Combined sewer overflows (CSOs) represent a common feature in combined urban drainage systems and are used to discharge excess water to the environment during heavy storms. To better understand the performance of CSOs, the UK water industry has installed a large number of monitoring systems that provide data for these assets. This paper presents research into the prediction of the hydraulic performance of CSOs using artificial neural networks (ANN) as an alternative to hydraulic models. Previous work has explored using an ANN model for the prediction of chamber depth using time series for depth and rain gauge data. Rainfall intensity data that can be provided by rainfall radar devices can be used to improve on this approach. Results are presented using real data from a CSO for a catchment in the North of England, UK. An ANN model trained with the pseudo-inverse rule was shown to be capable of predicting CSO depth with less than 5% error for predictions more than 1 hour ahead for unseen data. Such predictive approaches are important to the future management of combined sewer systems.

The accurate characterization of flow from urban surfaces to sewer/stormwater systems is important for urban drainage design and flood modeling/risk identification. However, the geometrical complexity and large variety of drainage structures (linking elements) available makes model calibration and verification difficult. In this study an extensive comparison between experimentally measured and numerically modeled flow characteristics in the vicinity of ten different designs of manhole grate was performed under drainage flow in subcritical conditions. Using a 2‐D surface PIV (sPIV) system the work presents the first detailed characterization of velocity fields around these linking elements. In addition, it provides the first detailed verification of the ability of a 2‐D numerical model to describe both velocity fields and drainage flows. The overall comparison shows a close relationship between numerical and the experimental results with some higher inflows in the experimental results as a consequence of a localized transition from weir to orifice condition near the void areas of the grates. It was also noted that velocity differences decreased further from the manhole, due mainly to the more directional flow. Overall the work demonstrates the potential for further use of 2‐D numerical models to describe flow conditions at linking elements, either directly within modeling simulations or indirectly via the characterization of energy loss coefficients.

Urban drainage systems are frequently analysed using hydraulic modelling software packages such as InfoWorks CS or MIKE-Urban. The use of such modelling tools allows the evaluation of sewer capacity and the likelihood and impact of pluvial flood events. Models can also be used to plan major investments such as increasing storage capacity or the implementation of sustainable urban drainage systems. In spite of their widespread use, when applied to flooding the results of hydraulic models are rarely compared with field or laboratory (i.e. physical modelling) data. This is largely due to the time and expense required to collect reliable empirical data sets. This paper describes a laboratory facility which will enable an urban flood model to be verified and generic approaches to be built. Results are presented from the first phase of testing, which compares the sub-surface hydraulic performance of a physical scale model of a sewer network in Yorkshire, UK, with downscaled results from a calibrated 1D InfoWorks hydraulic model of the site. A variety of real rainfall events measured in the catchment over a period of 15 months (April 2008–June 2009) have been both hydraulically modelled and reproduced in the physical model. In most cases a comparison of flow hydrographs generated in both hydraulic and physical models shows good agreement in terms of velocities which pass through the system.