Dams and Civil Structures, North America

Cashing in on Existing Infrastructure

Issue 3 and Volume 34.

Hydrokinetic turbines offer an exciting opportunity to produce power on existing federally- and state-owned infrastructure, but their impact on such systems remains uncertain. The U.S. Department of Interior’s Bureau of Reclamation and its partners are working to build a tool designed to answer some of these questions.

By Josh Mortensen

Josh Mortensen, P.E., is a hydraulic engineer for the U.S. Department of Interior’s Bureau of Reclamation.

Much attention has been given to the development of hydrokinetic power generation technologies for use in inland waterways. These hydrokinetic turbines utilize the energy of the flowing water to turn a turbine, which is connected through a shaft to a generator to produce power. Unlike traditional hydropower, these devices do not require a reservoir or significant drop in head to operate. The end goal of many of the private developers of these technologies is to eventually have them installed in rivers and canal systems throughout the U.S.

Since 2011, the U.S. Department of Interior’s Bureau of Reclamation has received requests from private hydrokinetic developers wanting to install their technologies in Reclamation canal systems for demonstration testing. As the impacts of this technology on existing systems are unknown, Reclamation initiated a study in early 2013 to determine the hydraulic effects of these devices on existing water delivery and hydropower operations.

As part of this effort, Reclamation’s Research and Development Office in conjunction with its Technical Service Center’s Hydraulic Investigations Group have partnered with hydrokinetic developer Instream Energy Systems Corp. and Sandia National Laboratories to study hydrokinetic performance and hydraulic impacts in canal systems. Results are being obtained from physical testing using Instream’s hydrokinetic installation in the Roza Main Canal near Yakima, Wash., in conjunction with numerical hydraulic modeling. The main objective of this study is to develop a numerical tool to accurately predict impacts from hydrokinetic deployments in canals.

The need for data

A variety of hydraulic structures typical in water delivery systems — such as flow measurement devices, diversions, check gates, siphons, and pumping and hydropower facilities — are dependent on appropriate hydraulic conditions for effective operation. To quantify potential hydraulic implications to canal operation arising from the installation of a hydrokinetic device, Reclamation and its partners have installed instrumentation at key locations along the Roza Canal to monitor flow deliveries, water surface elevations, and hydropower performance throughout the hydrokinetic testing.

Comparison of hydraulic and hydropower data during hydrokinetic testing will provide insight into any impacts on the existing system. Testing under a variety of flow conditions and with additional configurations is planned through 2016. This information is important for future decisions regarding “if and where” conjunctive use of hydrokinetic power generation in water delivery systems can be used to meet both power and water usage needs.

Selecting a test site

The Roza Main Canal — located in south/central Washington — spans about 11 miles from the headworks at the Roza Diversion Dam on the Yakima River to the 13-MW Roza hydropower plant. It includes two tunnels, two inverted siphons, two unlined sections and no turnouts. The main canal has a trapezoidal geometry with an average slope of 0.0004 ft/ft and a flow capacity of 2,000 cubic feet per second. Water is diverted to the Roza Irrigation District from a point just upstream of the Roza plant during the irrigation season, with remaining flows passing through the plant before returning to the Yakima River.

The canal system is ideal for hydrokinetic testing because it contains a wide variety of canal and hydropower features that are typical in the many thousands of miles of open-channel water distribution systems operated by Reclamation in the western U.S. The system meets a number of hydraulic criteria needed for a full range of hydrokinetic operation, and its long stretch of canal with no turnouts made it easier to compare operating and baseline conditions during different times of the year. The Roza Canal was identified as a potential test site when Instream approached Reclamation looking to conduct a field demonstration.

Instream’s hydrokinetic unit was installed near the mid-point of the Roza Canal. The vertical-axis turbine includes a rotor that is connected to a generator through a vertical shaft, with power generated as the rotor is turned by the velocity of the canal flow. The turbine, rotor and generator are connected to a horizontal shaft that is rotated 90 degrees, allowing it to be placed in or removed from the canal in a matter of minutes. Baseline and operating data can be taken on the same day because the units can be rotated up and out of the water in a matter of minutes.

A downstream look at Instream’s hydrokinetic turbine deployed on the Bureau of Reclamation’s Raza Canal.

Power produced by the generator is used at the site by Instream’s mobile office and control room when needed, or dispersed through a heat bank..

Collecting information

Water level measurements were made at 12 cross-sections spaced about 33 feet apart upstream and downstream of Instream’s installation, minus one on each immediate side of the rotor centerline. Velocity measurements were also made at most of the cross-sections. Water levels were also measured at a few other locations along the canal near inverted siphons, and canal contractions and expansions to help calibrate the numerical model.

Onset HOBO water level loggers recorded water level measurements and a Teledyne Rio Grande acoustic Doppler current profiler (ADCP) was used to measure velocities at each cross-section. Water surface data were used in conjunction with the velocity data to determine the total energy loss coefficients associated with the hydrokinetic turbine, with the information used to calibrate the numerical model developed by Sandia.

Creating a numerical model

Field data were used to calibrate two numerical models of the canal system — with and without a hydrokinetic installation.

The first is a one-dimensional Hydraulic Engineering Center River Analysis System (HEC-RAS) model of the Roza Canal, which was calibrated by Reclamation. The main use of the HEC-RAS model was to determine water level impacts throughout the canal system. The model was calibrated using water level and canal discharge data from the field, with HEC-RAS developed as the main predictive modeling tool for general hydraulic impacts because it is widely used in the industry and can be applied to a variety of systems.

While this model can only be accurately calibrated to the physical data available from the Roza Canal system, it is anticipated that energy loss characteristics of the hydrokinetic turbine can be applied in other similar systems to provide reasonable hydraulic impact predictions using the modeling tool.

The second model used in the study is a quasi-3D Environmental Fluid Dynamics Code (EFC) that is calibrated by Sandia to include hydrokinetic features. The EFDC model helps characterize the turbine’s impacts and incorporates cross-sectional velocity data to determine the total energy loss across the hydrokinetic device. Results from this model will be compared to HEC-RAS results to ensure accurate impact predictions within the range of physical test results.

Basic setups of each model were completed in 2013, and they will continue to be modified to match field data through 2016.

Hydraulic impact results and observations

Field testing was conducted in 2013 as well as May and August of 2014 under a range of operating conditions including canal discharges from 1,700 to 1,900 cubic feet per second and a range of hydrokinetic power output. The upstream water level increased about 0.10 feet on average up to 175 feet from the unit. Due to the shallow slope of the canal this increase propagated up to 2,400 feet upstream where the average increase was 0.07 ft. This increase is minimal and may allow certain systems to be available for hydrokinetic installation without impacts to their operation. However, it may be a concern to canals that already operate near their free-board limitation. Introducing multiple turbines into a canal could potentially compound this effect into a water level increase that is much more significant. Additional field testing with multiple units could help answer this question and verify numerical results. Downstream water levels decreased significantly in the near-field but recovered within 200 feet of the hydrokinetic turbine.

Velocity profiles upstream of the hydrokinetic device were mostly unaffected by the presence of the turbine in the flow, with an average velocity of about six feet per second and a maximum velocity of about seven feet per second near the center of the canal. Downstream velocity profiles were complicated by the turbine in the canal flow and are being analyzed by Sandia National Laboratories for hydrodynamic parameters to determine potential spacing and placement of additional turbines. These results will also be useful in quantifying other hydraulic effects to canals such as turbulence.

Impacts at the Roza plant

Index testing of the Roza hydropower plant’s single Francis unit was conducted simultaneously with the 2013 baseline and operational hydrokinetic tests in the canal. This comparison showed there was no difference in the Francis unit’s performance from the baseline and operational tests. This result is not surprising, as the Roza plant is located several miles downstream from the hydrokinetic test site in a canal with sub-critical flow.

However, testing was performed to verify that there were no negative impacts caused by the deployment, operation or removal of the hydrokinetic unit from the canal.Due to these results, index testing will not be conducted in the future. However, forebay and tailrace levels, unit flow, penstock pressure and power output at the plant will be recorded through the 2015 and 2016 testing seasons.

Due to the current results, performance testing will not be conducted in the future. However, forebay and tailrace levels, unit flow, penstock pressure and power output at the plant will be recorded through the 2015 and 2016 testing seasons.

Conclusions and future plans

Field testing was conducted to evaluate hydraulic impacts from hydrokinetic power generation to existing canal and power plant operations. These measurements are still being used to develop numerical modeling tools to predict system impacts and unit performance of hydrokinetic units in canal systems. Results from field tests in 2013 and 2014 showed slight increases in upstream water surface elevations, but overall impacts were minimal. There were no impacts to the existing Roza Powerplant downstream.

Additional field testing is planned for 2015 and 2016 with a larger hydrokinetic turbine as well as multiple turbines in series in the canal. Field data with these configurations will help determine the relationship of hydraulic impacts with hydrokinetic operation under a greater range of power output and multiple unit deployments. These results will continue to support the numerical modeling efforts to develop a tool to accurately predict hydrokinetic impacts to canals.

More HR Current Issue Articles
More HR Archives Issue Articles