Computational fluid dynamics characterization of a novel mixed cell raceway design

Authors: CHUN, CHAN WOO - Cornell University; VINCI, BRIAN - Freshwater Institute; TIMMONS, MICHAEL - Cornell University

Submitted to: Aquacultural Engineering
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 2/18/2018
Publication Date: 2/21/2018
Citation: Chun, C., Vinci, B., Timmons, M. 2018. Computational fluid dynamics characterization of a novel mixed cell raceway design. Aquacultural Engineering. 81:19-32. https://doi.org/10.1016/j.aquaeng.2018.02.002.
DOI: https://doi.org/10.1016/j.aquaeng.2018.02.002

Interpretive Summary: Fish culture tanks must address the biological and technical requirements of fish through their hydrodynamic design, providing optimal fish swimming speed, uniform water quality, and a self-cleaning environment. In the present study, computational fluid dynamics analysis was used to optimize design of a novel mixed cell raceway tank that used center drains in each cell plus linear flow down the length of the raceway using inlet and outlet weirs for the primary fraction of the total flow. A key finding was that when the mixed cell raceway was operated with a 15-minute hydraulic turnover time, at least 20% of the total flow must pass through the bottom-center drain of each cell to maintain effective vortex formation and rapid solids removal. Findings suggest that mixed cell raceway tank designs can be optimized to create hydraulic conditions that increase fish culture capacity while minimizing fixed and variable costs.

Technical Abstract: Computational fluid dynamics (CFD) analysis was performed on a new type of mixed cell raceway (MCR) that incorporates longitudinal plug flow using inlet and outlet weirs for the primary fraction of the total flow. As opposed to regular MCR wherein vortices are entirely characterized by the boundary conditions at inlet nozzles and outlet center drains in the center of each cell, the new MCR can develop a wider variety of fluid behaviors due to the additional boundary conditions at the inlet and outlet walls where the weirs are placed. In this study, we investigated how the primary longitudinal flow would affect vortex formations in the cells by designing three different MCR models and simulating three major cases for each model. Through this process, performances of two numerical CFD models (transition k-kl-' vs. k-e) were compared, along with two vortex quantification methods (Q-riterion vs. a proposed method). We found that the k-kl-' CFD model more accurately predicted vortex formation than the k-e model. The three MCR models differed by weir geometry or drain size only, in order to see their individual influence on cell vortex formation. Each case had its unique weir flow rate and center drain loading rate values, whose total rate would result in 15-minute hydraulic retention time (HRT) for the raceway. The percentage of center drain loading rate to total flow rate (R = 7.5%, 12.5%, and 20.1%.) was defined to establish a relationship between R and vortex strength or size. Simulations demonstrated that weir aspect ratio impacted cell vortex formation and strength. Unlike weir geometry effects, the drain size had non-significant impacts on fluid behavior other than the velocity very near the drains. While R did have positive correlations with vortex strength, vortex size, and self-cleaning performance, an R of 20.1% was enough for uninterrupted vortex formations. Too low of a center drain rate or R value can result in lack of any meaningful cell vortex formation which then obviates any self-cleaning action in an MCR. Our key finding through extensive computational analysis was that an R value of 20% was required in order to maintain effective vortex formation. Expressed more explicitly, this can be described as maintaining a center drain loading rate of 0.010094m3/s per cell (160 gpm), which correspond to unit loading rates of 16.3 lpm/m2 per cell (0.40 gpm/ft2 per cell).