Investigations on unconfined large-scale methane explosion with the effects of scale and obstacles

The strong pressure wave in gas explosion accident could cause great casualties and property loss, which has become the main threat to safety of chemical process and further popularization of clean energy. In this research, unconfined explosion experiments under different methane concentrations at 2...

Full description

Saved in:
Bibliographic Details
Published inProcess safety and environmental protection Vol. 155; pp. 1 - 10
Main Authors Zhou, Yonghao, Li, Yanchao, Jiang, Haipeng, Zhang, Kai, Chen, Xiangfeng, Huang, Lei, Gao, Wei
Format Journal Article
LanguageEnglish
Published Rugby Elsevier B.V 01.11.2021
Elsevier Science Ltd
Subjects
Online AccessGet full text

Cover

Loading…
More Information
Summary:The strong pressure wave in gas explosion accident could cause great casualties and property loss, which has become the main threat to safety of chemical process and further popularization of clean energy. In this research, unconfined explosion experiments under different methane concentrations at 27 m3 scale were performed with internal and external obstacles. The results showed that the flame buoyant effect was more obvious for the rich-fuel flame due to the excess methane accumulation. The reflection of the pressure wave could cause higher overpressure near the ground and the external obstacle enhanced the overpressure in the neighborhood. For 27 m3 scale explosion, the critical flame radius for the flame acceleration was larger than 1 m3 scale, because of the longer travelling distance for the reflection wave from the bottom. In addition, the acceleration exponent rose to 1.5 due to the full development of the hydrodynamic instability. The flame radius for peak overpressure, RM, was found to be at a certain distance from the outer obstacle bar center, which was 1.8 for Nb = 6, 1.35 for Nb = 4 and 0.9 for Nb = 2. The predicted results of peak overpressure agreed well with the experimental results, which extended the prediction model to the larger scale.
ISSN:0957-5820
1744-3598
DOI:10.1016/j.psep.2021.09.004