A Monte Carlo simulation study on probability distribution of urban tunnel fire

Abstract

Urban tunnel traffic has its inherent charateristics, such as frequently periodical traffic congestion/blocking, tidal phenomenon, large traffic flow and population, complicated traffic confluences and cross-flowing intertwined sections. Unlike ordinary roads, the potential fire scenarios are quite different and the consequences of fire are extremely severe in urban tunnel. The study adopts theoretical calculation, formula derivation, traffic survey, random sampling simulation to explore the probability distribution of urban tunnel fire. MATLAB is used to realize the numerical simulation of a million fire scenarios under different working conditions. The results identify the major factors affecting two-lane tunnel vehicle ignition model and three-lane model respectively. The probability of six combustion scales ( $ 5\,{\rm MW} $ , $ 20\,{\rm MW} $ , $ 25\,{\rm MW} $ , $ 30\,{\rm MW} $ , $ 40\,{\rm MW} $ and $ 60\,{\rm MW} $ ) is higher than that of others, and with the increase of combustion scale, the probability decreasing exponentially. The ratio of truck to bus has a significant impact on the probability of large-scale fire. And truck fire and bus fire have a risk of igniting vehicles in adjacent lanes. The research results provide a basis for the design parameters of evacuation safety in practical engineering.

Keywords

• urban tunnel
• monte carlo method
• tunnel fire
• probability distribution