Stormwater treatment: Examples of computational fluid dynamics modeling

none7siControl of rainfall-runoff particulate matter (PM) and PM-bound chemical loads is challenging; in part due to the wide gradation of PM complex geometries of many unit operations and variable flow rates. Such challenges and the expense associated with resolving such challenges have led to the relatively common examination of a spectrum of unit operations and processes. This study applies the principles of computational fluid dynamics (CFD) to predict the particle and pollutant clarification behavior of these systems subject to dilute multiphase flows, typical of rainfall-runoff, within computationally reasonable limits, to a scientifically acceptable degree of accuracy. The Navier-Stokes (NS) system of nonlinear partial differential equations for multiphase hydrodynamics and separation of entrained particles are solved numerically over the unit operation control volume with the boundary and initial conditions defined and then solved numerically until the desired convergence criteria are met. Flow rates examined are scaled based on sizing of common unit operations such as hydrodynamic separators (HS), wet basins, or filters, and are examined from 1 to 100 percent of the system maximum hydraulic operating flow rate. A standard turbulence model is used to resolve flow, and a discrete phase model (DPM) is utilized to examine the particle clarification response. CFD results closely follow physical model results across the entire range of flow rates. Post-processing the CFD predictions provides an in-depth insight into the mechanistic behavior of unit operations by means of three dimensional (3-D) hydraulic profiles and particle trajectories. Results demonstrate the role of scour in the rapid degradation of unit operations that are not maintained. Comparisons are provided between measured and CFD modeled results and a mass balance error is identified. CFD is arguably the most powerful tool available for our profession since continuous simulation modeling.noneYing G.; Sansalone J.; Pathapati S.; Garofalo G.; Maglionico M.; Bolognesi A.; Artina S.Ying G.; Sansalone J.; Pathapati S.; Garofalo G.; Maglionico M.; Bolognesi A.; Artina S