Abstract

A significant acceleration generated in the gorge terrain poses a serious threat to the operational safety of the transportation facilities downstream of the gorge. To determine the formation mechanism of the acceleration effect, this paper compares the results of wind tunnel tests with the improved delayed detached eddy simulation in terms of the spatial distribution, turbulence, and spectrum of the flow field on the gorge topography under crosswinds. The effect of the Reynolds number on the results is discussed in terms of the characteristic wind speed of flow field and the scaling ratio. Recurrent neural networks are used to attempt to get signals from unknown measurement points and to repair damaged signals. The results show that when the mountain spacing in the experimental and the computational fluid dynamics models is zero, the most dramatic acceleration reaches 1.28–1.4 times the incoming wind speed. In the wind tunnel tests, the peak of the power spectra density (1.065 × 10−2) at the downstream of the center of the gorge with a mountain spacing of zero is 3.37 and 14.77 times higher than the corresponding values (3.16 × 10−3 and 7.21 × 10−4) in the gorge topography with the mountain spacing of 0.01 and 0.02 m, respectively. The maximum difference of mean wind velocities in the leeward of the gorge is 6.4% when the Reynolds number ranges from 2.03 × 105 to 1.03 × 107. The results are expected to provide a reference for the design of windproof facilities in gorge terrains.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call