The Gaoligongshan Tunnel in Yunnan Province, China, serves as a representative example of a high ground-temperature and high-altitude tunnel. The tunnel’s construction through four heat-water conduction zones presented significant safety risks, which formed the primary motivation for this research. Correlation and regression analyses between on-site measured wet-bulb globe temperature (WBGT) values and environmental parameters, including air temperature, atmospheric pressure, wind speed, and relative humidity, reveal that air temperature, atmospheric pressure, and relative humidity exhibit significant correlations with WBGT. Based on Pearson correlation coefficients, the degree of correlation ranks in descending order as follows: air temperature relative humidity atmospheric pressure. A method utilising a Sparrow search algorithm (SSA)-BP neural network is proposed for the rapid and accurate prediction of WBGT values from conventional tunnel environmental parameters. This approach facilitates thermal environment assessment based on WBGT, supporting the development of a three-dimensional ‘3-D’ tunnel cooling system. Compared with predictions from a standardised regression equation, the SSA-BP neural network improves the R2 value between predicted and measured WBGT by 7.7%. Technical measures such as ventilation, cryocoolers, and ice cooling are demonstrated to reduce WBGT to ≈26°C, thereby effectively improving the tunnel thermal environment.
Luo et al. (Wed,) studied this question.