Experimental and Numerical Study on the Two-Dimensional Longitudinal Temperature Rise Behavior of Fire Smoke in the Shenzhen–Zhongshan Ultra-Wide Cross-Section Undersea Tunnel

The Shenzhen–Zhongshan Link is a key cross-sea corridor in the Guangdong–Hong Kong–Macao Greater Bay Area. As a representative ultra-wide cross-section undersea tunnel, it exhibits smoke spread behaviors that differ fundamentally from those of traditional road tunnels. In particular, the radial flow region of fire smoke is more pronounced, resulting in substantial lateral variations in smoke dynamics parameters. These characteristics render classical one-dimensional ceiling jet temperature rise theories insufficient for capturing the multidimensional thermal behavior in such geometries. In this study, the immersed-tunnel section of the Shenzhen–Zhongshan Link was investigated through a combination of full-scale fire experiments and Fire Dynamics Simulator (FDS) simulations. The longitudinal attenuation and lateral distribution characteristics of hot smoke temperature rise during spread in an ultra-wide tunnel were systematically obtained. Based on a simplified one-dimensional ceiling jet concept, differences in hot smoke diffusion distance were employed to characterize the lateral temperature rise ratio at any longitudinal location, from which a lateral distribution model was developed. The classical one-dimensional average temperature rise decay model was further reformulated to derive a modified longitudinal decay model applicable to the tunnel centerline of ultra-wide cross-sections. By integrating these characteristic models, a two-dimensional longitudinal prediction framework for hot smoke temperature rise in ultra-wide tunnels was established. Validation against full-scale fire experiments demonstrates that the proposed model can predict the two-dimensional thermal field with an accuracy within 25%. The findings of this study provide a theoretical basis for fire scenario reconstruction in the Shenzhen–Zhongshan undersea tunnel and offer a technical foundation for optimizing emergency ventilation strategies during fire incidents.