very wide popularization and engineering application prospects due to its advantages of the global optimal ability and computational efficiency as well as the low economic cost.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available