Flow Simulations Approach for Flocculation Tanks

Abstract

Flocculation in water treatment facilities plays a key role in the separation of colloidal inorganic and organic substances. Its optimization leads to a significant increase in its efficiency and savings of operational costs. However, it is currently based on trial-and-error experimental approaches. In this contribution, we focus on flow modeling in stirred flocculation tanks that would, after coupling with a calibrated model of particle aggregation, enable simulationbased flocculation optimization. Despite the abundance of literature on stirred tank modeling, there is no universal agreement on the methodology used to describe turbulence nor on the approach to the computational mesh creation. Consequently, there is no unified methodology for simulations and their validation. To address this, we present a best-practice methodology for economical, yet reliable flow simulations in the said device. This methodology includes the choice of the turbulence model, the approach to the design of a high quality mesh suitable for arbitrary geometries, and results evaluation. It is developed based on an extensive literature review, a multitude of flow simulations using several meshes of progressively higher quality and resolution, and various strategies to converge to steady-state flow conditions. The simulation quality indicators used here involve comparison with the experimental data on fluid velocity, stirrer power output, and flow rate through the impeller zone. Additionally, the resulting flow simulation models are compared using tracer transport simulations, hinting at their potential for coupling with particle aggregation models.