Journal of Ocean and Wind Energy (ISSN 2310-3604) Copyright by The International Society of Offshore and Polar Engineers Vol. 2, No. 4, November 2015, pp. 248 252; http://dx.doi.org/10.17736/jowe.2015.jcr30 http://www.isope.org/publications/publications.htm Parametric Study of the Power Absorption for a Linear Generator Wave Energy Converter Wei Li, Jan Isberg, Jens Engström, Rafael Waters and Mats Leijon Division of Electricity, Swedish Centre for Electricity Energy Conversion Uppsala University, Uppsala, Sweden This paper presents a performance optimization study of a linear generator wave energy converting system developed at Uppsala University in Sweden. The optimization is focused on the investigation of the power absorption with regard to the wave climate at the Lysekil test site for the developed wave energy converter. A frequency-domain simplified numerical model is built on the basis of the equation of motion of the system to simulate the behavior of the wave energy converter. The power take-off damping coefficient is chosen as the control parameter in this study. The results show that the power take-off damping coefficient can be further optimized with regard to the characteristics of the wave climate at the test site to increase the power absorption capacity of the linear generator wave energy converter. This study will provide significant performance optimization information for the development of the linear generator wave energy conversion technology. INTRODUCTION Nowadays, the potential of providing renewable energy from ocean waves has increased significantly with the development of wave energy conversion technology. The estimated global wave energy potential is in the order of 10 TW, which is comparable with the power demand of the whole world (Raffero et al., 2015). In the last decade, many different wave energy conversion technologies have been developed to harvest this huge power all around the world. Some of these have progressed to a full-scale test and the pre-commercial stage (Drew et al., 2009; Falcão, 2010). The scientific community will continue to work towards capturing ocean wave energy through the development of highefficiency wave energy converters. At Uppsala University in Sweden, a linear generator wave energy converter has been under development since 2001. The wave energy converter is based on a linear driven generator mounted on a gravity-based foundation located at the seabed. The generator is connected to a point absorbing buoy on the sea surface via a rope (Eriksson, 2007; Leijon et al., 2008). It has been in operation at the Lysekil wave energy converter test site since 2006. So far, more than twelve full-scale linear wave energy converters have been deployed and tested at the Lysekil test site (Hong et al., 2013). A schematic of the linear generator wave converter is shown in Fig. 1. In the near future, a wave energy farm with hundreds of wave energy converters will be built at a nearby site to supply electrical power to the local community. The efficiency of a wave energy converting system determines the success of its commercial development (Flocard and Finnigan, 2012). To develop high-efficiency wave energy converters, a performance study is important, which provides information on an optimal design of the considered wave energy converting system. Considerable effort has been made to evaluate the performance of different types of wave conversion systems on the basis of numerical modelling and an experimental approach (Babarit et al., 2011; Duclos et al., 2006; Silva et al., 2013). Some recent studies of the power absorption of point absorbing wave energy converter systems with linear generator converters based on the numerical method and practical experiment were conducted by Lejerskog et al. (2015) and Pastor and Liu (2014a, 2014b). In this paper, performance optimization is investigated for the linear generator wave energy converter developed by Uppsala University. In the wave energy conversion system, a toroidal buoy as the wave power absorber is investigated. A frequency-domain numerical model of the wave energy converting system is created to simulate its operation. In this paper, the damping coefficient of the power take-off is chosen as the control parameter for the optimization study, and the main focus is on the power absorption aspect of the linear generator wave energy converter. The power capture ratio is evaluated through different control strategies of the power take-off damping parameter to optimize the average annual power production with respect to the wave climate at the Lysekil test site. This study is intended to provide further information on the development and optimization of this wave energy conversion technology. Received March 11, 2015; revised manuscript received by the editors September 2, 2015. The original version was submitted directly to the Journal. KEY WORDS: Wave energy converter, linear generator, power absorption, wave spectrum, frequency domain. Fig. 1 Schematic of the linear generator wave energy converter
Journal of Ocean and Wind Energy, Vol. 2, No. 4, November 2015, pp. 248 252 249 where m is the mass of the buoy including the mass of the translator, F e is the excitation force, F r is the radiation force, F PTO is the external power take-off damping force, and F R is the hydrostatic restoring force (see Falnes, 2002). For simplicity, we assume a linear power take-off damper; in this case, F PTO is given by: F PTO = ż (2) where and ż are the damping coefficient and the velocity of the buoy, respectively. The hydrostatic restoring force can be expressed as: (a) Fig. 2 The linear generator wave energy converter L9 with connected torus buoy METHOD Equation of Motion The linear generator wave energy converter L9 will be considered in this optimization study (see Fig. 2a). It was deployed in October 2009 at the Lysekil test site. The buoy connected to L9 is a torus buoy with a radius of 2.5 meters and a height of 0.5 meter, as shown in Fig. 2b (Gravråkmo et al., 2010). In the modelling, we simplified the converting system to include only the point absorbing torus buoy connected to the power take-off (PTO) damper, as sketched in Fig. 3. For simplicity, the end stops will not be considered in the model; therefore, there is no constraint on the heave motion of the buoy. The draft of the buoy is determined by the weight of the buoy and the translator. We assume that the buoy motion is only in the heave direction and that the linear wave theory is applied for the evaluation of the hydrodynamic forces. The validity of using these assumptions has been previously confirmed for a cylindrical buoy connected to a linear generator through a comparison of simulations and open sea trials (Eriksson et al., 2007). Cylinder and torus buoys behave similarly, and there is no reason to believe that these assumptions do not yield a reasonable approximation in our case as well. The assumptions have been the starting point for a number of studies in the literature, e.g., Bosma et al. (2012), Bozzi et al. (2013), Eriksson et al. (2005), and Filianoti and Camporeale (2008). Given the vertical direction of the axis z, the acceleration of the buoy z can be written as: (b) F R = F b F g = kz (3) where F b and F g represent the buoyancy force of the buoy and the gravitational force, respectively, k is the hydrostatic stiffness, and z is the displacement of the buoy from its equilibrium position. In the frequency domain, the excitation force F e can be expressed as: F e = Af e (4) where A is the amplitude of the incoming wave and f e represents the transfer function of the excitation force. The radiation force is given by: F r = m a z B ż (5) where m a is the added mass of the buoy and B is the radiation resistance. Inserting Eqs. 2 to 5 into Eq. 1 gives: 2 m + m a + i + B + k z = Af e (6) The response amplitude operator (RAO) of the buoy in heave motion can be written as: RAO = z A = f e 2 m + m a + i + B + k The corresponding hydrostatic coefficients and added mass of the torus buoy for L9 are computed using WAMIT in this modelling. The RAO for the torus buoy of L9 with the installed damping coefficient 4 10 4 N s/m is shown in Fig. 4, and the corresponding (7) m z = F e + F r + F PTO + F R (1) Fig. 3 Simplified model of the converter Fig. 4 RAO for the torus buoy of L9
250 Parametric Study of the Power Absorption for a Linear Generator Wave Energy Converter Fig. 5 Power function p of L9 power function p of the wave energy converting system (see Eq. 8) is given in Fig. 5. p = 1 2 B PTO 2 RAO 2 (8) where B PTO is the power take-off damping coefficient. We can see a maximum peak at a certain wave frequency in the power function of the wave energy converting system in Fig. 5. This indicates the importance of the wave climate to the power absorption of the wave energy converting system. Wave Climate The wave climate measured at the Lysekil test site by a Datawell Wave Rider buoy in the period of 2007 to 2009 is used to optimize the power absorption for the L9. The joint distribution of the significant wave height H s and the energy period T e of the measured wave data at the Lysekil site is given in Fig. 6. The wave power can be approximately estimated by analysis of Fig. 7 Comparison of the spectrum of the measured wave climate at the Lysekil site with the Bretschneider spectrum the measured wave data, as given by Eq. 9: J = g2 64 H 2 s T e (9) where J is the wave power flux per unit of wave crest length, is the density of sea water, 1,025 kg/m 3, g is the acceleration of gravity, and T e is the energy period of the wave. In the frequency domain, the power absorption for the converting system can be evaluated by the use of Eq. 10: P abs = 2S p d (10) where S is the spectrum of the wave climate and p is the power function of the converting system, as defined in Eq. 8. A comparison of the evaluated spectrum of the measured wave data with the Bretschneider spectrum is shown in Fig. 7. The Bretschneider spectrum is given by Eq. 11: S B = 5 4 m 16 H 2 5 1/3 e 5 4 m /4 4 (11) where is the frequency in radians per second, m is the modal frequency of any given wave, and the significant wave height is represented by H 1/3 (Bretschneider, 1959). We can see that the evaluated spectrum matches the Bretschneider spectrum quite well except that it is slightly wider than the Bretschneider spectrum. Thus, the Bretschneider spectrum can be used in Eq. 10 to represent the wave spectrum at the Lysekil site for the power absorption evaluation with reasonable accuracy (Mackay, 2011). RESULTS AND DISCUSSION Fig. 6 Joint distribution of the significant wave height H s and the peak period T p of the wave climate at Lysekil The energy production of the L9 with the default power takeoff damping coefficient with regard to the statistics of the wave climate at the test site is shown in Fig. 8. The evaluated annual average output power in this case is 2 kw. The optimization of the performance of the considered wave energy converter L9 is based on different control strategies for the power take-off damping coefficient (Duclos et al., 2006; Genest et al., 2014). First, the power absorption is optimized by seeking the optimal constant power take-off damping coefficient with respect to the wave climate at the test site. This means that the performance of the
Journal of Ocean and Wind Energy, Vol. 2, No. 4, November 2015, pp. 248 252 251 Fig. 8 Annual energy production of the L9 with the default power take-off damping coefficient converting system can be optimized by adapting the power takeoff damping coefficient to the considered wave climate at the test site. The wave energy conversion system with a certain installed power take-off damping would harvest energy from the waves more efficiently than for other values of the power take-off damping. For the wave climate of a certain site, the size of the buoy and the power take-off damping coefficient are essential parameters to maximize the power output. In our case, we considered only the power take-off damping. Figure 9 shows the evaluated annual average output power with different power take-off damping coefficients. We can see that with the chosen damping coefficient, 107.23 kn s/m, the highest annual average power output is given, which is 3.9 kw. This is twice the annual average output power compared to that with the use of the default damping coefficient. With the optimization above, we chose a constant power takeoff damping coefficient for entire sea states within the statistics diagram of the wave climate at the Lysekil test site. Furthermore, we considered applying a control to the system by varying the power take-off damping coefficient for different sea states, which means that the power take-off damping will be optimized in every sea state within the statistics diagram of the wave climate at the Fig. 9 Annual average power output as a function of the power take-off damping coefficient Fig. 10 Annual energy production of L9 with control of the damping coefficient Annual energy Annual average Power Damping type production power capture ratio Installed 1 57 10 4 kwh 2.0 kw 14.11% Constant 3 43 10 8 kwh 3.92 kw 27.66% Controlled 3 62 10 4 kwh 4.13 kw 29.15% Table 1 Comparison of the optimized performance of L9 test site to achieve an overall maximum annual power output. This gives further optimization of the output power for the wave energy converting system. The energy production for L9 with the controlled power take-off damping coefficient with regard to the statistics of the wave climate at the test site is shown in Fig. 10. The evaluated annual average output power becomes 4.13 kw. The incoming wave annual average power for the torus buoy with a diameter of 5.05 m at the Lysekil test site is 14.17 kw. A comparison between the performance of L9 with the installed power take-off damping coefficient and the optimized performance of the power absorption is shown in Table 1. As we can see from Table 1, the power capture ratio increased to 27.66% with the optimized constant damping coefficient, compared to 14.11% with the default damping coefficient. Under the controlled damping coefficient, the power capture ratio increased to 29.15%. However, it did not increase that much compared to the constant damping coefficient. The reason could be that the wave power from the incoming waves at the Lysekil test site is mainly within the peak periods of 4 to 6 s and the significant wave height of 1.5 to 3 m. This gives less space for the controlled damping coefficient to improve the power capture ratio of the converting system. Hence, we can see that for the power absorption optimization of the wave energy conversion system, the wave climate is essential when control strategies to maximize the power output are considered. A wave energy conversion system should be designed according to the characteristics of the installed site for efficient performance. In this performance optimization study, the power take-off damping coefficient is the only control parameter that we take into account in the model. However, another parameter, i.e., the dimension and shape of the point absorber buoy, is also very important for the optimization of the power capture ratio. Here L9 is connected to the torus buoy, which can reduce the excitation force in comparison to the cylindrical buoy in extreme wave conditions. This is important in order to secure the stability of the moor-
252 Parametric Study of the Power Absorption for a Linear Generator Wave Energy Converter ing system since it is mounted on a gravity-based foundation. We also considered a linear damping system that we actually installed in L9. Including nonlinear damping may result in a higher power capture ratio. This will be investigated in future work. CONCLUSIONS The purpose of this paper is to optimize the power absorption performance of the linear generator wave energy converter developed at Uppsala University. On the basis of the equation of the motion of the converting system, a frequency-domain model is built to investigate the power absorption performance with regard to the wave climate at the Lysekil test site. The power take-off damping coefficient is taken into account as a control parameter in this optimization study. The results show a large improvement in the power absorption when the optimal damping coefficient is chosen for the considered wave energy converter L9. This optimization study provides important information on the development and optimization of the linear generator wave energy converter. ACKNOWLEDGEMENTS The authors would like to thank StandUp for Energy, Uppsala University, the China Scholarship Council (CSC), and the Swedish Energy Agency for their support. REFERENCES Babarit, A, Hals, J, Kurniawan, A, Krokstad, J, and Moan, T (2011). Power Absorption Measures and Comparisons of Selected Wave Energy Converters, Proc 30th Int Conf Ocean Offshore Arct Eng, Rotterdam, Netherlands, OMAE, 1, 437 446. http://dx.doi.org/10.1115/omae2011-49360. Bosma, B, et al. (2012). Wave Energy Converter Modeling in the Frequency Domain: A Design Guide, Energy Convers Congr Expo (ECCE), IEEE, 2099 2106. http://dx.doi.org/10.1109/ecce.2012.6342553. Bozzi, S, Miquel, A, Antonini, A, Passoni, G, and Archetti, R (2013). Modeling of a Point Absorber for Energy Conversion in Italian Seas, Energies, 6(6), 3033 3051. http://dx.doi.org/10.3390/en6063033. Bretschneider, CL (1959). Wave Variability and Wave Spectra for Wind Generated Gravity Waves, Technical Memo No 118, Beach Erosion Board, US Army Corps of Engineers, Washington, DC, USA. Drew, B, Plummer, AR, and Sahinkaya, MN (2009). A Review of Wave Energy Converter Technology, Proc Inst Mech Eng, Part A: J Power Energy, 223(8), 887 902. http://dx.doi.org/10.1243/09576509jpe782. Duclos, G, Babarit, A, and Cleìment, AH (2006). Optimizing the Power Take Off of a Wave Energy Converter With Regard to the Wave Climate, J Offshore Mech Arct Eng, 128(1), 56. http://dx.doi.org/10.1115/1.2163877. Eriksson, M (2007). Modelling and Experimental Verification of Direct Drive Wave Energy Conversion: Buoy-Generator Dynamics, Doctoral Thesis, Uppsala University, Uppsala, Sweden. Eriksson, M, Isberg, J, and Leijon, M (2005). Hydrodynamic Modelling of a Direct Drive Wave Energy Converter, Int J Eng Sci, 43(17 18), 1377 1387. http://dx.doi.org/10.1016/j.ijengsci.2005.05.014. Eriksson, M, Waters, R, Svensson, O, Isberg, J, and Leijon, M (2007). Wave Power Absorption: Experiments in Open Sea and Simulation, J Appl Phys, 102(8), 084910. http://dx.doi.org/10.1063/1.2801002. Falcão, AFDO (2010). Wave Energy Utilization: A Review of the Technologies, Renewable Sustainable Energy Rev, 14(3), 899 918. http://dx.doi.org/10.1016/j.rser.2009.11.003. Falnes, J (2002). Ocean Waves and Oscillating Systems, Cambridge University Press, 286 pp. Filianoti, P, and Camporeale, SM (2008). A Linearized Model for Estimating the Performance of Submerged Resonant Wave Energy Converters, Renewable Energy, 33(4), 631 641. http://dx.doi.org/10.1016/j.renene.2007.03.018. Flocard, F, and Finnigan, TD (2012). Increasing Power Capture of a Wave Energy Device by Inertia Adjustment, Appl Ocean Res, 34, 126 134. http://dx.doi.org/10.1016/j.apor.2011.09.003. Genest, R, Bonnefoy, F, Clément, AH, and Babarit, A (2014). Effect of Non-ideal Power Take-off on the Energy Absorption of a Reactively Controlled One Degree of Freedom Wave Energy Converter, Appl Ocean Res, 48, 236 243. http://dx.doi.org/10.1016/j.apor.2014.09.001. Gravråkmo, H, et al. (2010). Description of a Torus Shaped Buoy for Wave Energy Point Absorber, Presented at: Proc Renewable Energy Conf, Pacifico Yokohama, Japan. Hong, Y, et al. (2013). Status Update of the Wave Energy Research at Uppsala University, Proc 10th Eur Wave Tidal Energy Conf, Aalborg, Denmark. http://urn.kb.se/resolve?urn= urn:nbn:se:uu:diva-212701. Leijon, M, et al. (2008). Wave Energy from the North Sea: Experiences from the Lysekil Research Site, Surv Geophys, 29(3), 221 240. http://dx.doi.org/10.1007/s10712-008-9047-x. Lejerskog, E, Boström, C, Hai, L, Waters, R, and Leijon, M (2015). Experimental Results on Power Absorption from a Wave Energy Converter at the Lysekil Wave Energy Research Site, Renewable Energy, 77, 9 14. http://dx.doi.org/10.1016/j.renene.2014.11.050. Mackay, EBL (2011). Modeling and Description of Omnidirectional Wave Spectra, Proc 9th Eur Wave Tidal Energy Conf, Southampton, UK, 122 131. Pastor, J, and Liu, Y (2014a). Frequency and Time Domain Modeling and Power Output for a Heaving Point Absorber Wave Energy Converter, Int J Energy Environ Eng, 5(2 3), 1 13. http://dx.doi.org/10.1007/s40095-014-0101-9. Pastor, J, and Liu, Y (2014b). Power Absorption Modeling and Optimization of a Point Absorbing Wave Energy Converter Using Numerical Method, J Energy Resour Technol, 136(2), 021207. http://dx.doi.org/10.1115/1.4027409. Raffero, M, et al. (2015). Stochastic Control of Inertial Sea Wave Energy Converter, Sci World J, 2015, 980613. http://dx.doi.org/10.1155/2015/980613. Silva, D, Rusu, E, and Soares, C (2013). Evaluation of Various Technologies for Wave Energy Conversion in the Portuguese Nearshore, Energies, 6(3), 1344 1364. http://dx.doi.org/10.3390/en6031344.