Local loss model of dividing flow in a bifurcate tunnel with a small angle

• Published in: Journal of Zhejiang University. A. Science 2019, Vol.20 (1), p.21-35
• Main Authors: Zhang, Xin, Zhang, Tian-hang, Hou, Yun-ge, Zhu, Kai, Huang, Zhi-yi, Wu, Ke
• Format: Article
• Language: English
• Subjects: Air flow; Bifurcation theory; Coefficients; Computer simulation; Deflection; Diversion; Division; Downstream; Flow deflection; Flow separation; Flow velocity; Mathematical models; Physics & Engineering; Separation; Uniform flow; Velocity gradient
• ISSN: 1673-565X; 1862-1775
• DOI: 10.1631/jzus.A1800298

To provide a theoretical basis for the flow diversion control of a bifurcate tunnel, the flow separation characteristics and local loss model at the tunnel bifurcation were analyzed by combining numerical simulation and theoretical derivation. The results showed that the sudden change of boundaries interrupts uniform flow when air flows through a tunnel bifurcation, causing changes in flow velocity and direction. When the diversion ratio β is small, the flow is separated on the downstream mainline tunnel sidewall close to the bifurcation point and the ramp sidewall away from bifurcation point; when β is large, the flow is separated on the downstream mainline sidewall away from bifurcation point and the ramp sidewall close to bifurcation point. The local loss on flow division is caused mainly by velocity gradient changes and flow deflection and separation. When the air flux ratio q of the downstream mainline tunnel to that of the ramp is equal to their cross-sectional area ratio ϕ , local loss coefficients are at their minimum; when q > ϕ , the loss coefficients decrease with the increase of β , but the loss coefficient for the ramp increases as the bifurcation angle rises. When q < ϕ , the loss coefficients increase with the increase of β , but the loss coefficient for the ramp declines as the bifurcation angle rises. Finally, a theoretical formula to predict the dividing flow local loss coefficient of a bifurcate tunnel is established based on the airflow deflection angle assumption. The proposed model has a higher precision in prediction than other formulas.