Session B2: Improvement of Computational Tools for Biological Performance Assessment of Hydroelectric Turbines

Abstract: The technical evaluation of fish passage conditions is becoming an integral part of projects to install new or replace existing hydroelectric turbine units. In the design phase, these evaluations depend on laboratory and computational modeling techniques that simulate prototype conditions. For instance, laboratory experiments with beads have yielded observational data on collision occurrence and frequency in turbine physical models. On the computational model side, simulations have been used to assess hydraulic conditions hazardous to fish passage such as rapid decompression, extreme shear flow, and likelihood of collision on turbine runner blades. The present work expands on the previous modeling studies of biological performance assessment at projects in the Columbia River Basin, in the U.S. Pacific Northwest. The preceding studies have laid the groundwork for the use of computational fluid dynamics (CFD) in biological assessment of turbine passage. This presentation describes continued work by Pacific Northwest National Laboratory (funded by US Department of Energy) to improve its biological performance assessment (BioPA) software and to test the model against field observations. Specifically, two model enhancements will be discussed: (1) the generation of multiple passage pathways through the turbine to account for turbulent dispersion in BioPA, and (2) the implementation of overset (moving) mesh simulations to better account for the actual motion of fish and autonomous sensor devices. We incorporated these recent developments into the BioPA for an existing Kaplan turbine unit installed at the Ice Harbor Dam on the Snake River. Autonomous sensor devices were used in past field studies to evaluate the hydraulic conditions when the unit operated at three discharges. We calculate and compare statistics from both the field data and BioPA modeling results to assess the accuracy of the simulation model and its assumptions. Specifically, we compare signatures of absolute pressure and probability of collision with the runner blades.
Research Projects
Organizational Units
Journal Issue
Publisher Version
Embedded videos