Buxton, A.P., 2002, The development of a computational methodology for the prediction of the sediment retention performance of invert traps in combined sewer systems, PhD thesis, Department of Civil and Structural Engineering, University of Sheffield.
Early types of sediment trapping structures were designed to retain coarse inorganic sediment transported within sewer systems, and thus prevent problems associated with their deposition. They were found to trap all types of material during dry weather flow, which could be re-entrained. An improved trapping technique was proposed to selectively trap inorganic grits.
The principle aim of this research was, therefore, to develop a computationally based methodology that can be used to assess the sediment retention performance of an invert trap, and therefore assess the selectivity of a trap design.
A laboratory study was carried out to obtain a set of sediment retention performance curves for an invert trap. The study examined three trap configurations with two types of artificial sediment over a range of discharges. The study showed that trap performance was sensitive to discharge and that the use of a smaller trap entrances improved the selectivity of the trap.
A 3-dimensional Computational Fluid Dynamics model, using the FLUENT software, was set-up and validated against laboratory in-trap flow field data. This data was obtained using the Particle Image Velocimetry visualisation technique. A Reynolds stress turbulence model using the Quadratic Pressure-Strain model was found to give the best predictions, successfully recreating the 3-D flow structures within the channel and the invert trap.
The particle tracking routine was implemented to predict the sediment retention performance of the range of tests applied within the laboratory. Parametric analysis was carried out to assess the sensitivity of the predictions to a range of parameters that were used to calibrate the methodology. After calibration, the study displayed an excellent correlation with laboratory data, resulting in a methodology that could potentially be applied to full-scale invert traps. During the course of the study, several issues involving the use of particle tracking were highlighted for further examination.