The Oscillating Water Column

AC 2008-1054: AN OCEAN ENERGY PROJECT: THE OSCILLATING WATER COLUMN

Craig Somerton, Michigan State University CRAIG W. SOMERTON Craig W. Somerton is an Associate Professor and Associate Chair of the Undergraduate Program for Mechanical Engineering at Michigan State University. He teaches in the area of thermal engineering including thermodynamics, heat transfer, and thermal design. He also teaches the capstone design course for the department. Dr. Somerton has research interests in computer design of thermal systems, transport phenomena in porous media, and application of continuous quality improvement principles to engineering education. He received his B.S. in 1976, his M.S. in 1979, and his Ph.D. in 1982, all in engineering from UCLA. Page 13.197.1

Introduction Though an important alternative energy source, the topic of ocean energy in an alternative energy class often receives less attention than some of the more popular alternative energy technologies such as solar, wind, fuel cells, and biofuels. To address this concern in the alternative energy course taught at Michigan State University, an ocean energy project is assigned. This paper presents one of those projects, the oscillating water column (OWC).

ME 417 Design of Alternative Energy Systems is a senior level design intensive elective course [1]. It is a project based course for which the students carry out three technical projects. In its last offering the projects were: design of a fuel cell system, design of a wave energy system, and design of a solar energy system. The focus of the course is for students to use simple engineering principles in developing predictive models for alternative energy systems. Some of the projects require students to develop their own calculation tool (a spreadsheet or MATLAB program), while other projects use commercially available software, for example spreadsheet programs from RETScreen International [2], which is managed under the leadership and ongoing financial support of Natural Resources Canada’s (NRCan) CANMET Energy Technology Centre - Varennes (CETC-Varennes). However, for some topics in the course, such as ocean energy, software is not readily available. The OWC project presented in this paper uses an in-house MATLAB program that allows the students to perform design studies with respect to power production and energy costs for several design parameters, including location and size.

This paper continues with an overview on ocean energy, along with details of the various wave energy systems. The particulars on the thermodynamic models, the project statement, and some typical design analysis results are then be presented. Next, student feedback on the project will be reviewed. Lessons learned conclude the paper

Ocean Energy Oceans cover approximately 71% of the earth’s surface [3] and are the largest energy reservoir on the earth’s surface. The United States has a general coastline length of 12,000 miles [4], so there is adequate opportunity to interact with these vast energy reservoirs. One may argue that there are four different forms of ocean energy: tidal energy, current energy, thermal gradient energy, and wave energy. The first three will be briefly reviewed with wave energy being the focus of this section.

Tides are created by the gravitational pull of the moon and the sun. Coastal tide heights can very significantly due to local geographical features such as bays and inlets. Current technology requires a five meter difference between high and low tides for the production of electricity [5]. It is estimated that worldwide there are about 40 such sites. Locations for some of the sites with the highest tides is provided in Table 1

Page 13.197.2 Table 1 World’s Highest Tides [6]

Location Difference between low tide and high tide Bay of Saint-Malo, France 12 m Bay of Fundy, Canada 15 m Puerto Gallegos, Argentina 13.2m Mont Saint Michel, France 12.3m Bristol, United Kingdom 14.6m Bhaunagan, India 12.2m Derby, Australia 11.8 m Cook Inlet, Alaska 11.5 m

Two main technologies have been proposed and implemented to harness tidal energy. Dams and barrages capture sea water in a reservoir during the rise to high tide. The reservoir sea water is then released through water turbines as the tide falls. Hence, they function much like a conventional hydroelectric inland dam system. The largest tidal power station in the world (and the only one in Europe) is in the Rance estuary in northern France, built in 1966 [7]. It has a peak rating of 240 Megawatts for its 24 turbines and an annual output of about 600 million kWh resulting in an average power 68 MW. Since 1984, Nova Scotia Power has been generating up to 20 megawatts of power at a tidal barrage in the Bay of Fundy [8].

Tidal turbines operate underwater much in the same way as wind turbine farms. In December 2006, Verdant Power installed two underwater turbines in New York’s East River [9]. Each turbine can capture up to 35 kilowatts of power from the river’s tidal currents. Since the initial installation, four additional turbines have been added.

Ocean current energy is analogous to wind energy and, similar to wind energy, is also due to the different rates of natural convection in the ocean arising through different solar heating rates because of the distance from the sun. A typical ocean current map is shown in Figure 1.

Figure 1 Ocean Current Map of the Northwest Atlantic Ocean [10] / Page 13.197.3 The energy of ocean currents may also be harnessed with the use of underwater turbines that have an advantage over tidal turbines due to more consistent and less periodic flow rates.

Thermal energy in the ocean comes from the absorption of solar radiation. This solar energy inflow will heat the surface of the ocean, but does not penetrate very deep, and hence gives rise to a temperature gradient between the water on the surface and that at depth. Very simply, the ocean may be considered as having two heat reservoirs, a high temperature heat reservoir at the surface and a low temperature heat reservoir at some depth. These heat reservoirs can be utilized to run a heat engine operating between them, so as to produce power. This approach is called Ocean Thermal Energy Conversion (OTEC). It is schematically shown in the Fig. 2

Ocean Surface (High Temperature Heat Reservoir) at T H

Heat Engine Wnet

Ocean Depth (Low Temperature Heat Reservoir) at T L

Figure 2 OTEC Interaction Diagram

The potential of OTEC can be seen in the map of Fig. 3 that indicates the temperature difference available in various locations.

Figure 3 OTEC Potential [11] Page 13.197.4 Waves are primarily produced by the wind, so wave energy is actually a version of solar energy. Figure 4 shows worldwide wave power in terms of kW per meter of wave width.

Figure 4: Wave Energy Potential [12]

Three strategies have been developed in an attempt to harness wave energy. An onshore reservoir approach is used in the tapered channel system that accelerates the incoming wave by decreasing the flow area, so that it can fill an elevated reservoir. The reservoir water is then released back to sea level, passing through a turbine to extract energy. Mechanical surface systems use the wave motion to move a solid object. The simplest would be a buoy system that rises and falls with the waves and whose tethers turn a generator. The third strategy is the oscillating water column. An oscillating water column system behaves very much like a piston- cylinder system, with the ocean surface acting as the piston, which will compress or expand the air or sea water within the cylinder (or column) as waves rise and fall. The modeling of the oscillating water column system is discussed in detail below.

Modeling of Oscillating Water Column For the course project, a series of simple thermodynamic models are used to simulate the operation of the oscillating water column. A two chamber model is used. As shown in Fig. 5, when a wave rises up through the open chamber, its air is compressed and forced through a turbine into the closed chamber, pressurizing it. When the wave falls, the pressure in the open chamber reduces allowing the closed chamber air to return to the open chamber. A five process thermodynamic model is used to simulate this operation.

Process 1 Polytropic Compression of Open Chamber as a Closed System Process 2 Isentropic Expansion through the Turbine from Open to Closed Chamber Isotropic, Adiabatic Flow from the Open Chamber as a Transient System Isotropic, Adiabatic Flow into the Closed Chamber as a Transient System Process 3 Polytropic Expansion of Open Chamber as a Closed System Process 4 Isentropic Expansion through the Turbine from Closed to Open Chamber Isotropic, Adiabatic Flow into the Open Chamber as a Transient System Isotropic, Adiabatic Flow from the Closed Chamber as a Transient System Process 5 Venting of Closed and Open Chambers to Atmosphere Page 13.197.5 Open Closed Chamber Chamber

Turbine Turbine

Figure 5 Operation of a Two Chamber Oscillating Water Column

A MATLAB program has been developed to facilitate the required calculations and produces performances results such as power produced and efficiency. The monthly wave data for five different coastal locations have been incorporated into the MATLAB program. An economic analysis has been built into the MATLAB program, so that an energy cost ($/kW hr) is calculated and can be used as the objective function in a design optimization. The MATLAB program is available for download at

http://www.egr.msu.edu/classes/me417/somerton/#projects

The energy cost is calculated by taking the annual energy production predicted by the thermodynamic model (at the coastal location specified) and dividing it by the annual cost of the system. The annual cost of the system is the capital cost of the system converted to an annual cost through the appropriate interest rate/lifetime factor equation. The capital costs are determined from estimates of construction costs and equipment costs (both tied to the size of the system).

Design Project The statement of the assigned design project is provided in Appendix I. Students perform a design study to determine the chamber heights and diameters that will maximize performance during the year. Students work in teams of two and perform the design analysis at two different locations of their choosing. The following locations are available in the program:

Aberdeen,WA (315 NM West) Aberdeen,WA (78 NM South SW) S Aleutians Hilo,HI (185 NM SE) Pensacola,FL (115 NM East)

This provides the students with a variety of different locations and wave conditions to choose from.

A written report is required to document their work and conclusions, with an important aspect being the comparison of the OWC plants design for the two different sites. The grading rubric is given in Fig.6. Page 13.197.6 Project Grade Evaluation Project 3 Design of an Oscillating Water Column Power System

Student Names: ______

Assigned Assigned Maximum Topic Score Score Score Site #1 Site #2 Month 1 Optimization 10

Month 2 Optimization 10

Month 3 Optimization 10

Month 4 Optimization 10

Graph of Power Output vs Open 15 Chamber Diameter Graph of Power Cost vs Closed 10 Chamber Volume Final Design Recommendation 5

Graph of Power Cost vs Month 10

Comparison of Sites 10

Quality 10

Total 100

Figure 6 Grading Rubric for Project

The optimization is performed to set four design parameters: open chamber height, open chamber diameter, closed chamber height and closed chamber diameter. To perform the optimization, the students used a directional walk method in which they vary one parameter, keeping the others constant, until a minimum in the energy cost is determined. This process is repeated for the other three parameters, and then the whole process is repeated again to assure determination of the “true” minimum point. Not surprising, the optimum dimensions tended towards large diameters and small heights. This optimization is done at four different months. From these results, the student team must choose the diameters and heights to use in their final design. An example of month by month performance of the final design is given in Fig. 7.

Optimized Model for Pensacola, Florida- Power Cost vs. Month

0.1

0.0914 0.09

0.0789 0.08 0.0789

0.07 0.0691

0.06

0.05

0.0414 0.0414 0.04 0.0414 0.0357 0.0357 0.0316 0.0357 0.03 0.026

0.02 l r r r ry ry h ri st e a rc uly u be Ap May June J to nu Ma emb a Aug Page 13.197.7 J ebrua ptembe Oc v F e S No December Month Figure 7 Monthly Energy Cost A project grade was assigned using the grading rubric of Fig. 7. The distribution of grades for this project are shown in Fig.8 and indicates that the vast majority of the students produced a high quality project,

4 Number of Students Number of

0 90 95 100 Project #3 Score

Figure 8 Project Grade Distribution

Student Feedback At the end of the semester a survey was administered to the students. For each of the alternative energy technologies covered by the three technical projects the students were asked to evaluate their level of confidence using a 5-1 scale with 5 being best for the course learning objectives for the technologies. For ocean energy these were:

a. Students are able to understand the nature of the ocean as an energy source b. Students are able to understand and evaluate different types of ocean energy sources, such as ocean thermal energy conversion, wave energy, and tidal energy c. Students are able to calculate the performance of ocean power systems d. Students are able to design an ocean power system

Table 2 shows the numerical averages of the students’ scores.

Table 2 Averages of Student Responses

Objective Average Score a 4.83 b 4.50 c 4.38 d 4.36

These are quite good numbers and may indicate that learning happened. Clearly, objectives c and d are at a higher level of Bloom’s taxonomy than objectives a and b, so it is not surprising to Page 13.197.8 see that slight falloff in the student ratings. A distribution of the scoring is shown in Fig. 9.

35 30 25 Number of 20 Students 15 10 5 3 0 4 Confidence a 5 Level b c d Learning Objective

Figure 9 Distribution of Student Scoring for Ocean Energy Learning Objectives

Note that all of the objectives were scored no lower than 3 by any student in the class. For comparison purposes, Table 3 shows the student response for objectives c and d for all three technologies. The projects used for the fuel cell and solar technology are tried and tested, so the lack of a discernable difference indicates that the oscillating water column project was quite successful.

Table 3 Averages of Student Responses for Different Technologies

Objective Fuel Ocean Solar Cells Energy Energy c 4.45 4.38 4.52 d 4.12 4.36 4.44

Lessons Learned

1. A simple thermodynamic model for an oscillating water column works well in predicting performance. 2. Students were quite uncomfortable deciding on a final design based upon the four monthly optimizations. 3. Student learning appears to be strong using this project. 4. Few students took advantage of the availability of the source code to create loops for the optimization. 5. The project compared favorably to more tested and tried projects.

References 1. ME 416 Spring 2007 website, www.egr.msu.edu/classes/me417/somerton, visited 2/27/08. 2. RETScreen International Web Site, www.retscreen.net/ang/home.php, visited 2/27/08. Page 13.197.9 3. Wikpedia website, en.wikipedia.org/wiki/Ocean, visited 2/27/08. 4. Infoplease website, www.infoplease.com/ipa/A0001801.html, visited 2/27/08. 5. Department for Business, Enterprise and Regulatory Reform, United Kingdom Web Site, www.berr.gov.uk/energy/sources/renewables/news-events/press- materials/background/tidal/page24345.html, visited 2/27/08. 6. Derby Visitor Center website, www.derbytourism.com.au/pages.asp?code=38, visited 2/27/08. 7. Renewable Energy UK website, www.reuk.co.uk/La-Rance-Tidal-Power-Plant.htm, visited 2/27/08. 8. Nova Scotia Power website, ww.nspower.ca/environment/green_power/tidal/index.shtml, visited 2/27/08. 9. Verdant Power website, www.verdantpower.com/what-initiative, visited 2/27/08. 10. NOAA Regional Ocean Forecast System at http://polar.ncep.noaa.gov/cofs/, visited 1/15/07. 11. National Renewable Energy Laboratory, www.nrel.gov/otec/what.html, visited 1/15/07. 12. New and Renewable Energy Centre (NAREC), www.global-renewables.com/technologies- wave-tidal.php#, visited 1/15/07.

Page 13.197.10 Appendix I OWC Project Statement

ME 417 Design of Alternative Energy Systems

Project 3 Design of an Oscillating Water Column Power System Due Friday, March 30, 2007

The U.S. Senate subcommittee on coastal energy resources has hired the engineering firm of Bénard and Somerton (aka BS Engineering) to provide a preliminary design for an oscillating water column system power within the territory of the United States. Two associates of the firm have been assigned to investigate two different sites in the United States. Each associate will perform the design analysis for a single site. The following sites have been targeted for this study.

315 NM W of Aberdeen (WA) 78 NM SSW of Aberdeen (WA) S. Aleutians (AK) Hilo (HI) Pensacola (FL)

For each site, the design team will undertake a design study to determine

• the optimal open chamber height (within 0.5 m) • the optimal open chamber diameter (within 0.5 m) • the optimal closed chamber height (within 0.5 m)

The closed chamber diameter will be taken to be equal to the open chamber diameter.

The objective function used in the design study will be the predicted cost of the electricity in $/kW ⋅hr, under the constraint that the system power output will be 5 MW. The team will want to compare the results of the design studies at the different sites. The design team should perform a design study for each of four different months at each site.

The following parameter values should be used in the design analysis:

Interest Rate: 12% Turbine Life: 10 years

A MATLAB computer program, OWCPower.m, has been provided to assist you in the calculations. The user’s guide for the program is attached.

To get started the student needs to choose a site and a month. Using the OWCPower.m program,

the student should start at a chamber height (say 7 meters) and a chamber diameter (say 10 Page 13.197.11 meters), and then change the open chamber height until a maximum power output is achieved. The open chamber and closed diameters should then be changed to obtain the 5 MW power output. The economic analysis should then be conducted. The closed chamber height should then be changed and the process repeated until the closed chamber height that minimizes the power cost is found. The month is then changed and the search for an optimum chamber diameter/height combination for the new month is conducted. This optimization search is done for any four months. Using the results for these four months, the student needs to choose an final chamber diameter/height combination based on minimum power cost and viability. The OWCPower.m program should then be run with this final design for each month of the year, which would produce a graph of Power Cost vs. Month. Appropriate documentation (tables and graphs) for the optimization and design study must be provided in the report that will support the design decision. The documentation should include the data of the directional walk optimization, a graph of power output vs. open chamber diameter at various open chamber heights, and a graph of power cost vs. closed chamber volume.

Page 13.197.12