CFD modeling of the flow patterns in a water storage tank
Please login to view abstract download link
Water is essential for human life and an efficient management is critical to ensure its availability and quality [1]. Drinking water tanks regulate flow and pressure to meet demand and ensure minimum pressure, but climate change and prolonged droughts threaten supply and water quality [2]. Therefore, to identify internal flow patters is necessary to detect critical zones, particularly low-velocity regions to prevent algae growth, microbial proliferation, and other external factors that could compromise water quality [3]. Computational Fluid Dynamics (CFD) provides an alternative to experimental techniques for predicting flow patterns, residence times, and mixing improvements [4-5], but simulations must be adapted to each specific geometry. In this work, three dimensional, unsteady RANS simulations were performed using the RNG kɛ turbulence model with Enhanced Wall Treatment to capture the mean flow features of complex flow structures [3], especially in the high-velocity regions occurring inside the tank. Since the experiment is focused on the surface, the VOF approach enables a more direct comparison as it does not impose a totally flat water surface profile. This method gives a more accurate representation of how surface effects propagate into other regions. The governing equations were solved using a multiblock structured mesh of over 1.5 million hexahedral cells. The model was validated through laboratory experiments using floating balls to visualize surface flow. Ball motion was tracked using a Python neural network code analyzing video frames. The validated model identified stagnation zones and low-velocity regions, allowing detailed analysis of critical flow areas. The methodology can be applied to drinking water tanks to predict flow patterns and guide design improvements that enhance water homogenization. Extending this approach to real tanks will help to optimize design, improve mixing, and reduce low-velocity zones while providing a non-intrusive way to assess flow patterns.