Abstract

ABSTRACT INTRODUCTION Remote ocean instrumentation and monitoring techniques often rely on floating buoys with a variety of sensors to acquire time series measurements such as ambient noise, acoustic tracking or communications. The operating lifetime of small remote buoys is limited by onboard battery power. Remote acoustic sensors with hydrophone arrays, onboard RF transmitters, GPS receivers and other support electronics can draw up to 100-200W of continuous power in operation, limiting battery life in many cases to 12 to 24 hours between recharge. Recharging is inconvenient, and often impractical to the point that many compact sonobuoys are designed to scuttle themselves after about a day. The associated cost, as well as the environmental impact of sending large amounts of battery and electronic hardware to the bottom of the ocean, is a strong driver for developing renewable ocean power sources for semi-permanent unattended buoy deployments. This paper describes the first phase of what will hopefully be a multi-stage development effort for small ocean energy scavenging buoys. The primary technical objectives of the Phase 1 effort were: • Design and modeling of a float-pull cord wave energy generation system. • Preliminary sizing and conceptual design of the ultimate 100-200W open ocean power generating system. Phase 1 concentrated on proof of concept of the power generation mechanism and float. The next phase of the project will begin the transition to open water operation by examining the sea anchor that will ultimately be necessary for open water operation as well as scaling the power upwards from 50W to 100-200W. In support of this vision, a simple, low-cost buoy size platform capable of generating power by scavenging energy from ocean wave motion is developed and demonstrated. The phase 1 prototype is designed to deliver a minimum of 50W of average power from the wave motion characteristics. The motions and the resulting tension in the mooring line are calculated through a linear potential-based computer program. The heaving-body WEC (Wave Energy Converter) design and the modeling are discussed. The design data and the calculations are presented as part of proof of concept of the power generation mechanism and the buoy. Phase 1 consisted of: • A modeling and simulation phase to size the concept, predict how much power it will generate under various conditions, and calculate the resulting forces and reactions on the various subcomponents. • A design phase for the prototype (and some preliminary conceptual designing of the ultimate product -enough to ensure that the prototype is representative of the product). to the analysis (and in Phase 2 enhance the models accordingly). Phase 1 helped to prove the viability of the concept of a pullcord generator coupled to an inflatable surface float, and verify that model and experiments correspond for the 50W prototype. It will help to answer questions such as the sea state required to generate a given amount of power for this arrangement given a certain size float and set of allowable forces. The Phase 1 baseline models will also ultimately be used to examine the anticipated sizing of the components for the ultimate full 100-200W operation. The device that we describe here to generate energy from wave motion is more compact and lighter weight than many that are available today (see for example, The inflatable float at the surface (which is often used in sonobuoy systems anyway for radio communications antennas) will move up and down with the action of the waves. The underwater portion of the sonobuoy (where the hydrophones and acoustic devices are typically located) was anchored to the bottom in Phase 1 (see To calculate how much energy can be generated from such an arrangement, we look at typical sea states: • Sea State 3: 2.87ft significant wave height and 7s peak period. • Sea state 4: 6.15ft significant wave height and 12s peak period. For example, in order to generate 100W (average) power from a 1m average wave amplitude with this approach, 700 N (157 lbf) of force would have to be developed on the tether (Energy = Force * Distance, hence 700 N * 1 m = 700 J, Power = Energy / Time, hence an average power of 100 W). The 700 N force is generated by the float rising with the wave, and the underwater sail or anchor creating an equal and opposite counterforce underwater (the sail feature of the underwater portion of the system allows it to generate a larger counterforce than it could achieve by its mass alone given the buoyancy effects when submerged in water). This mechanical energy is converted into electrical power by a generator coupled to a spool mechanism that the tether winds and unwinds about In deep water, wave energy dissipates rapidly with depth. However there is some wave energy present down to a depth equivalent about one half the wavelength of the surface wave. For a typical open ocean wave, wavelengths can be 100m, which means that to avoid significant motion of the wave the sea-anchor portion of the system would have to be submerged 50m or deeper. However, due to the rapid fall off in wave energy with depth, it may not be necessary to place the anchor portion this deep in actual practice. Trex and University of Hawaii have modeling and simulation tools available which allow us to model the dynamics of the tether cable and its effects on the generator, float, and submerged portions of the system. Ultimately we hope that buoys using this technology may be used for: MODELING The physical model consists of a buoy floating on the surface, a submerged Mechanism Box below the buoy and the mooring lines that connect them to each other and to the sea floor. The model can be seen in The results of the analysis are RAOs for body motions and mooring line forces. These RAOs are used, together with the wave spectra, to obtain estimates of the short-term extreme response. Bretschneider spectrum is used in the irregular-sea analysis for the two different significant wave height and peak period combinations (see the last section) used in the study. The average water depth of 33ft is used in the hydrodynamic calculations. The buoy is a sphere of 27in diameter. It weighs 20 lbf in air but with 50lbf of buoyancy there is a pretension of 30lbf in the connector that connects it to the Mechanism Box below. Only 6in of the buoy is submerged below the SWL. Therefore, the mass of the buoy is 1.553slug and the mass moment is 0.7862slug-ft^2. The center of gravity is on the SWL where the body coordinate origin is. The panel model of the buoy is shown in The buoy is connected to the Mechanism Box by a very stiff connector (axial stiffness 1x10^6 lbf/ft). There is pretension of 30lbf in this connector and therefore the two transverse components of the stiffness matrix are 60lbf/ft each since the length of the connector is 0.5ft. Other components of the stiffness are assumed zero, i.e., the connector provides no resistance to any rotations. The Mechanism Box is modeled as a rectangular box of 8in width, 16in height and 16in length. It weighs 60lbf in air. Thus its mass is 1.863slug. As a result, the steel rope that connects the Mechanism Box to the sea floor is pretensioned an additional amount of 16lbf so that the total pretension in the line is 46lbf. The mass moment of inertias of the Mechanism Box are 0.2483slug*ft^2 in the x (longitudinal) and z (vertical) directions, and 0.552slug*ft^2 in the y (transverse) direction. The x axis corresponds to the zero degree wave heading. The center of gravity is assumed to be 0.5ft below the body coordinate origin which is at the centroid of the Mechanism Box. The Mechanism Box is connected to the sea floor (Anchor Box) by a steel rope. The steel rope is pretensioned a total of 46lbf and thus the transverse components of the stiffness matrix is 1.58lbf/ft. The axial stiffness of the steel rope is determined through experiments. The stainless steel cable is 3/16" diameter stranded cable made up of 7 sub-cables wrapped together in a helix. The "sub-cables" are each made up of 19 individual solid stainless steel wires each 10 mils in diameter. The helix of the cable bends around to the right and it is 7 x 19 (7 major strands of 19 filaments each). Its pitch is 1.15". The laboratory experiments provided us with an axial stiffness of 3,697 lbf/ft (244lbf of tension over 0.2% strain). Even though we have the stiffness of the steel rope, it is not possible to directly use it in the calculations since the rope winds about the spool mechanism. Electricity is generated only when the buoy-mechanism-box combo moves up, i.e., over one-half of the wave cycle only. One way to model such a physical system is through the use of a viscous dashpot (damper) instead of the stiffness of the steel rope. During the dry laboratory experiments, it is determined that the winding velocity of the steel rope is approximately 2ft/s when the pull force is about 150lbf. Therefore, we have decided to use a viscous dashpot coefficient of c=75lbf/(ft/s) in HYDRAN calculations. SIMULATION RESULTS Fig The amplitude of the vertical force or the axial tension (dynamic) in the steel rope that connects the Mechanism Box to the sea floor can be obtained through the following equation: where ω is the angular wave frequency, c is the viscous dashpot coefficient and x 3 is the heave amplitude. This force is shown in 3 Copyright © 2009 by ASME LINE TENSION VERSUS WAVE POWER The Phase 1 preliminary/conceptual design was altered as shown in • Preserve buoyancy of float (currently ~300 lbf) • Not puncture and have to re-seal the inflatable skin • Allow mechanism to also displace water, minimizing the resulting loss of buoyancy This appears to be acceptable from preliminary calculations: • 700N = 157lbf over a 1m displacement every 7s (waves) would yield 100W • 25 lbf return spring brings the total extension force to 182 lbf • Maximum submerged buoyancy of about 300lbf appears possible from the buoy • This yields a design margin of almost 2 on most key parameters -likely satisfactory in this early stage of the design We have developed a crude yet surprisingly effective method of testing, sizing, and determining the necessary gearing and load characteristics of our electrical generators. By placing a candidate generator in a clamp on a mill or drill press, and then changing the electrical loads and rotational speed, we have developed families of curves for different generators to obtain the power as a function of the RPM. What was evident from our earlier efforts is that to achieve 50W at a low rotational speed (which is desirable in terms of limiting gearing), and with a larger margin of safety, we needed to order a larger generator than those we already had on the shelf. We have received this generator, and the resulting test data are shown in We also tried lowering the load impedance to 3 ohms as shown in the blue trace on the graph (again A prototype was fabricated in the laboratories of TREX Enterprises Corporation in order to test all major subsystems of the design prior to final integration (and more expensive tests in the ocean will follow). It is important to note that the first initial prototype was not waterproof -it is simply meant to hold the components in place for dry land tests in the lab to confirm their proper performance prior to being incorporated into an appropriate water tight enclosure Over 150 Watts of peak power was generated from this system in a laboratory environment, at relatively modest pull-forces (150lbf and less out of about 300lbf buoyancy potentially available from the A5 buoy). We can now predict the available wave power. The average wave power over a wave cycle is obtained by multiplying Eq. (1) by heave velocity and integrating over the wave cycle. Recall further that we can only produce power in one-half of the wave cycle. The resulting (average) wave power can be shown to be 4 / 2 3 2 x c P AV Since about 140lbf of pull force is predicted by the calculations when the wave amplitude is 2ft and the wave period is 5-7s, we estimate that we can generate over 140W of power. 142W peak power is needed for 50W average power since only in one-half of the wave cycle the steel rope will be providing the rotation to the drum, and in the other half, it will simply be retracted by the return spring. On average, about 35.4% (100*0.707/2) of the peak power obtained from The significant value of the tension in the steel rope is 102lbf in Sea State 3 and 151lbf in Sea State 4. The short term extreme spectral results for the tension in the steel rope for two different sea states based on the Bretschneider spectrum showed that in Sea State 3, the extreme tension is 190lbf and for Sea State 4, it is281lbf (for zero wave heading). Finally, the average wave power obtained from Eq. (2) is shown in CONCLUSIONS Phase 1 helped to prove the viability of the concept of a pullcord generator coupled to an inflatable surface float, and verify that the model and the laboratory (dry) experiments correspond to the 50W prototype. We have answered questions such as the sea state required to generate a given amount of power for this arrangement given a certain size float and set of allowable forces. The Phase 1 baseline models will also ultimately be used to examine the anticipated sizing of the components for the ultimate full 100-200W operation. As a further demonstration of the concept, we plan to conduct in ocean (near-shore) experiments with the prototype float/anchor/pull-cord system to generate 50W of power -at the end of a pier for example. One potential site for this test is Kilo Nalu Near-shore Reef Observatory on Oahu, which experiences waves enhanced by the shallow bottom (breakers). We may also look for other similar sites, piers, or points in Hawaii. ACKNOWLEDGMENTS We would like to thank the Office of Naval Research for their support and funding of this project, as well as allowing us to publish these results. We are also grateful to Prof. Ronald Knapp of the Univ. of Hawaii for providing us with the laboratory data on the stiffness of the steel rope used here.