Measuring water velocity using DIDSON and image cross-correlation techniques
To design or operate hydroelectric facilities for maximum power generation and minimum ecological impact, it is critical to understand the biological responses of fish to different flow structures. However, information is still lacking on the relationship between fish behavior and flow structures despite many years of research. Existing field characterization approaches conduct fish behavior studies and flow measurements separately and coupled later using statistical analysis. These types of studies, however, lack a way to determine the specific hydraulic conditions or the specific causes of the biological response. The Dual-Frequency Identification Sonar (DIDSON) has been in wide use for fish behavior studies since 1999. The DIDSON can detect acoustic targets at long ranges in dark or turbid dark water. PIV is a state-of-the-art, non-intrusive, whole-flow-field technique, providing instantaneous velocity vector measurements in a whole plane using image cross-correlating techniques. There has been considerable research in the development of image processing techniques associated with PIV. This existing body of knowledge is applicable and can be used to process the images taken by the DIDSON. This study was conducted in a water flume which is 9 m long, 1.2 m wide, and 1.2 m deep when filled with water. A lab jet flow was setup as the benchmark flow to calibrate DIDSON images. The jet nozzle was 6.35 cm in diameter and core jet velocity was 1.52 m/s. Different particles were used to seed the flow. The flow was characterized based on the results using Laser Doppler Velocimetry (LDV). A DIDSON was mounted about 5 meters away from the jet nozzle. Consecutive DIDSON images with known time delay were divided into small interrogation spots after background was subtracted. Across-correlation was then performed to estimate the velocity vector for each interrogation spot. The estimated average velocity in the core zone was comparable to that obtained using a LDV. This proof-of-principle project demonstrated the feasibility of extracting water flow velocity information from underwater DIDSON images using image cross-correlation techniques.
- Research Organization:
- Pacific Northwest National Lab. (PNNL), Richland, WA (United States)
- Sponsoring Organization:
- USDOE
- DOE Contract Number:
- AC05-76RL01830
- OSTI ID:
- 973733
- Report Number(s):
- PNNL-SA-64821; TRN: US201006%%956
- Resource Relation:
- Conference: Waterpower XVI: New Roles for Hydro in a Changing World, July 27-30, 2009, Spokane, Washington, Paper No. 130
- Country of Publication:
- United States
- Language:
- English
Similar Records
Particle-image Velocimetry Sensitivity through Automatic Differentiation
Response Relationship Between Juvenile Salmon and an Autonomous Sensor in Turbulent Flows