NExperimental and numerical model studies on flash flood inundation processes over a typical urban street

2021 
Abstract Accurate prediction of flood inundation processes in urban areas is challenging, due to the complexity of street layouts and the variety of infrastructures. In this study, based on a laboratory model of urban flooding with a sewer system underneath, a series of laboratory experiments were conducted to investigate the influences of different street layouts and infrastructures on flood inundation processes. Key hydrographs of water depth and flow velocity were recorded at several measuring points to provide comprehensive information about the hydrodynamic characteristics of urban flooding. Furthermore, a 2D shallow water equation numerical model, based on the finite volume method, was also utilized to replicate the experimental scenarios considered. An analysis of the mesh resolution and discharge capacity formulae for street inlets was also performed through a series of numerical tests. The following conclusions are drawn from this study: (i) the sewer system has a strong influence on the flood inundation processes in terms of reducing both the surface water depth and flood wave velocity, as compared with street layouts and other infrastructures; (ii) the results from the numerical simulations agree well with the experimental findings, with the NSE values being greater than 0.7 and the RMSE values less than 3×10−2; (iii) a further increment in the mesh resolution may benefit slightly the model predictions, but at the expense of an increasing computational cost; and (iv) of all the inlet discharge capacity formulae used in this study, the weir and orifice formulae considering the influence of rain boxes were the most appropriate for representing the geometric features of street inlets and showed the best performance in calculating the flow drainage between the surface runoff and the underground sewer system.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    49
    References
    3
    Citations
    NaN
    KQI
    []