Assessing and Testing

Hydrokinetic Turbine
Performance and Effects
on Open Channel
Hydrodynamics: An
Irrigation Canal Case
Study
March 2017
(This page intentionally left blank)
Acknowledgments
This research was made possible through funding from the Department of Energy’s EERE Office’s Wind
and Water Power Technologies Office and the Bureau of Reclamation’s Science and Technology Program.
Instream Energy Systems Corp. provided the installation and operation of the HK unit and Reclamation’s
Pacific Northwest Region allowed use of the Roza Main Canal and site access. Sandia National
Laboratories is a multi- mission laboratory managed and operated by Sandia Corporation, a wholly owned
subsidiary of Lockheed Martin Corporation, for the U.S. Department of Energy’s National Nuclear Security
Administration under contract DE-AC04-94AL85000.

This report was prepared by:

Budi Gunawan, Water Power Technologies Department, Sandia National
Laboratories, Albuquerque, NM, USA
Vincent S. Neary, Water Power Technologies Department, Sandia National
Laboratories, Albuquerque, NM, USA
Josh Mortensen, Technical Service Center, Hydraulic Investigations and
Laboratory Services, United States Bureau of Reclamation, Denver, CO, USA
Jesse D. Roberts, Water Power Technologies Department, Sandia National
Laboratories, Albuquerque, NM, USA

Acknowledgments i
HK IN OPEN CHANNELS REPORT

List of Acronyms
ADCP acoustic Doppler current profiler
ADV acoustic Doppler velocimeter
AEP annual energy production
BP battery-powered
CEC current energy converter
CFD Computational Fluid Dynamics
CH conventional hydropower
CS cross-section
D Dimension
DOE Department of Energy
EGL energy grade line
FEMA Federal Emergency Management Agency

FIS flood insurance studies

GPS global positioning system
HGL hydraulic grade line
HK hydrokinetic
MBE multi-beam depth echo sounders

MOU Memorandum of Understanding

MHK marine hydrokinetic
MV moving vessel
NGVD National Geodetic Vertical Datum
RC remote controlled
RTK real-time kinematic
SBE single-beam depth echo sounders
TSR tip-speed ratio

WPTO Water Power Technologies Office

ii List of Acronyms
Nomenclature
A flow area ūx 
hub height velocity an x distance downstream
of the turbine
b bottom width
ūdef hub height velocity deficit
Cd turbine drag coefficient
Ub bulk or section mean velocity
Cp turbine power coefficient
Vi streamwise velocity at section i
di flow depth normal to flow at section i
Vm velocity component of the velocity head
D hydraulic depth
x, y, z streamwise, cross stream and vertical axes
Dt instantaneous drag force
xi x, y, z axes
E Modulus of Elasticity of the strut
y flow depth
Fr Froude number
yi vertical distance from channel bottom to water
g gravitational acceleration surface at section i
hb energy loss due to turbine blockage zi distance from datum line to channel bottom at
hf friction losses section i

hm minor losses Z side slope

ht energy extracted by turbine αi Coriolis coefficient at section i

L reach length ε energy dissipation rate per unit mass

Ls characteristic length scale of the channel ε_ bending strain

geometry h Kolomogorov microscale
M bending moment l tip-speed ratio
p pressure
ν kinematic viscosity
P wetted perimeter
ρ density of water
PT instantaneous mechanical turbine power
t instantaneous torque
Q flow discharge
ω instantaneous turbine angular velocity
r strut diameter
R hydraulic radius cfs cubic feet per second
Re Reynolds number kW kilowatt
So longitudinal channel bed slope m meter
T top width cm centimeter
u, v, w instantaneous streamwise, cross stream and m/s meters per second
vertical velocities
ui instantaneous velocity components u, v, w
ui’ fluctuating part of ui
ūi mean part of ui or characteristic mean velocity
ū∞ upstream approach velocity at hub height

Nomenclature iii
HK IN OPEN CHANNELS REPORT

Executive Summary
Hydrokinetic energy from flowing water in open channels has the potential to support local electricity
needs with lower regulatory or capital investment than impounding water with more conventional means.
MOU agencies involved in federal hydropower development have identified the need to better understand the
opportunities for hydrokinetic (HK) energy development within existing canal systems that may already have
integrated hydropower plants. This document provides an overview of the main considerations, tools, and
assessment methods, for implementing field tests in an open-channel water system to characterize current
energy converter (CEC) device performance and hydrodynamic effects. It describes open channel processes
relevant to their HK site and perform pertinent analyses to guide siting and CEC layout design, with the
goal of streamlining the evaluation process and reducing the risk of interfering with existing uses of the site.
This document outlines key site parameters of interest and effective tools and methods for measurement and
analysis with examples drawn from the Roza Main Canal, in Yakima, WA to illustrate a site application.

iv Executive Summary
Table of Contents
1 INTRODUCTION........................................................................................................................................................ 1

1.1 Project Motivation................................................................................................................................................ 1

1.2 Hydrokinetics and Canals................................................................................................................................ 1

1.3 About This Document..................................................................................................................................... 3

2 PROPERTIES TO BE MEASURED...................................................................................................................4

2.1 Study reach bathymetry.................................................................................................................................4

2.2 Turbine Power Performance........................................................................................................................ 5

2.2.1 Power................................................................................................................................................................. 5

2.2.2 Drag...................................................................................................................................................................6

2.3 Hydrodynamic Effects....................................................................................................................................6

2.3.1 Flow Field Properties.................................................................................................................................6

2.3.2 Energy (Head) Parameters....................................................................................................................9

3 INSTRUMENTATION, DEPLOYMENT AND MEASUREMENT PROTOCOLS................................11

3.1 Bathymetric mapping recommendations..............................................................................................11

3.2 Water level monitoring..................................................................................................................................12

3.3 Velocity and turbulence measurements.............................................................................................. 14

3.3.1 Acoustic Doppler current profiler...................................................................................................... 14

3.3.2 Acoustic Doppler velocimeter............................................................................................................ 18

3.4 Power measurement..................................................................................................................................... 19

3.5 Drag measurement....................................................................................................................................... 20

4 Predicting the effects of HK turbine deployment using numerical modeling........................22

5 Summary.................................................................................................................................................................. 25

6 References................................................................................................................................................................27

Table of Contents v
HK IN OPEN CHANNELS REPORT

List of Figures
Figure 1. Canal cross-section (left) and profile (right) parameters...................................................4

Figure 2. Sketch defining flow coordinate system and mean, instantaneous flow profiles,
and flow changes induced by a turbine (Neary et al. 2011)..............................................8

Figure 3. Energy grade line between two locations along a channel in open-channel flow,
adapted from Te Chow (1959). .................................................................................................... 10

Figure 4. The remote control survey boat with real-time kinematic GPS correction..............11

Figure 5. An example of a coarse bathymetric survey using a single-beam echo sounder
at the Roza Canal site, showing the raw data (left) and the interpolated data
(right).........................................................................................................................................................12

Figure 6. Water level time series at the Roza Canal site, at 50 m and 700 m upstream of
the turbine. ............................................................................................................................................13

Figure 7. Top left: An example of water level pressure transducer; the length of the
transducer in picture is 15 cm. Top right: Transducer, deployed using a PVC pipe
at the channel’s side wall.
Bottom: angle iron profiles mounted on the channel’s side wall.................................. 14

Figure 8. Different types of ADCP beam configurations (RDI 2011)................................................15

Figure 9. Boat capsize due to highly turbulent flow and uneven water surface
downstream of a turbine...................................................................................................................15

Figure 10. A cableway deployment with a tensioned tagline (the white line), that is
attached to the ADCP boat........................................................................................................... 16

Figure 11. ADCP velocity contours (looking downstream), as shown in the RD Instrument’s
WinRiver software, at four cross-sections at the Roza Canal test site: 50 m
upstream of the turbine, 10 m downstream of the turbine, 20 m downstream of
the turbine, and 30 m downstream of the turbine. The x axis shows the length
of the ADCP travel path while traversing the canal............................................................17

Figure 12. Spatiotemporally-averaged velocity contour (looking downstream), 10 m

downstream of the turbine, at Roza Canal under high tip-speed ratio. Flow
discharge was approximately 55 m3/s. Black scatters in the top figure indicate
the locations of velocity measurements...................................................................................17

vi List of Figures
Figure 13. Streamwise velocity contours, interpolated from MV ADCP survey. Left: near-
surface velocities, right: hub-level velocities. The white circle indicates the
locations of the Instream turbine. .............................................................................................. 18

Figure 14. Cable-deployed ADVs (in circles) with sounding weight (Photograph courtesy
of Bob Holmes, USGS, 2010)......................................................................................................... 19

Figure 15. Sandia’s mobile ADV deployment system for canal measurements. On top is the
schematic of the system and a picture of the unit deployed in the field is shown
on the bottom. ..................................................................................................................................... 19

Figure 16. T
 he torque sensor system mounted to the Instream’s turbine shaft at Roza
Canal. The small brown rectangular sensor to the left of the torque sensor
system is a strain gauge used for drag measurements................................................... 20

Figure 17. An example of a battery-powered wireless strain sensor system, for torque or
thrust measurement (www.binsfeld.com)............................................................................. 20

Figure 18. An example of a turbine performance curve for different tip-speed ratios......... 20

Figure 19. Strain gauge mounting location for measuring turbine drag on cross-flow wind
turbine (Griffith et al. 2011).............................................................................................................21

Figure 20. Model-predicted velocity contours for Roza Main Canal site (at turbine mid-
span, flow is from top to bottom). The left figure shows the whole simulation
domain, with a turbine positioned near the outflow boundary. The right figure
is the velocity contour at the turbine location. Legend units are in meter/
second.................................................................................................................................................... 23

Figure 21. Measured and simulated wake velocities at turbine mid-span at Roza Main
Canal, as a function of distance from the turbine............................................................. 23

Figure 22. Simulated Energy Grade Lines and Hydraulic Grade Lines (water surface
elevations along the reach) at Roza Main Canal with and without the turbine’s
presence................................................................................................................................................. 24

List of Tables
Table 1 Recommended measurements for the assessment of potential impacts from open-channel
HK operations................................................................................................................................................ 25

List of Figures and Tables vii

(This page intentionally left blank)
1 INTRODUCTION
1.1 Project Motivation
In 2010, a Memorandum of Understanding (MOU) regarding federal hydropower development in the U.S. was
signed by the Department of Energy (Office of Energy Efficiency and Renewable Energy), the Department of the
Interior (Bureau of Reclamation), and the Department of the Army (Corps of Engineers), and was later re-signed
in 20151. The 2015 Phase II MOU Action Plan developed by the MOU agencies identified the need to better
understand the potential for hydrokinetic (HK) energy development
within existing canal systems that may already have integrated
hydropower plants. Because HK generation of electricity from Hydrokinetic energy from
flowing water in open channels may be less infrastructure intensive flowing water in open channels
than conventional hydropower, that impounds free flowing water, has the potential to support
it has the potential to support local electricity needs with lower
regulatory or capital investment. This work is specifically aligned local electricity needs with
with renewable energy development goals of the Department lower regulatory or capital
of Energy (DOE) and the Department of the Interior (DOI) by
investment than impounding
investigating how much excess energy is available for electricity
generation via hydrokinetic technologies, while still meeting the water with more conventional
design requirements of the existing water supply systems and needs means.
of the water users. As an ancillary benefit of this work, many of
the methods and tools developed will be directly transferable to
the MHK industry for future potential developments in large rivers, tidal channels, and open ocean currents. This
aligns well with the primary goal of the DOE Water Power Technologies Office (WPTO) to efficiently develop and
utilize the country’s marine hydrokinetic (MHK) and conventional hydropower (CH) resources.

1.2 Hydrokinetics and Canals

Current energy conversion (CEC) technologies are a class of marine and hydrokinetic (MHK) technologies that
convert kinetic energy of canal, river, tidal or ocean water currents to generate electricity. Most CEC technologies
under development are water turbines, or hydrokinetic (HK) turbines, that are modeled after wind turbines. The
two common classes of HK turbines are axial-flow turbines, in which the rotor shaft is oriented parallel to the water
current, and cross-flow turbines, in which the rotor shaft is oriented perpendicular to the water current. The shaft
of cross-flow turbines can be oriented either vertically or horizontally. Both of these HK turbine classes can be
deployed in rivers and canals with additional design work and considerations to maximize energy generation from
natural and man-made features.

National MHK resource assessments have been conducted for wave, ocean current, tidal and river sites (EPRI
2011; Yang et al. 2015; Defne et al. 2012; EPRI 2012, respectively) with funding from the DOE. The EPRI 2012
report determined a theoretical resource availability of 1,381 TWh/year from river currents in the U.S., a significant
resource. Nearly 72,000 river segments with mean annual flow greater than 1,000 cubic feet per second (cfs)
were included in the assessment. The assessment, however, did not include canals and waterways. Therefore, the
overall potential for hydrokinetic energy production in these types of water conveyance channels is uncertain.
Nevertheless, high current speeds and resource predictability in canals and their general accessibility may be
favorable for energy generation through HK technologies.

The US canal system comprises tens of thousands of miles of canals. Existing canals come in a variety of shapes
and sizes to meet their primary objective of conveying water to support irrigation, navigation, and hydropower
developments. Canals are either earthen (unlined) or lined (often with concrete) to minimize unwanted changes

1 More information on the Federal Hydropower MOU is available at:

http://en.openei.org/wiki/Federal_Memorandum_of_Understanding_for_Hydropower

Introduction 1
HK IN OPEN CHANNELS REPORT

to canal specifications that may result from scour, vegetation,

etc. Those for navigation usually have low current speeds, often Although resource
well below turbine cut in speeds, and generally do not hold great
potential for hydrokinetic development. Many irrigation canals assessments have been
and power canals that feed into or out of hydropower dams conducted for wave, ocean
display the characteristics needed for HK energy development
current, tidal, and river
(i.e. have sufficient current speeds and water depths). In all cases,
the development of HK in canals requires strict coordination with sites, the potential for HK
existing canal owners to avoid interference with the canal’s primary energy production in canals
objective. Each canal type presents unique design challenges and
opportunities for HK energy development.
and waterways is uncertain.
However, certain canals,
While not all of these canals are suitable for HK energy including irrigation and
development, some will have flow speeds, water depths, and
other attributes, e.g., proximity to grid connection, favorable for power canals, have flow
commercial HK energy development. Favorable characteristics speeds, water depths, and
for cost-effective HK energy development can include high current site characteristics that are
speeds (>1.5 m/s) corresponding to high resource; high free-board
level (vertical distance between the water surface and top of the favorable for commercial HK
channel) to allow greater flexibility of water level variation prior to energy development.
water exceeding freeboard limits of the canal; and good site access
(ability to bring equipment to the canal’s edge). Lined channels are
also generally more favorable than unlined channels, as they are more resistive to scour. Channels that do not have
protected organisms living in them, will have comparatively, less environmental compliance requirements.

Potential concerns include disrupting water supply operations (by affecting head-discharge conditions at irrigation
canal intakes), increasing flood risks (by increasing water levels as a result of blockage and backwater effects),
reducing power generation of hydropower plants (by affecting plant inflow, tailwater levels and net head at
hydropower dam or discharge), and causing channel instabilities that lead to unfavorable morphological changes.
Furthermore, man-made canals are generally designed for a specific purpose and operating regime, such that the
design velocities and flow depths are given careful consideration to meet the canals primary objective. HK canal
deployments would necessarily change the hydrodynamics of the canal. It is critical to account for these changes to
ensure adverse effects, e.g., unwanted sediment deposition (silting), scouring, overtopping, diversion, and reduction
in CH energy production, are avoided.

The amount of water conveyed by a canal or a river is variable

throughout the year due to, for example, seasonal rainfall and Potential concerns, such
irrigation demands. Therefore, the HK resource in rivers and
canals varies seasonally. Adding hydrokinetic turbines to these as disrupting water supply
systems would also change the local/reach hydrodynamics, which operations, increased
if not carefully accounted for could cause unwanted events, such as
flood risks, reduced power
flooding, silting and scouring. All of these factors should be taken
into account in the design, operation, and overall development at hydropower plants,
of hydrokinetics in these resources. Furthermore, there may be unfavorable morphological
opportunities to design new canals to maximize HK potential
through features or geometries and also to deploy HK in rivers. For changes, and seasonal water
example, unconventional techniques could allow for harvesting variability, should be taken
energy from rapids, maximizing natural and/or engineered into account in the planning,
characteristics of rivers and canals.
design, and operation of a HK
Feasibility studies for HK energy development must demonstrate energy project.
that HK operations will not adversely affect canal operations
and other stakeholders. A combination of field measurements
and numerical modeling can be used for impact assessment of HK projects by evaluating changes to basic

2 Introduction
hydrodynamic parameters, such as changes to water surface elevations and velocities at different locations along
and across the canal resulting from an HK project.

1.3 About This Document

This document is intended as a “how to” manual for conducting hydrokinetic (HK) performance testing in canals
and rivers, and assessing the effects of HK device deployment on hydrodynamics. Although this manual provides
detailed information based on a single HK case study in a conduit canal with known flow velocity, the methodology
presented here is highly applicable to a variety HK design opportunities and infrastructure additions in other types
of canals and rivers. This manual is largely written for engineers and researchers who have background knowledge
on turbine performance and open channel hydraulic measurement, analysis, and field testing. Other users will
benefit from the methods and experience shared in this manual. The methods presented in this manual may not
be suitable for all sites, but users are expected to adapt the presented measurement methods, or other methods, for
their site and technology specific conditions.

The manual provides information for designing and implementing

testing in an open channel system for characterizing device
performance and the hydrodynamic effects of single device Using the Roza Main Canal in
and array deployments. Chapter 2 outlines the recommended Yakima, WA as a case study,
parameters for analyzing device performance and hydrodynamic this document provides
effects. These parameters include power coefficient, turbine
tip-speed ratio, inflow velocity, and water level. Some of the readers with an overview of
Chapter 2 contents are summarized from the case study manual key considerations, tools,
for river and tidal sites (Neary et al. 2011). Chapter 3 focuses
and protocols to conduct
on the types of instrumentation and methods to be used for
characterization, including data post-processing methodology. open-channel system site
Emphasis is placed upon acoustic-Doppler current profiler (ADCP) assessments and performance
and acoustic Doppler velocimeter (ADV) instruments for collecting
testing.
hydrodynamic measurements, as these instruments represent
the most practical and economical tools for use in the MHK
industry. Chapter 4 describes the ability of numerical models, of
varying fidelity, to predict the hydrodynamic changes induced by HK deployments and how they may be used in
conjunction with field measurements.

Much of the insights and materials presented in chapters 3 and 4 are gained from a hydrokinetic turbine case study
at the Roza Main Canal, Yakima, WA. The Roza Main Canal is approximately 11 miles long and diverts flow
from the Yakima River to the Roza Irrigation District where it supplies water for approximately 72,000 acres of
valuable farm land. It typically operates for about 11 months out of the year and shuts down for one month for
inspections and maintenance. The 11 mile reach features both lined and unlined sections as well as a bifurcation at
the downstream end where a portion of the flow is diverted to a power plant and then returns to the Yakima River.
Instream Energy Systems Corp. is currently using the Roza Main Canal site for testing a vertical-axis hydrokinetic
turbine. Readers interested in the details of ADV and ADCP measurement and data post processing can refer to
the guidance manuals Gunawan and Neary (2011) and Gunawan et al. (2011), which provide detailed methods and
protocols for the application of these instruments for MHK reconnaissance, feasibility, design and testing studies .
This document provides an overview of the main considerations for an open-channel CEC deployment site and the
tools and protocols to help guide the reader through a thorough site and technology assessment. It describes open
channel processes relevant to their HK site and perform pertinent analyses to guide siting and CEC layout design,
with the goal of streamlining the evaluation process and reducing the risk of interfering with existing uses of the
site. This document outlines key site parameters of interest and effective tools and methods for measurement and
analysis with examples drawn from the Roza Main Canal, in Yakima, WA to illustrate a site application.

Introduction 3
HK IN OPEN CHANNELS REPORT

2 PROPERTIES TO BE MEASURED
2.1 Study reach bathymetry
The bathymetry of the study reach provides critical information for modeling canal hydrodynamics, detailed in
Section 2.3., and developing numerical models for predicting turbine power performance and hydrodynamic
effects. Canals are defined herein as man-made open-channels with long reaches over which the section geometry
is uniform and the slope is constant. Accurate bathymetry of the canal reach of interest may be available in the
form of design or as-built drawings that include information on the canal bed profile, and cross-section geometry.
These may be available from the canal’s operator or flood insurance studies (FIS) from the Federal Emergency
Management Agency (FEMA). These drawings should be used
with caution as channel bathymetry can change significantly due to
natural erosion and deposition, turbine operation, or dredging and It is important to obtain
filling operations, particularly in unlined channels. Bathymetric accurate bathymetry from
surveys are recommended to verify these drawings or collect new design or as-built drawings.
bathymetry data if it is determined that substantial morphological
changes have occurred. In addition to accurate bathymetry, most Because channel bathymetry
canals have design or operational features that should be accounted can change, surveys are
for. These features include but are not limited to tunnels, inverted recommended to verify
siphons, flow measurement weirs and flumes, and emergency
wasteways and drains. the drawings or to collect
new data if changes have
The bathymetric properties for canals are illustrated in Figure 1 for
occurred. Additional design
a trapezoidal shaped channel, which is the most common section
geometry. The section geometry parameters include the shape, or operational features should
bottom width, b, side slope, Z:1, and top width, T. Definitions for also be accounted for.
these and other section parameters are given in Table 1. Profile
parameters include the reach length, L, which is the horizontal
distance between the most upstream section (1) and downstream section (2) for which the section geometry
parameters remain constant, and longitudinal bed slope.

Figure 1. Canal cross-section (left) and profile (right) parameters

The section design parameters, with L in parentheses indicating length, e.g., meters, are:

• Top flow width = T [L] = width at free-surface

• Flow depth = y [L] = vertical distance from channel bottom to free water surface

• Depth of flow sec = d [L] = flow depth normal to flow, d = y cosq

4 2 Properties to be Measured
 • Flow area = A [L2] = cross-sectional area normal to flow direction

• Wetted perimeter = P [L] = length of channel boundary in contact with water

• Hydraulic radius = R [L] = A/P

• Hydraulic depth = D [L] = A/T

2.2 Turbine Power Performance

Turbine power performance tests typically measure the mechanical
power of the turbine, the drag (thrust) force acting on the turbine, Measuring the mechanical
and the hydrodynamic force of the approach flow over a range of power of the turbine, drag
tip-speed ratios (TSR). The ratio of the mechanical power to the force on the turbine, and
hydrokinetic power of the approach flow, the power coefficient,
is a measure of the efficiency of the CEC technology. The ratio of hydrodynamic force of the
the drag force to the hydrodynamic force of the approach flow is approach flow is essential to
defined as the drag coefficient, which is equivalent to the thrust
assess the performance of
coefficient when the turbine is stationary (fixed position) because
the drag and thrust forces are in balance. These metrics provide turbine technologies and to
essential information that allow CEC technology developers to calculate the annual energy
assess the performance of their turbine technologies relative to other
production for HK power
turbine technologies, and to calculate the technical annual energy
production (AEP) for hydrokinetic power projects. They are also projects.
required data for certifying CEC technologies.

2.2.1 Power
Instantaneous turbine power can be determined from synchronous torque and rotor position measurements
(mechanical power). Instantaneous mechanical turbine power, PT , is the product of instantaneous torquet, and
instantaneous turbine angular velocity, ω.

The TSR is defined as the ratio of the rotor tip speed to the speed of the approaching flow,

where w is the instantaneous angular velocity. Turbine power coefficient, Cp, can be calculated as,

where A is the turbine’s flow facing area, p is the density of water and u is the instantaneous inflow velocity
over the turbine’s flow facing area. The turbine’s flow facing area is the multiple between the rotor diameter and
rotor height for vertical axis turbines (essentially a rectangular flow facing area) while an axial flow turbine has a
circular flow facing area calculated from the diameter of the turbine only. In the calculation of Cp, the divisor term
indicates the theoretical resource availability in the channel, therefore the inflow velocity needs to be measured far
enough upstream of the turbine such that turbine drag has little to no effect on the upstream velocity. A distance of
three to five turbine diameters is typically sufficient for this purpose (Hill et al. 2014; Hill et al. 2014; Neary et al.
2013).

Note that the term Cp is often used to describe power coefficient derived from the electrical power output instead of
mechanical power. The electrical power output is the power generated after taking into account the total efficiency
of the system. The total efficiency of the system can include efficiencies of the gearbox, drive train, and generator.
Example values of the efficiencies of these components are presented in publications, such as Hagerman et al.

2 Properties to be Measured 5
HK IN OPEN CHANNELS REPORT

2006, which states a range of efficiencies between 95% to 98%. It is therefore always recommended to state the
type of power (mechanical or electrical output) used for calculation when reporting a Cp value.

2.2.2 Drag
The hydrodynamic force acting on the rotor can be evaluated using the drag coefficient. The turbine drag
coefficient, Cd, can be calculated as,

where Dt is instantaneous drag force. The drag coefficient often varies with the TSR.

2.3 Hydrodynamic Effects

Any object placed in moving water will obstruct the flow and introduce a drag force, or drag. Under subcritical
flow conditions, this drag will cause water levels to rise upstream of the obstruction (backwater) and drop
downstream (drawdown). Hence deployment of HK turbines can alter water levels at various locations along the
canal. One important consideration is whether a HK turbine deployment may raise the water levels enough to
encroach the canal’s free-board requirement (maximum flow depth relative to top of canal bank) which increases
the risk for overtopping. If a threat of overtopping exists, the extent of the affected area needs to be determined to
predict damage and assess mitigation options. One of the mitigation options would be to operate the canal below
the normal condition, e.g. reducing the flow in the canal, which is
likely undesirable. Continuous water level monitoring along the
canal and knowledge of canal’s bathymetry are essential to be able Continuous water level
to determine if overtopping could occur and its potential extent. monitoring and knowledge
Another important consideration is whether a HK turbine of the canal’s bathymetry
deployment could create sediment deposition or scouring due to is needed to determine the
modifications to the flow velocities around the turbine. Too low of
a velocity can cause sediment deposition, while too high a velocity potential for overtopping.
can cause sediment scouring. The velocity magnitudes at different Velocity magnitudes should
locations around the turbine should be sufficient enough to keep
be sufficient enough to
sediments in suspension without scouring. Permissible velocities
vary for channels with different lining materials and other channel prevent sediment deposition
parameters. Permissible velocity can be found in the literature, such or scouring.
as a manual developed by the US Army Corps of Engineer
(USACE 1994).

2.3.1 Flow Field Properties

The flow field within a canal reach is the distribution of the instantaneous streamwise x, cross stream y, and vertical
z components of velocity (u, v, w) and pressure p over space and time.These flow field properties are typically time-
averaged to reduce the amount of information to a tractable description of the flow field for engineering analysis
(Neary et al. 2011). The time-averaged u can also be spatially averaged over the entire section to determine the
bulk or section mean velocity Ub. Instantaneous velocities are also important for determining instantaneous loads
that contribute strongly to fatigue failure.

2.3.1.1 Bulk Flow and Section Geometry Properties

Bulk flow properties that characterize the flow at a canal cross-section at the time of measurement include the bulk
velocity

6 2 Properties to be Measured
where Q is the flow discharge in the channel and A is the cross-section area; and section parameters (flow depth,
y, top width, T, flow area, A, hydraulic depth D, wetted perimeter P, and hydraulic radius R) which are calculated
from field measurements using the equations described in section 2.1. Methods for estimating the flow resistance
parameter, Manning’s n, can be found in common references on open-channel hydraulics (e.g., Sturm (2001)).
As the channel discharge (flow) was steady at the time of measurements, these bulk flow properties did not vary
with time at any given section in the canal reach. However, the flow in the study reach was nonuniform, meaning
these properties vary over the length of the channel.

These properties are used to calculate important non-dimensional parameters that indicate the flow state and flow
regime, the Reynolds number Re and the Froude number Fr, which are calculated as,

For Re values above about 2,000 - 4,000 inertial forces dominate over viscous forces, creating flow instabilities
and turbulence. Only canals with relatively fast moving, therefore, turbulent flow are suitable for HK deployment.
The Re value in these canals is typically well over 105. Fr value below one indicates that the celerity or speed of
propagation of a small surface wave is greater than the bulk velocity (i.e. gravitational forces dominate
over inertia). Most reaches of canals are designed to ensure subcritical flow regimes to reduce the potential
for scour. As a result, flow conditions upstream, e.g., velocity and water depth, are influenced by downstream
conditions. For example, the placement of a hydrokinetic turbine at a section in a canal causes a local obstruction
that increases the water surface elevation. This local rise in the water surface is transmitted upstream, resulting in
higher water surfaces, and reduced velocities upstream.

2.3.1.2 Mean (Time-averaged) inflow velocity and velocity deficit

The instantaneous velocity component is decomposed into its time-mean and turbulent fluctuation,
ui = ui + ui' along its respective axis as illustrated in Figure 2. The instantaneous velocities u, v, w are
defined herein as the streamwise, cross-stream, and vertical velocity components, respectively. The most common
means of characterizing a flow field is to measure the statistics of the flow from the mean velocity up through
higher order statistics such as the flow skewness and kurtosis.
These statistical measures can provide a wealth of information
related to the time or space averaged flow, however, they do not Inflow velocity measurements
provide a measure of the scale and of the instantaneous, unsteady
are important for
flow structures that may be important in generating unsteady loads,
vibration and noise. These unsteady flow structures tend to have determinining device.
smaller spatial scales with higher magnitude fluctuations and local
gradients than the time averaged profiles would suggest.

The instantaneous mean inflow velocity, as discussed in section 2.2., is important for determining device
performance and loading. Additionally, the mean streamwise velocity deficit is a common metric used to assess the
velocity recovery in the wake downstream of a turbine, and to optimize the turbine layout in a turbine farm. The
velocity deficit can be defined as

where is the upstream approach velocity at hub height and is the hub height velocity at position x
downstream of the turbine.

2 Properties to be Measured 7
HK IN OPEN CHANNELS REPORT

u
Turbulent inflow Tip vortices
u

Blade Rate:
blade passage
frequency

mean flow profile Blade & tower

Wakes

z,w
Flow shedding

Neary et al 2010
Junction vortices

Figure 2. Sketch defining flow coordinate system and mean, instantaneous flow profiles, and flow changes
induced by a turbine (Neary et al. 2011).

2.3.1.3 Turbulence and periodic motion in the flow

When flows are turbulent, the flow field properties at any given point, for any instant in time, depart from the
mean (time-averaged) flow over relatively small space and time scales as illustrated in Figure 2. These time
scales can be characterized from small (i.e., Kolmogorov microscale = ) to large (i.e., convective time-scale
= ), where ν is the kinematic viscosity, ε is the energy dissipation rate per unit mass, L is a characteristic
length scale of the channel geometry (e.g., flow depth) and is a characteristic mean velocity. Turbulent flow is
three-dimensional and three-component, and is characterized by a continuous range of flow scales in the form of
rotational motion (vortices or eddies). This range of spatial scales in a turbulent flow is dependent on the Reynolds
number, Re, where the higher the Reynolds number the broader the range of scales. The smallest scale in a
turbulent flow is limited by the fluid viscosity and is estimated by the Kolmogorov (spatial) microscale, η = (ν3/ε)¼,
while the largest spatial scale is characterized by the channel bounding geometry, L or a multiple thereof.

The turbulence intensities are dimensionless parameters that describe the level of turbulence within the flow along
each spatial direction. Turbulence speeds up the wake velocity recovery, which allows closer spacing between
turbines. Turbulence intensities are defined as the root-mean square of the fluctuating velocity component divided
by the mean velocity magnitude,
.

; ;
Turbulence occurs in both wall bounded (solid surfaces such as the bed floor) and in free shear flows such as
wakes formed behind HK devices, shown in Figure 2. In addition to these turbulent inflow structures, unsteady
downstream flow structures, that may be locally turbulent, may exist such as vortical flow structures and unsteady
or cyclic flow shedding. The types of vortical flow structure created is dependent on device design and mounting
configuration, but may consist of blade tip vortices, tower/bed-floor junction vortices and Karman vortex street
associated with flow shedding off of structures. The downstream wake of the device will consist of both large
scale (on the order of the rotor plane) and smaller scale flow structures. The large or macro-scale momentum
deficit wake of the device may have a rotational flow pattern related to blade rotation and will have characteristic
cyclic frequencies associated with the blade passage frequency. The
smaller scale structures will consist of individual wakes, vortices and flow Information on flow
unsteadiness as illustrated in Figure 2. Proper characterization of both the
inflow and downstream flow features is important in determining overall turbulence is useful for
device performance in single and array deployments. While the mean flow numerical models to
resource directly impacts long term device power output and powering accurately predict wake
performance, it is the short term unsteady flow characteristics which create
unsteady loads on the device components, vibration, and sound generation. velocity recovery.

8 2 Properties to be Measured
2.3.2 Energy (Head) Parameters
The section averaged 1D conservation of energy equation is typically applied, along with continuity (conservation
of mass) and the conservation of momentum, to model steady open-channel flow hydraulics in canals, and changes
in velocity and water depth due to a variety of modifications, including the deployment of instream structures,
water diversion, and energy extraction. All terms in the energy equation are normalized by the unit weight of
water, which results in energy head in length units [e.g., N-m/N = m]. The total energy head at any given section
consists of the sum of the elevation head, z, the piezometric head (water depth), y, and the velocity (kinetic energy)
head, aVm2/2g. As the water flows downstream energy losses due to friction, turbulence, and energy extraction
are incurred. Due to the nonuniform velocity distribution in a cross-section, the average velocity head does not
equal the velocity head of the average velocity,

A correction factor, the Coriolis coefficient, must, therefore, be applied to make the adjustment in the energy
equation,

, where

The hydraulic grade line (HGL) is the water surface profile, which is the line drawn through z + y values at all
sections along the canal reach. The energy grade line (EGL) is the profile drawn through the total energy at all
sections along the canal reach. Methods for determining friction and minor losses are given in standard open
channel hydraulics texts (e.g., Sturm 2001).

= Bed elevation at an upstream location, relative to a datum (m)

= Water sdepth upstream (m)

= Coriolis coefficient upstream (-)

= Mean streamwise velocity upstream (m/s)

g = Gravitational acceleration (m/s2)

= Bed elevation at a downstream location, relative to the datum used for (m)

= Water depth downstream (m)

= Coriolis coefficient downstream (-)

= Mean streamwise velocity downstream (m/s)

= Minor losses (m)

= Friction losses (m)

= Energy extracted by turbine (m)

= Energy loss due to turbine blockage (m)

2 Properties to be Measured 9
HK IN OPEN CHANNELS REPORT

Figure 3. Energy grade line between two locations along a channel in open-channel flow, adapted from
Te Chow (1959).

The HGL describes the variation in the water surface upstream, and the potential for overtopping the banks and
flooding. The EGL, with HGL, describes variation of velocity head and velocity over the channel reach upstream
of the turbine. The velocity is an important bulk parameter (section-averaged) that indicates potential for scour or
deposition due to hydraulic changes.

10 2 Properties to be Measured
3 INSTRUMENTATION, DEPLOYMENT AND
MEASUREMENT PROTOCOLS
3.1 Bathymetric mapping recommendations
Bathymetric mapping techniques recommended for a canal site are survey-grade single and multi-beam depth echo
sounders (SBE, MBE) coupled to a global positioning system (GPS) with real-time kinematic (RTK) correction.
An RTK correction consist of a rover receiver, mounted to the survey boat, and a stationary on-shore base station
which provides corrections to the rover receiver to obtain up to centimeter positioning accuracy. The RTK system
is a significant upgrade from the common single-receiver differential GPS system that only has decimeter accuracy.
The RTK GPS is especially recommended for sites, such as a narrow canal, where positioning errors can adversely
affect the study result significantly.

These instruments can be deployed from a

manned or unmanned (remote controlled,
RC) boat (Figure 4). RC boats are much
smaller and have a better maneuvering
capability than manned boats. As such, RC
boats are more suitable for surveying canals
with small widths and low clearance (such
as due to a bridge or ropes across the canal).

It is also safer for the operator because he

(or she) does not need to be present on the
boat when conducting the measurement. An
RC boat has a reasonably long range. For
example, the RC boat that Sandia National
Laboratories used for surveying at the
Roza Canal hydrokinetic site, Yakima, WA,
can be operated remotely from up to one
kilometer away. The RC boat (Z-boat 1800
model (OceanScience 2015)) is capable to
withstand high current speed (up to 5 m/s) Figure 4. The remote control survey boat with real-time
typically observed in MHK sites. kinematic GPS correction.

Bathymetric surveys are best conducted when the water level

is high to maximize data capture, as unsubmerged parts of the
Recommended bathymetric channel cannot be mapped. The bathymetry data are stored as x,
y, z positions, where x and y corresponds to the horizontal position
mapping techniques include obtained from GPS reading and z is the depth as measured by the
using survey-grade single- echo sounder. For consistency, the x, y, z coordinates should be
reported in a standard coordinate reference frame that includes
and multi-beam depth echo
latitude, longitude, and National geodetic vertical datum (NGVD).
sounders. The latter can be determined from the depth measurement and
elevation information provided by the GPS, taking into account the
vertical distance between the echo sounder and GPS receiver.

Surveying software, such as Hydromagic (Eye4Software 2015), is capable to display the x, y, z measurements on
top of a satellite image of the study reach in real-time. This feature improves the mapping efficiency because it
greatly helps distinguish the locations that have been and still need to be mapped.

3 Instrumentation, Deployment and Measurement Protocols 11

HK IN OPEN CHANNELS REPORT

The quality of the bathymetry data correlates highly 3.2 Water level monitoring
with the measurement point density. Generally, the
To understand the influence a turbine has on the
higher the point density, the better the data quality. A
flow field it is important to perform a hydrodynamic
recommended approach for conducting the survey
assessment for baseline and turbine operation
is to follow the lines of the “checkerboard,” starting
conditions. The baseline condition is defined as the
with measuring the outer boundary of the region to be
condition when a turbine is not present in water,
measured. The next step is to measure along multiple
while the turbine operation condition is defined as the
parallel paths within the canal, starting from one
condition when a turbine is submerged and subject
bank and finishing at the other bank, starting from the
to various states of operation. To enable a valid
upstream boundary and finishing at the downstream
comparison between baseline and turbine operating
boundary (or vice versa). Creating parallel paths is
conditions, measurements should be made under
often difficult in practice, especially if the channel is
the same flow conditions. Often flow conditions
reasonably narrow. For narrow channels, a zig-zig
in a canal are relatively steady for up to several
measurement path can be adopted as it is less difficult
hours and sometimes days. The best way to ensure
to perform than creating parallel paths (see Figure
that canal flow conditions vary is little as possible
5). The measurement can be repeated to increase data
between measurements in the presence and absence
density as required. One of the main challenges in
of the turbine is to perform the measurements one
surveying bathymetry in canal sites is controlling the
after the other with as little time as possible between
boat movement, which can be difficult due to high
measurements to avoid changes in the flow conditions.
current speeds in the canal.
The main goal of the water level measurements is to
determine the changes in water level under various
The raw bathymetry can be interpolated using software
states of turbine operation and its extent along the
such as Matlab, Tecplot, ArcGIS or Hydromagic, to
canal.
visualize the overall bathymetry of the study reach.
Alternatively, it can be used directly as input for
Measurements should be conducted at several locations
generating numerical modeling meshes (Figure 5).
upstream and downstream of the turbine location(s) to
meet specific project goals. Within 10 to 20 turbine
diameters from the turbine it is recommended that
measurements are closely spaced (1-5 turbine diameter
spacing) in order to obtain high resolution hydraulic
and energy grade lines. Further upstream and
downstream of the turbine, measurement locations can
be more sparsely spaced to simply determine the extent
of the disturbance caused by the turbine. The flow in
most canals is likely to be subcritical, which means
that a relatively persistent water level rise upstream
of the turbine, proportional to the turbine thrust, will
commonly be observed (i.e. backwater effect). For this
reason, measuring water levels at several locations far
upstream of the device is necessary for quantifying the
extent of the flow disturbance. As an illustration, water
level measurements at the Roza Main Canal site show
that turbine operation continued to affect water levels
= TURBINE up to 700 meters (~233 turbine diameters) upstream
of the turbine (furthest upstream water level sensor as
Figure 5. An example of a coarse bathymetric restricted by a siphon and logger installation) (Figure
survey using a single-beam echo sounder at the 6). A 25 kW vertical-axis hydrokinetic turbine, with a
Roza Canal site showing the satellite image with, no 3-meter diameter and 1.5 meter height was deployed in
data (left), raw data (middle), and the interpolated the Roza Main Canal site when the measurements were
data (right).
taken. Water level increases at 700 m (~233 turbine
diameters) upstream were about 70 percent of increases
measured at only 50 m (~17 diameters) upstream of the
turbine (~2 cm and ~3cm, respectively).

12 3 Instrumentation, Deployment and Measurement Protocols

 While the near bank water level may differ from
Hydrodynamic assessment should the centerline water level, the bank water level is
be performed for both baseline and the primary interest for water operators such as the
USBR and local irrigation districts, because spillways,
turbine operation conditions. Water which release water into neighboring fields to prevent
level measurements should be taken overtopping of the canal, are located at the bank of
the canal. For that reason, measuring water levels at
simultanouesly at several locations
the channel’s sidewall is appropriate for overtopping
upstream and downstream of the assessment. Energy grade line should ideally be
turbine(s). calculated using mean cross-section depths to take into
account depth variations across the channel. This can
be done using multiple pressure transducers mounted
This indicates that the back water effect from the
across the channel. However, installation of additional
turbine deployment reached well upstream for this
mounting hardware on the bottom is often required, and
particular canal geometry. The extent of the flow
this is most easily done when no water is present in the
disturbance upstream of the turbine is influenced by
canal.
flow conditions and the canal geometries. As such,
it can be difficult to optimize the location of water
Measuring continuous water levels across the channel
level sensors to get the most useful information
requires an instrument that is capable of moving across
with the minimum amount of expense (i.e. sensor
the channel, such as an echo sounder mounted on a
and data processing costs, etc.); however numerical
survey boat. Measurements using this method in the
hydrodynamic models can prove useful for this
wake region are often challenging because of the highly
purpose.
turbulent flow and dynamic water surface elevation in
that region.
A common instrument for monitoring water level is a
water level pressure transducer. Pressure transducers
make point measurements, are battery-powered and,
depending on the data sampling rates, can be remotely
deployed for measuring water levels for days or weeks
(Figure 7). Water level could fluctuate significantly
within seconds, therefore the highest sampling rate
should be used whenever possible, to allow accurate
assessment on water level dynamics. The transducer is
commonly deployed at the side wall of a channel, such
as the one shown in Figure 7. One way to deploy the
transducer is by mounting it on a PVC pipe and sliding
the pipe into an angle iron anchored to the channel wall.
The water elevation at the measured location relative to
a datum can be determined by taking into account the
transducer’s depth measurement, the elevation of the
top of the angle iron (which can be measured using a
total station) and the vertical distance between the top
of the angle iron and the transducer location. Figure 6. Water level time series at the Roza Canal
site, at 50 m and 700 m upstream of the turbine.

A water level pressure transducer

can be used to continuously monitor
water levels, and the highest sampling
rate should be used if the levels
fluctuate. To assess overtopping,
water levels should be measured at
the channel’s sidewall.

3 Instrumentation, Deployment and Measurement Protocols 13

HK IN OPEN CHANNELS REPORT

Figure 7. Top left: An example of water level

pressure transducer; the length of the transducer
in picture is 15 cm. Top right: Transducer, deployed High frequency ADCPs are the
using an angle iron anchored to the channel’s side primary instrument for measuring
wall. Bottom: angle iron profiles mounted on the
channel’s side wall.
water velocities and flow discharge in
shallow water environments, such as
3.3 Velocity and turbulence small rivers and canals.
measurements
3.3.1 Acoustic Doppler current profiler environments from small rivers to estuaries. It is
also important to consider ADCP sample volume
Acoustic Doppler current profilers (ADCP) have
size, especially for measuring velocity in areas with
continuously been developed since their inception
steep gradients in velocity (i.e. in the near field wake
in the early ‘80s (Christensen 1982; Gordon 1989;
region of a turbine). The inherent principle of ADCP
Simpson 2001), and are starting to become the primary
measurements is to average velocity measurements
instrument for measuring water velocities and flow
from all transducer beams. Because the transducer
discharge. ADCP’s robustness and capability of
beams leave the ADCP at an angle, the sampling
collecting data over a large region within a relatively
volume size will be higher further from the ADCP.
short time are appealing for MHK measurements.
Smaller sample sizes will better capture velocity
ADCP measurement and data post processing and
variation in flows with higher shear and turbulence.
quality control guidelines specifically for MHK
Data collection software that comes with shallow water
applications are outlined in Neary et al. (2011) and
ADCPs often has a discharge measurement feature,
Gunawan and Neary (2011). Examples of MHK field
which is useful for monitoring the flow discharge and
and laboratory studies involving the use of ADCP
its steadiness in the canal.
measurement are described in Goddijn-Murphy et al.
(2012), Gunawan et al. (2014), Neary et al. (2013), and
ADCPs come with two to nine transducers, but only
Thomson et al. (2012).
those with three or more transducers can measure
3D velocity (Figure 8). Common ADCP deployment
The profiling range over which an ADCP can resolve
methods include mooring and bed mounted for
water velocities depends upon the frequency of the
stationary measurements. Stationary measurement
acoustic signal. Generally, the lower the frequency,
is particularly useful for monitoring the change of
the farther the ADCP can measure through the water
velocity direction and magnitude over time. However,
column. Low frequency ADCPs are used for deep
the measurement is limited to one location. Mooring
water environments, such as in the ocean, while
and mounted ADCP measurement may not be feasible
high frequency ADCPs are used for shallow water
because it modifies the canal’s geometry, which is often

14 3 Instrumentation, Deployment and Measurement Protocols

 pulling the ADCP while it measures velocity and is
Common ADCP deployment methods commonly used for measuring flow discharge in rivers
or canals (Instruments 1996; Simpson 2001). This
for stationary measurements include method is not recommended for hydrokinetic canal
mooring and bed mounted, while the sites because of the high current speeds in such sites.
moving-vessel method can be used The high current speed causes boat instability and
difficulties in controlling its path, which often yields
for floating or moving platforms. poor data and can capsize the boat (Figure 9). This
method is also exhausting for the operators because
prohibited by the water district that manages the canal they have to resist the drag force acting on the boat.
operation. Another common ADCP deployment method
is the moving-vessel method, where the ADCP takes A preferred method is to use a cableway system with a
velocity profile measurements from a floating platform support structure on each bank and a straight-tensioned
(often a boat) while it is moving. The ADCP can be tagline across the canal (e.g. Gunawan et al. (2010)) ,
moved by tethering the ADCP boat or by driving the which can guide the boat through a relatively straight
boat if the boat is equipped with a motor. path and transfer the drag force acting on the boat to the
support structure. An example of a cableway system is
shown in Figure 10, where the white tagline helps resist
the drag force acting on the boat by transferring it to
two rebar’s, one on each bank, that are anchored in the
ground.

Flow discharge in a cross-section should be calculated

based on the standard approach for ADCP discharge
measurement outlined in (Simpson 2001). Four ADCP
transects, each with a discharge difference of less than
5% from the mean discharge of the four transects, are
recommended for determining the discharge at the
measured cross-section (which is the mean discharge of
the four transects itself). When using velocity contours
at different cross-sections for calculating velocity
deficit or creating a wake velocity map, it is critical
to ensure that the flow discharges at different cross-
sections have similar values (conservation of mass).

Figure 8. Different types of ADCP beam

configurations (RDI 2011)

Cross-section velocity mapping is one of the most

important measurements in canal sites as it provides
a base for calculating inflow velocity for assessing
turbine performance and loading; inflow discharge
for determining flow steadiness and numerical model
boundary condition; bulk velocity for calculating
Reynolds and Froude numbers; and wake velocities for
velocity deficit/recovery calculations and numerical
model validation. The simplest method for conducting
a moving vessel (MV) ADCP cross-section (CS)
measurement is by pulling the ADCP vessel back and
forth across the channel using a looped line. This Figure 9. Boat capsize due to highly turbulent flow
method requires one person on each bank holding and and uneven water surface downstream of a turbine.

3 Instrumentation, Deployment and Measurement Protocols 15

HK IN OPEN CHANNELS REPORT

speed will likely be site specific, and is dependent on

A rebar support, anchored to the ground many parameters that can include, mean water speed,
turbulence, suspended sediment concentration, and
ADCP settings. Therefore, ADCP operators are advised
to figure out the optimal traverse speed on their test site
by taking measurements with different traverse speeds,
and comparing the results.

Inflow velocity should be measured at an upstream

cross-section where the velocity dip due to the presence
of the turbine is not observed. The velocity dip
diminishes typically at three or more turbine diameters
upstream of the turbine. Velocity within the turbine
wake begins to recover almost immediately but takes
between 10 to 20 turbine diameters downstream of
the turbine before the velocity reaches 80%-90%
of its original inflow value (Chamorro et al. 2015;
Maganga et al. 2010; Myers and Bahaj 2007; Myers
and Bahaj 2010; Neary et al. 2013). For that reason,
Figure 10. A cableway deployment with a tensioned wake velocity measurements should include the
tagline (the white line), that is attached to the ADCP region within 20 turbine diameters downstream from
boat. the turbine whenever possible. Figure 11 shows
examples of an inflow cross-section velocity contour
When conducting cross-section measurements, the and three wake flow velocity contours measured at
ADCP should be traversed slowly and at a constant three different cross-sections at the Roza Canal site.
speed to reduce noise in the data. Traversing too fast The inflow velocity magnitudes are relatively uniform
could introduce measurement errors, which could across the channel. The wake flow velocity contours
cause inaccurate discharge measurements. These show a deceleration in the middle of the channel due
errors can include invalid bottom tracking (Mueller to the presence of the turbine, whereas accelerations
and Wagner 2009) and bin resolution that is too coarse are observed near the banks. The deceleration and
(Gunawan et al. 2011). At the Roza Canal site, which accelerations are more pronounced immediately
has a typical bulk velocity value of around 2 m/s downstream of the turbine, but gradually diminish with
and a typical water surface width of around 13 m, a increasing distance from the turbine.
traverse speed of 0.05 m/s was used to obtain consistent
single-transect discharge values. The ADCP bin sizes The MV ADCP measurement method essentially
during these measurements were set to 0.2 - 0.25 collects short period (instantaneous) velocity data at
m. For these conditions, traverse speeds of greater each position of the ADCP. As a result, the velocity
than 0.10 m/s typically caused inconsistent single- measurements contain large fluctuations due to flow
transect discharge values. While there are currently no turbulence and instrument noise and, hence, do not
guidelines on optimal ADCP traverse speeds, a general resemble the smooth-time-averaged velocity profiles
rule of thumb is for the boat travel speed to be less generally used for open-channel flow analysis. This
than the water speed (Muller et al. 2013). The optimal method, however, is favorable for most HK-related
measurements due to its capability to achieve data
from a large region within a short time. Data post-
Cross-section velocity mapping is processing, such as spatio-temporal data averaging, is
recommended prior to analyzing the data. Dinehart and
important to calculate inflow velocity,
Burau (2005), Szupiany et al. (2007), Gunawan et al.
inflow discharge, bulk velocity, and (2010), proposed spatio-temporal averaging methods
wake velocities. The preferred method to smooth the velocity contours measured using the
MV method. Traversing the same transect several
of measurement is to use a cableway
times and averaging the measurements at each reduces
system and to traverse the ADCP the fluctuations in velocity profiles, caused by flow
slowly and at a constant speed. turbulence and instrument errors.

16 3 Instrumentation, Deployment and Measurement Protocols

 The moving-vessel ADCP
measurement method is favorable
for most HK-related measurements
because it can achieve data from
a large region within a short time.
However, spatio-temperal data
averaging is recommended prior to
analyzing the data to smooth velocity
contours.

An example of a spatio-temporally averaged wake flow

velocity at 10 m downstream of the turbine, at the Roza
Main Canal site is shown in Figure 12. It is convenient
to always have measured data along a cross-section.
However, this is nearly impossible to achieve in the
field. In practice the locations of the measured data
will always have some deviations from a straight
path. The spatio-temporal averaging process generally Figure 12. Spatiotemporally-averaged velocity
involve interpolating/extrapolating the ADCP data to a contour (looking downstream), 10 m downstream
representative cross-section that will be used for further of the turbine, at Roza Canal under high tip-speed
analysis, such as for numerical model comparison. ratio. Flow discharge was approximately 55 m3/s.
The black-colored scatters on the top figure show the Black scatters in the top figure indicate the locations
locations of the measurements obtained from four of velocity measurements.
ADCP transects. The contours in both the top and
bottom figures are the results of extrapolating the ADCP MV measurements can be conducted together
measured velocities to a representative cross-section. with the bathymetric survey, to obtain quasi-
instantaneous velocity data along the surveyed reach.
As with bathymetric surveys, a cable way or manned
/ remote control survey boat is required. A GPS is
required to obtain good accuracy of ADCP positions
during the survey, and is a significant upgrade from
the ADCP bottom tracking capability. The MV ADCP
survey is an efficient method to measure velocities
over a large region. However, the boat speed has to be
relatively slow and the boat operation has to be smooth
(gradual accelerations and decelerations, maintaining
a uniform speed, and slow turns) when measuring
velocity using this method (Muller et al., 2013). High
boat speed generally introduces more errors in MV
ADCP measurements, as shown, for example, in
Gunawan et al. (2016). As stated previously, a good
rule of thumb, recommended by Muller et al. (2013),
is to maintain the boat speed equal or less than the
average water speed. An example of the MV ADCP
survey results are shown in Figure 13.

Figure 11. ADCP velocity contours (looking downstream), as shown in the RD Instrument’s WinRiver
software, at four cross-sections at the Roza Canal test site: 50 m upstream of the turbine, 10 m downstream
of the turbine, 20 m downstream of the turbine, and 30 m downstream of the turbine. The x axis shows the
length of the ADCP travel path while traversing the canal.

3 Instrumentation, Deployment and Measurement Protocols 17

HK IN OPEN CHANNELS REPORT

Figure 13. Streamwise velocity contours, ADV inflow measurements should ideally be taken at a
interpolated from MV ADCP survey. Left: near- distance greater than three turbine diameters upstream
surface velocities, right: hub-level velocities. The of the turbine for the same reasons described in the
white circle indicates the locations of the Instream ADCP section. An ADV is commonly deployed using
turbine. a stationary tower or tripod mounted on the channel
or ocean bed. Mounting a structure on the canal’s bed
3.3.2 Acoustic Doppler velocimeter alters the canal’s geometry altogether, which is often
Acoustic Doppler velocimeter (ADV) measurements prohibited by the water district that manages the canal
provide higher order information than ADCP to more operation. Even though bed deployment is permitted,
accurately calculate turbulence statistics. To achieve towers and tripods obstruct a greater percentage of the
this, ADV’s use a smaller sampling volume and higher flow section compared to large river or tidal channels
sampling frequency. Higher order turbulence statistics and care must be taken when considering stationary
are typically required for parameterizing high- and mid- deployment so as not to create vortex shedding and
fidelity CFD models, and can be valuable for validating flow alteration within the sample volume of the ADV.
the results, e.g. turbulence kinetic energy at different
locations upstream and downstream the turbine.
ADVs use a smaller sampling
These higher order measurements of instantaneous volume and higher sampling
velocity are also useful for investigating the interaction frequency to calculate higher
between the flow turbulence, wake development and
order turbulence statistics. ADV
power generation, as well as for resource assessment
(Gunawan et al. 2014). Readers interested in the deploymentdeployment methods
details of ADV measurement principles and data post include mounting a stationary tower
processing and quality control guidelines specifically
for MHK applications are referred to (Gunawan et al. or tripod on the channel or using a
2011). cable system.

18 3 Instrumentation, Deployment and Measurement Protocols

The ADV sampling volume, whenever feasible, should
be placed at the turbine hub height or rotor mid height
level. This ADV position is anticipated to provide
the most representative inflow to the turbine and the
maximum deficit in the wake of the turbine. Further, it
is beneficial to cover the ADV stem (mounting pole)
with a hydrofoil to minimize drag on the ADV support
structure. This helps in reducing the vibration of the
measurement probe and lessen flow diversions to the
location of the sampling volume; both introduce error
in the velocity measurement.

Another ADV deployment method suitable for canal Figure 15. Sandia’s mobile ADV deployment system
measurements is cable deployment. An example for canal measurements. On top is the schematic of
of a cable-deployed ADV using a sounding weight, the system and a picture of the unit deployed in the
deployed from a boat, is outlined in Holmes and field is shown on the bottom.
Garcia (2008) (Figure 14). If the width of the canal
is relatively small, a cableway system, mounted at the 3.4 Power measurement
canal edges, may be used to deploy the ADV (Figure Turbine power time series for assessing power
15). The cabled ADV was mounted to two tensioned- production performance can be obtained from generator
steel wires using an aluminum plate. The system is output or by measuring the rotor’s mechanical torque
designed to be able to move the ADV laterally and and angular velocity. Common methods for torque
vertically, and significantly withstand the water flow- measurements include mounting a strain gauge at the
induced drag force to up to 1-meter depth below the shaft. The technical challenges for this method are
water surface. It should be noted that the ADV will supplying power to the strain gauge through the rotating
likely move (swing) during cable deployment due to shaft and sending the signal out of the rotating shaft.
the flow drag and turbulence. If the errors caused by Slip rings can be used to interface to the rotating shaft,
the ADV motions are significant, the measurement may but they are susceptible to wear and need to be replaced
need to be discarded. Accounting for flow induced over time. A non-contact method for transferring the
motion of the ADV on the velocity measurements is power into and signal out of the shaft eliminates the
an ongoing research topic. Several researchers have needs of a slip ring. This method typically uses radio
suggested correction methods and performed limited telemetry and wireless induced power and an example
testing using the methods (Thomson et al., 2015, can be seen in Figure 16. The system in this example
Durgesh et al., 2014, Neary et al., 2012). Readers consists of a stationary power coil and a rotating collar
interested in learning more about these correction that needs to be mounted on the shaft. A torque strain
methods are advised to review these references. gauge is mounted to the shaft and connected to the
rotating collar. Wireless power and data transmission
occur in the small gap between the stationary coil
and rotating collar. The size of the gap is maintained
constant while the collar is rotating with the shaft. The
collar is equipped with six magnets that can be tracked
by the magnet sensor located in the master control unit
to enable angular velocity measurement of the shaft.

If only temporary torque measurement is required, a

battery-powered wireless system can be used (Figure
17). The torque measurement principle of the battery-
powered (BP) system is similar to the wireless collar
system. However, a power coil is not needed for the BP
system, which reduces the complexity of the system
installation significantly. The BP system consists of a
Figure 14. Cable-deployed ADVs (in circles) with radio transmitter that can be mounted to the shaft using
sounding weight (Photograph courtesy of Bob
a strong tape, such as a fiberglass one, and a receiver
Holmes, USGS, 2010).

3 Instrumentation, Deployment and Measurement Protocols 19

HK IN OPEN CHANNELS REPORT

A rotor’s mechanical troque can

be measured using strain gauges
mounted at the shaft

for receiving the measurement signal. A torque strain

gauge is mounted to the shaft and connected to the
radio transmitter. A single 9V battery will typically last
for a few hours of measurements. Multiple batteries
allow longer measurements to be performed, but may
not be practical because all of the batteries need to be
mounted to the shaft.
Figure 18. An example of a turbine performance
An example of a turbine performance curve of curve for different tip-speed ratios.
the Instream Energy System turbine, tested at the
Roza Canal, is shown in Figure 18. The curve was
determined using generator output and inflow velocity 3.5 Drag measurement
measurements. Note that Cp and l are normalized
to maximum values of Cp and l respectively Instruments commonly used for measuring turbine
(Cpmax and lmax). drag are load cells and strain gauges. One of the
main challenges for measuring turbine drag is finding
appropriate locations for mounting the load cells or
strain gauges. For measuring drag on a wind turbine,
an often used mounting location for a strain gauge is
near the base of the structure where maximum strain
is present, as shown in Figure 19. The strain value at
the base can be used to determine the drag forces at the
turbine and tower using the bending moment equation,
provided the material and geometric properties of the
structure are known. The same method can be applied
for calculating drag force on hydrokinetic turbines. For
example, the bending moment at a position at a cylinder
Figure 16. The torque sensor system mounted to strut relates to the strain value at that position through
the Instream’s turbine shaft at Roza Canal. The the equation
small brown rectangular sensor to the left of the
torque sensor system is a strain gauge used for drag
measurements.

where ε is bending strain (microstrain), M is bending

moment (inch-pounds), E is the Modulus of Elasticity
of the strut (pounds per square inch) and r is the strut
diameter (inch). The drag force can be determined
from the moment if the length of the moment arm is
known using analytical or numerical methods. For
complex structures and heterogeneous structure
materials, Finite Element Analysis may be required for
determining the drag force.

Figure 17. An example of a battery-powered

wireless strain sensor system, for torque or thrust
measurement (www.binsfeld.com).

20 3 Instrumentation, Deployment and Measurement Protocols

Load cells generally need to be coupled with the
turbine support structure to be able to measure turbine
drag. The support structure may need to be modified
for this purpose. For this reason, using a load cell may
be impractical if the structure is not designed or easily
modified for accommodating a load cell.

Figure 19. Strain gauge mounting location for

measuring turbine drag on cross-flow wind turbine
(Griffith et al. 2011).

3 Instrumentation, Deployment and Measurement Protocols 21

HK IN OPEN CHANNELS REPORT

4 Predicting the effects of HK turbine

deployment using numerical modeling
Developers generally aim to deploy multiple turbines at a site to improve the levelized cost of energy of a current
energy conversion project. Prior to deploying multiple turbines, regulators and other stakeholders often request
that the expected effects to the water system are generally understood and quantified using a numerical model
simulation. High-fidelity CFD models, such as StarCCM+ and Fluent, are capable of accurately simulating flow
around turbine structures, however, they are computationally demanding and, therefore, generally limited to
computational domains that extend just several turbine diameters from the device.

One-dimensional models, such as HEC-RAS or ISIS 1D, predict the variation of the section averaged water level
and velocity along the longitudinal or streamwise length of the channel reach and can approximate hydraulic
changes due to turbine deployment with existing software functionality. One-dimensional hydrodynamic models
are most beneficial when operational impacts (mainly raised water levels) are the main concern and can predict the
quantity and locations of HK installations as an initial estimation over a large domain. Feasibility level results can
be obtained faster and cheaper compared to more complex modeling tools and generally do not require a numerical
modeling expert to run the simulation. One of the limitations of 1D models is that they are not able to predict the
variation of water levels and velocity over the cross-section; only the mean cross-section water level and velocity
are predicted. As a consequence, near-field velocity changes such as convective acceleration around a turbine and
flow recovery in the wake of a turbine are not predicted. Further, they may not predict overtopping as accurately as
2D and 3D models; especially at bends in the canal where super-elevation effects cause the water surface elevation
along a cross-section to tilt, with higher water surface elevations on the outer (convex) bank.

Multi-dimensional hydrodynamic models, e.g., such

as SNL-Delft3D-CEC, allow a compromise between
computational cost and accuracy. These models provide Numerical models can predict the
3D simulations over large domains to capture both hydrodynamic effects of blockage and
near- and far-field effects of HK turbine deployments drag caused by hydrokinetic turbines
(Sandia National Laboratories 2016). This enhanced
version of the base code Delft3D (Deltares 2014) and assess optimal device spacing
includes code development to accurately simulate to maximize energy production and
the 3D flow-field changes induced by HK turbines,
minimize potential impacts, including
accounting for turbine induced drag, momentum loss,
and turbulence production and dissipation. Although local water level rise, scour, and
these models presently do not accurately simulate deposition.
velocity recovery very near the turbine (within 3-5
diameters), they strike a good balance between ease
of use, computational efficiency, and accuracy that allows them to simultaneously determine optimal locations for
turbine deployment to maximize array power production and minimize unwanted changes to the flow field that
could affect other human uses and aquatic life.

Numerical modeling of hydrokinetic turbines requires measurement datasets for model input and calibration/
validation to improve confidence in their use. Measurement variables useful for calibration and validation include
water level (hydraulic grade line, HGL), energy grade line (EGL), water surface slope, energy slope, velocity,
turbulence parameters (turbulence intensity, turbulence kinetic energy), flow discharge, and turbine thrust (drag).
Measurements of these parameters (or a subset) around one or more turbines aid in site model calibration and
validation. At a minimum, model validation for a single turbine deployment is recommended prior to modeling the
effects of a multiple turbine deployment.

An example numerical model result using SNL-Delft3D-CEC for flow in the Roza Main Canal in the presence of
a single turbine is shown in Figure 20. Illustrated is the increase in flow speed around the turbine with a decrease
and trend towards recovery behind the turbine. Approximately 12 diameters downstream of the turbine, the canal

22 4 Predicting the effects of HK turbine deployment using numerical modeling

boundary surface transitions from concrete lining to
earthen lining (unlined), and nearly doubles in width. Water level, energy grade line
Geometry transitions increase the challenges for in-situ
water surface slope, energy slope,
field data collection and numerical modeling. It also
complicates the assessment of flow recovery in the lee of velocity, turbulence parameters, flow
the turbine. A common calibration/validation approach discharge, and turbine drag are all
for MHK turbine simulations is to compare measured
and simulated centerline hub/mid rotor height wake
measurement variables useful for
velocities, as shown in the example in Figure 21. Note calibrating and validating numeric
that the measurement data shown was obtained from models.
MV ADCP measurements (interpolated), which include
several along cross-section MV measurements, and a
survey measurement in which the ADCP/survey boat was moved in a zig-zag pattern over the measurement area.
This is best done in a straight section of channel to minimize aforementioned super-elevation effects. Results from
the model compare well with measurements from the collected field data of velocity recovery behind the deployed
turbine starting 4 diameters in the wake of the turbine. To maximize power production, it is best to deploy turbines
far enough apart such that velocity recovery and rate of recovery from the upstream turbine to the downstream
turbine has reached a balance. This often occurs within 7-20 diameters downstream of the upstream turbine
(Myers and Bahaj 2010, Stallard et. Al 2013, Chamorro et. Al 2015). Although SNL-Delft3D-CEC doesn’t capture
the very near field-field velocity deficit (3-5 diameters) it is still well suited for determining the optimal spacing
requirements.

Figure 20. Model-predicted velocity contours for Roza Main Canal site (at turbine mid-span, flow is from top
to bottom). The left figure shows the whole simulation domain, with a turbine positioned near the outflow
boundary. The right figure is the velocity contour at the turbine location. Legend units are in meter/second.

Numerical simulations also need to be able to predict the Hydraulic and Energy Grade Lines (HGL and EGL)
accurately, with and without the turbine’s presence, because they are main parameters for quantifying HK
deployment effects. Water level and velocity measurements can be used for assessing the model’s performance
in this aspect. The predicted changes of water level and bulk (cross-section averaged) velocity along the reach
without the turbine and with the turbine in place can be assessed against measurements. The same measurements
can also be used for validating the water surface and energy slope predicted by the model. For example, the
model predicted HGLs and EGLs at the Roza Main Canal for baseline (without turbine’s presence) and in the
turbine’s presence are shown in Figure 22. The model predicts that adding a turbine at the present location
increases water levels by around 6 centimeters immediately upstream of the turbine and around 3 centimeters at
700 m (~233 diameters) upstream of the turbine; these values are close to the 4 and 3 centimeters obtained from

4 Predicting the effects of HK turbine deployment using numerical modeling 23

HK IN OPEN CHANNELS REPORT

measurements respectively. In addition, the predicted water surface slope along the reach also agrees well with
the measured value. Once the single turbine model is calibrated and validated against measurements (e.g. thrust
(drag) coefficient), the turbine calibration parameters can be used for multiple-turbine simulation cases, for project
optimization.

Figure 21. Measured and simulated wake velocities at turbine mid-span at Roza Main Canal, as a function of
distance from the turbine.

Figure 22. Simulated Energy Grade Lines and Hydraulic Grade Lines (water surface elevations along the
reach) at Roza Main Canal with and without the turbine’s presence.

24 4 Predicting the effects of HK turbine deployment using numerical modeling

5 Summary
Although a detailed hydrokinetic canal resource assessment does not exist at this time, it is clear that some
existing canal systems contain favorable flow speeds, water depths, and other characteristics that are desirable for
hydrokinetic power development. Feasibility studies for HK projects must demonstrate that HK operations will
not adversely affect water operations. Potential concerns include disrupting water supply operations (by affecting
head-discharge conditions at irrigation canal intakes), flooding (when turbine deployment and operation increase
water levels above acceptable limits), reduction of power generation in nearby hydropower plant (by affecting
the water levels and net head at hydropower dam or discharge), and unfavorable erosion-deposition patterns.
Carefully thought out physical and numerical tests can be used to address these issues and determine the viability
of a particular canal or section of a canal for various amounts of HK energy development. This document should be
viewed as a starting point to identify the parameters to be measured, instrumentation and deployment strategies for
their measurements, and protocols for acquiring and processing data, for HK flow effects and turbine performance
assessments within canals (and potentially small rivers).

To help estimate the level of effort required for a hydrokinetic assessment and impact study, Table 1 summarizes
required data and potential methods for collecting that data. Table 1 assumes a full hydrodynamic impact study
with an operational HK turbine installed. This effort could take a few weeks to several months depending on
scheduled discharges, site conditions, and HK turbine design. For example, an HK installation that can hoist
its own turbine in and out of the flow can be compared to the exact same baseline conditions on the same day
and testing could be completed within a matter of days for a given flow condition. In contrast, a HK design that
requires a flow shutdown to install or remove the unit may require months to acquire sufficient baseline data and
several months after the installation for comparison to HK operational data under similar flow conditions.
For a feasibility level study, bathymetry, discharge, and water level data at key locations of the channel (transitions,
bifurcations, siphons, flow measurement structures, etc.) are sufficient to help determine if HK deployment is
an option. This information is often available from owners and operators of the canal and can be verified using
methods identified in Table 1. As previously mentioned, these data can be used with numerical models to support
feasibility studies.

Table 1 Recommended measurements for the assessment of potential impacts from

open-channel HK operations
Table 1 Recommended
Measurementmeasurements for the assessment of potential impacts from open-channel HK
Instrument
No
operations Locations Main Purpose
Parameters Example

1 Bathymetry (or Along the channel, within Determine channel echo

Note geometry
that the measurements
for lined listed
20 -in
30table 1 are also
diameter fromrequired
the forbed
numerical modelor
elevations, input, and HGL and EGL
sounder
calculations.channels) turbine, and far upstream verify the existing and
of the turbine, at the bathymetry data (as- remotely-
same locations with far built geometry data controlled
upstream water level can suffice for lined survey boat
measurements channels if verified by
a field survey)

2 Water level Cross sections Determine impact water level

immediately upstream on water level at logger
and downstream of locations adjacent to
the turbine, e.g. every the turbine, where
diameter within 5 significant difference
diameters from the from baseline
turbine, and every 3-5 (without HK) is often
diameters between 5 to expected
20 diameters from the
turbine.

5 Summary 25
HK IN OPEN CHANNELS REPORT

Measurement Instrument
No Locations Main Purpose
Parameters Example

3 Water level Cross sections far Determine impact water level

upstream of the turbine, on water level at logger
e.g. at -100, -200 and -300 far upstream of the
x/D turbine. Impact
at far upstream is
typically expected for
open channels with
subcritical flow.

4 Downstream Cross sections every 1 Determine local ADCP

local velocity or 2 diameters up to 5 velocity variations
measurement over diameters downstream; downstream of the
entire cross-section every 2 to 5 diameters turbine, where high
between 5 and 20 velocity gradients
diameters downstream are expected. This
information is useful
for turbine array
design and erosion/
deposition/scouring/
silting analysis.

5 Upstream Cross sections at 5 and 10 Determine inflow ADCP

local velocity diameters upstream velocity for
measurement over establishing turbine
entire cross-section performance curves,
as well as velocity
gradients.

6 Upstream and Ideally at the same cross Determine turbulence ADV

downstream velocity sections as the upstream level and unsteady
and turbulence, at and downstream ADCP coherent structures
a high sampling measurements, at turbine on the flow. This
resolution centerline. A minimum of information is useful
3 locations downstream for identifying and
and one location upstream quantifying cyclical
(between 5 to 10 diameter load on the turbine,
upstream) is required for and is a critical
numerical model input. numerical model
input for accurately
predicting wake
profiles.

Note that the measurements listed in table 1 are also required for numerical model input, and HGL and EGL
calculations.

26 5 Summary
6 References
Chamorro, L. P., Hill, C., Neary, V. S., Gunawan, B., Arndt, R. E. A., and Sotiropoulos, F. (2015). “Effects of
energetic coherent motions on the power and wake of an axial-flow turbine.” Physics of Fluids, 27(5), 055104.

Christensen, J. L. H., L.E. (1982). “Mississippi River test “ Vol. 1, U.S. Geological Survey, El Cajon, CA.

Defne, Z., Haas, K. A., Fritz, H. M., Jiang, L. D., French, S. P., Shi, X., Smith, B. T., Neary, V. S., and Stewart,
K. M. (2012). “National geodatabase of tidal stream power resource in USA.” Renewable and Sustainable Energy
Reviews, 16(5), 3326-3338.

Deltares (2014). “Delft3D: Hydro-Morphodynamics.” Deltares: Delft, Netherlands. p. 712.

Dinehart, R. L., and Burau, J. R. (2005). “Averaged indicators of secondary flow in repeated acoustic Doppler
current profiler crossings of bends.” Water Resour Res, 41(9).

Durgesh,V., Thomson, J., Richmond, M.C., Polagye,B. (2014) “Noise correction of turbulent spectra obtained from
acoustic doppler velocimeters.” Flow Measurement and Instrumentation, Volume 37, June, Pages 29-41.

EPRI (2011). “Mapping and Assessment of the United States Ocean Wave Energy Resource. .” R. b. E. P. R. I. E. a.
N. R. E. L. (NREL). ed., pp 176.

EPRI (2012). Assessment and mapping of the riverine hydrokinetic resource in the continental United States. No.
1026880. Electric Power Research Institute (EPRI).

Eye4Software (2015). “Hydromagic: Professional hydrographic survey software.” Software manual.

Goddijn-Murphy, L., Woolf, D. K., and Easton, M. C. (2012). “Current Patterns in the Inner Sound (Pentland Firth)
from Underway ADCP Data.” Journal of Atmospheric and Oceanic Technology, 30(1), 96-111.

Gordon, R. L. (1989). “Acoustic Measurement of River Discharge.” Journal of Hydraulic Engineering-ASCE

115(7):925-936.

Griffith, D. T., Mayes, R., and Hunter, P. (2011). “Excitation Methods for a 60 kW Vertical Axis Wind Turbine.”
Structural Dynamics and Renewable Energy, Volume 1, T. Proulx, ed., Springer New York, 329-338.

Gunawan, B., Neary, V., Roberts, J., Dallman, A., Grovue, S., Mortensen, J., and Heiner, B. (2014). “Field
measurement test plan to determine effects of hydrokinetic turbine deployment on canal test site in Yakima, WA,
USA.” Proceedings of the 2nd Marine Energy Technology Symposium (METS2014) Seattle, WA.

Gunawan, B., and Neary, V. S. (2011). “ORNL ADCP Post processing guide and Matlab algorithms for MHK site
flow and turbulence analysis.” ORNL/TML-2011/404.

Gunawan, B., Neary, V. S., and Colby, J. (2014). “Tidal energy site resource assessment in the East River tidal
strait, near Roosevelt Island, New York, New York.” Renewable Energy, 71, 509-517.

Gunawan, B., Neary, V. S., Hill, C., and Chamorro, L. (2011). “Accuracy of spatial and temporal averaging of
acoustic Doppler current profiler (ADCP) moving boat measurements.” American Geophysical Union Fall Meeting,
San Francisco, CA.

Gunawan, B., Neary, V.S. and Hill, C. (2016) “Comparison of fixed and moving vessel ADCP measurements in a
large laboratory flume”. Journal of Hydraulic Engineering (in press).

6 References 27
HK IN OPEN CHANNELS REPORT

Gunawan, B., Neary, V. S., and McNutt, J. R. (2011). “ORNL ADV Post processing guide and Matlab algorithms
for MHK site flow and turbulence analysis.” ORNL/TML-2011/338.

Gunawan, B., Sterling, M., and Knight, D. W. (2010). “Using an acoustic Doppler current profiler in a small river.”
Water Environ J, 24(2), 147-158.

Hill, C., Neary, V. S., Gunawan, B., Guala, M., and Sotiropoulos, F. (2014). “U.S. Department of Energy Reference
Model Program, RM1: Experimental Results.” Sandia Report, ed. No. SAND2014-18783R. Sandia National
Laboratories (SNL-NM), Albuquerque, NM (United States).

Hill, C., Neary, V. S., Gunawan, B., Guala, M., and Sotiropoulos, F. (2014). “U.S. Department of Energy Reference
Model Program, RM2: Experimental Results.” Sandia Report, ed. No. SAND2014-16561 R. Sandia National
Laboratories (SNL-NM), Albuquerque, NM (United States).

Holmes, R. R., and Garcia, M. H. (2008). “Flow over bedforms in a large sand-bed river: A field investigation.”
Journal of Hydraulic Research, 46(3), 322-333.

Instruments, R. (1996). “Acoustic Doppler Current Profiler: Principles of operation, a practical primer.” Technical
manual.

Maganga, F., Germain, G., King, J., Pinon, G., and Rivoalen, E. (2010). “Experimental characterisation of flow
effects on marine current turbine behaviour and on its wake properties.” IET Renewable Power Generation, 4(6),
498-509.

Mueller, D. S., and Wagner, C. R. (2009). “Measuring Discharge with Acoustic Doppler Current Profilers from a
Moving-Boat.”

Mueller, D.S., Wagner, C.R., Rehmel, M.S., Oberg, K.A,, and Rainville, Francois, (2013), “Measuring discharge
with acoustic Doppler current profilers from a moving boat” (ver. 2.0, December 2013): U.S. Geological Survey
Techniques and Methods, book 3, chap. A22, 95 p. http://dx.doi.org/10.3133/tm3A22.

Myers, L., and Bahaj, A. S. (2007). “Wake studies of a 1/30th scale horizontal axis marine current turbine.”
Ocean Engineering, 34(5-6), 758-762.

Myers, L. E., and Bahaj, A. S. (2010). “Experimental analysis of the flow field around horizontal axis tidal turbines
by use of scale mesh disk rotor simulators.” Ocean Engineering, 37(2-3), 218-227.

Neary, V. S., Gunawan, B., Hill, C., and Chamorro, L. P. (2013). “Near and far field flow disturbances induced by
model hydrokinetic turbine: ADV and ADP comparison.” Renewable Energy, 60(0), 1-6.

Neary, V. S., Gunawan, B., Polagye, B., Thomson, J., Richmond, M. C., Durgesh, V., Muste, M., and Fontaine, A.
(2011). “Field measurements at river and tidal current sites for hydrokinetic energy development: Best practices
manual.” ORNL/TML-2011/419.

Neary, V.S., Gunawan, B., Mellin, A. and Smith, B.T (2012) Sensor-fusion technology for accurate measurements
in deep water environments. Invention Disclosure 201202876, S-124,449.

OceanScience, T. (2015). “Z-Boat 1800 Remote hydrographic survey boat.” Brochure.

Sandia National Laboratories (2016) “SNL-Delft3D-CEC.” Web article (http://energy.sandia.gov/energy/

renewable-energy/water-power/market-acceleration-deployment/snl-delft3d-cec/). Last accessed: 12/19/2016.

28 6 References
Simpson, M. R. (2001). “Discharge Measurements Using a Broad-Band Acoustic Doppler Current Profiler “ open-
file report 01-1., S. United States Geological Survey, ed.

Stallard, T., Collings, R., Feng, T., Whelan, J. (2013) “ Interactions between tidal turbine wakes: experimental
study of a group of three-bladed rotors.” Phil. Trans. R. Soc. A 2013 371 20120159; DOI: 10.1098/rsta.2012.0159.
Published 14 January 2013

Sturm, T. W. (2001). Open channel hydraulics.

Szupiany, R. N., Amsler, M. L., Best, J. L., and Parsons, D. R. (2007). “Comparison of fixed- and moving-vessel
flow measurements with an aDp in a large river.” ASCE Journal of Hydraulic Engineering, 133(12), 1299-1309.

Te Chow, V. (1959). Open-channel hydraulics, McGraw-Hill.

Thomson, J., Polagye, B., Durgesh, V., and Richmond, M. C. (2012). “Measurements of Turbulence at Two Tidal
Energy Sites in Puget Sound, WA.” IEEE Journal of Oceanic Engineering, 37(3), 363-374.

Thomson, J., Talbert, J., Klerk, A. D., Zippel, S., Guerra., M. and Kilcher, L. (2015) Turbulence measurements
from moving platforms, 2015 IEEE/OES Eleveth Current, Waves and Turbulence Measurement (CWTM), St.
Petersburg, FL, 2015, pp. 1-5.

US Army Corps of Engineer (1994) “Channel stability assessment for flood control projects.” EM 1110-2-1418, 31
October 1994 (retreived from http://www.publications.usace.army.mil/Portals/76/Publications/EngineerManuals/
EM_1110-2-1418.pdf in December 8, 2016).

Yang, X., Haas, K. A., Fritz, H. M., French, S. P., Shi, X., Neary, V. S., and Gunawan, B. (2015). “National
geodatabase of ocean current power resource in USA.” Renewable and Sustainable Energy Reviews, 44(0), 496-507.

6 References 29
For more information, visit: energy.gov/eere

DOE/EE-1537 • March 2017