Press Room - CFD helps guide fish through the Bonneville dam

	CFD helps guide fish through the Bonneville dam Laurie Ebner, US Army Corps of Engineers, Portland District, USA Cindy Rakowski, Pacific Northwest National Laboratory, Richland, Washington, USA
	The Columbia-Snake River system, in the Pacific Northwest of the U.S., plays a vital role in the economic well-being of the region. Various types of numerical and physical model investigations and field efforts are being conducted by the US Army Corps of Engineers to re-establish fish runs. One of these studies addresses improvements to a fish guidance system and in this article, it is explained how CFD studies for the Bonneville Project are being used to support engineering and operational decisions. Multiple structural modifications have been made to Unit 15 of Bonneville’s 2nd Powerhouse (B2) that have improved the fish guidance efficiency. These structural improvements have been designed to increase the overall gatewell flow and maintain uniform velocities into the gatewell. These conditions are believed to increase the number of fish guided into the gatewell and the uniform velocity is believed to decrease the possibility of escape from the gatewell into the turbine. The majority of the design effort was conducted in a physical model with all flow going directly into the turbine intake. The plan was to make the same improvements at Unit17 although, given the large lateral flow components known to exist near the unit, it is unknown if the improvements will work as well as those of Unit 15. A STAR-CD model of the Bonneville Forebay was used to investigate the lateral flow component near the B2 powerhouse and into the gatewell slot (Figure 1).

	Complex flows near the powerhouse result from the excavated bathymetry in front of the turbine intakes and the Bonneville Project operations. The flow at B2 is most uniform when all of the B2 turbine units (Units 11-18) are operating. When the total powerhouse discharge requires that not all turbines be used, then the load is typically split between the end units (11,12,17, and 18). This split flow operation results in lateral flow across the non-operational center units (13-16), and partially across Units 12 and 17. Our approach was at 2 scales: a numerical flume model was used to test the implementations or parameterizations of intake modifications; then those parameterizations were implemented in a full forebay model and various operational and structural scenarios run. The numerical model was used to test the sensitivity of the flow near the powerhouse to changes in spill volume, and the spill volume with the largest lateral components at the powerhouse was then used in all of the remaining scenarios. These included combinations of removable structural features and operational alternatives. In total, 15 combinations of operational scenarios and structural alternatives were run. The results from all of the operational scenarios highlighted that project operations were the most significant cause of lateral flow in the intakes at B2. The various structural or operational configurations had little influence on the lateral flow inside the intake. In addition, the lateral flow tends to be eliminated by the time the flow gets to the gatewell slot. With the lateral flow being minimized if not eliminated at the gatewell slot, the same design prototype tested at Unit 15 was installed at Unit 17, and will be tested this year.