Urban drainage networks are generally designed to operate in a free-surface flow condition. However, as a consequence of heavy rainfall events or network malfunctions, the filling of sewers (pressurisation) and network overflow may occur. Several modelling software products are commonly used to simulate floods in drainage networks, and their results are usually thought to be reliable and robust. However, no specific studies have been carried out on the behaviour of these modelling products during the pressurisation transition. Mathematical models often use the Preissmann slot concept to handle pressurisation. In this paper, on the basis of laboratory pipe tests, the reliability of such a scheme is studied by means of a popular and open-source software product: SWMM (Storm Water Management Model). Many numerical tests were carried out with SWMM, varying the spatial and time steps and the Preissmann slot width, in order to examine the performance of the modelling software over intervals of these parameters even wider than what is usual in practical applications. The comparison between simulated and experimental surges allows one to draw interesting conclusions regarding the effectiveness of software products analogous to SWMM in simulating pressurisation, as well as the choice of the parameters themselves.
Keywords: numerical modelling, pipe pressurisation, unsteady flow, urban drainage