Experimental Investigation on the Power Generation Performance of Dielectric Elastomer Wave Power Generator Mounted on a Square Type Floating Body

Transcription

1 Journal of Materials Science and Engineering B7 (9-10) (017) doi: /161-61/ D DAVID PUBLISHING Experimental Investigation on the Power Generation Performance of Dielectric Elastomer Wave Power Generator Mounted on a Square Type Floating Body Mikio Waki 1, Seiki A. Chiba, Z. Song 3, S. Zhu 3 and K. Ohyama 3 1. Wits Inc., Oshiage, Sakura-City, Tochigi , Japan. Chiba Science Institute, Muguro-ward, Tokyo , Japan 3. Fukuoka Institute of Technology, Wajiro-Higashi, Higashi-ku, Fukuoka Fukuoka , Japan Abstract: In this paper, a simple buoy was used for the power generation tests with a DE powered by waves in the two-dimensional wave tank and the experimental results were analyzed. This research result shows that the optimization of equipment shape and buoy based on the characteristics of the wave for power generation is very important. The extension and contraction of the DE are determined by the motion characteristics of the Heave of the simple shaped buoy model. Furthermore, in order to have freedom in tuning the motion characteristics of Heave, we found that a device capable of setting the spring characteristics of the DE to the smallest possible size with respect to the size of the floating body is necessary. By well matching these factors, high power generation with high efficiency can be realized in a real sea area. Key words: Dielectricelastomer, DE, electroactive polymer artificial muscle, wave power generator, ship. 1. Introduction To harness one kind of renewable energy obtained from Japan s vast surrounding sea areas, ocean wave energy utilization techniques are currently being developed [1, ]. Although ocean wave energy has the potential for large-scale power generation, it is inferior to other renewable energies with regards to its higher cost. Therefore, it is necessary to improve the utilization technique of wave energy in the future. The conventional method of wave power generation requires two conversion processes: the first is to convert wave energy to drive the turbine, and the second is to convert the kinetic energy of the turbine into electric energy. A wave power generator using a DE (dielectric elastomer) artificial muscle developed by Chiba et al. can directly obtain electrical energy by deforming the DE film with wave energy [3, 4]. In other words, no redundant energy conversion Corresponding author: Seiki Chiba, Ph.D., professor, research field: dieleastic elastomers. processes are required in the system. Therefore, it is very likely that the energy loss during conversion can be reduced as compared with other methods. The DE is low-cost, lightweight and highly shape-flexible. The high degree of flexibility for both mounting modes and the apparatus dimensions makes it possible to assemble multiple power generators within a small space [5]. In addition, since the deformation frequency of the elastomer hardly affects the power generation efficiency, there is no element whose frequency dependence appears other than the frequency characteristic of the moving object motion in the waves. The generation efficiency of the power generation motor varies depending on the rotation speed, but it is a major feature of elastomer that there is no such characteristic [6]. In this paper, a simple buoy was used for the power generation tests with a DE powered by waves in the two-dimensional wave tank and the experimental results were analyzed. In this test, experiments were

2 180 Experimental Investigation on the Power Generation Performance of Dielectric carried out using a square type floating body which was not designed by considering the fluid mechanics. As a result, only 50% of the power generation conversion efficiency was obtained at the incident wave of 0.7 second. Chiba et al. [5-7] realized high power generation efficiency (70-90%) via optimization of both the equipment and the buoy shape. This research result shows that the optimization of equipment shape and buoy based on the characteristics of the wave for power generation is very important. By well matching these factors, high power generation with high efficiency can be realized in a real sea area. 1.1 Background of the Dielectric Elastomer Artificial Muscle The DE developed for the actuator is thinly stretched by the Coulomb force when a voltage is applied, and then returned to its original state by the elasticity when the voltage is removed [8, 9]. As shown in Fig. 1, the electrodes coated on the elastomer surfaces attract each other because of the potential difference across the elastomer. The DE can be thinly extended via Coulomb force. The principle of power generation is reversed. The static electric energy is generated when the DE is stretched and subsequently elastically recovered. 1. Operating Principle of Generator There are several ways in which DE can be used to produce electrical energy from the mechanical work used to stretch and contract it. Here we present one method based on a constant voltage cycle [10]. Application of mechanical energy to stretch the DE causes a reduction in thickness and an expansion of the surface area (see Fig. a). At this moment, a voltage may be placed upon the polymer (i.e. positive charges are placed on one side and negative charges on the other side). When the stretching forces are removed, the elastic recovery force of the DE acts to restore the original thickness and to decrease the surface area (see Fig. b). The increase in thickness upon relaxation acts to push the opposite charges apart from each other, effectively raising the voltage applied to the DE. Even though the capacitance of the DE reduces upon relaxation, there is a net increase in the energy stored on the DE compared to that put on by the original application of the voltage. This increase can be harvested as electrical energy. The voltage can be measured and compared with predictions based on this DE theory developed by Chiba et al. [10]. In general, experimental data based on high impedance measurements are in excellent agreement with this prediction [10]. Functionally, this mode resembles piezoelectricity, but its power generation mechanism is fundamentally different. With a DE, electric power can be generated even by a slow change in the DE shape, while for piezoelectric devices impulsive mechanical forces are needed to generate the electric power [3]. Also, the amount of electric energy generated and the conversion efficiency from mechanical to electrical energy can be greater than that from piezoelectricity [3, 11-14]. At present, two approaches, (1) DE materials and mechanical systems using DEs [11-19], Fig. 1 Load Voltage off Load Basic operational principle of DE. Vout (high) Vout (high) Application of mechanical energy (a) DE Stretched Voltage on Vin (low) Vin (low) Dielectric Elastomer Compliant Electrodes (b) DE Relaxed Fig. Operating principle of DE power generation [4].

3 Experimental Investigation on the Power Generation Performance of Dielectric 181 and () operating strategies (circuit designs) [1, 16-19] are studied. The electric energy generation is most secured by utilizing the permanently existing sea wave. Based on this idea, Chiba et al. [7, 0] conducted an experiment with a small power generation buoy mounted with a DE in a real sea area. As the small power generation buoy responds to the waves, the DE film elongates, so the power generation efficiency was evaluated by measuring and analyzing the response in the waves of the small power generation buoy in addition to the momentary power generation amount. Moreover, in this real sea area experiment, the output cycle for the wave of the small buoy with the DE was clarified. However, since the wave drifting force received by the small buoy greatly affects the power generation, it is not possible to evaluate the power generation amount of the DE only with the vertical motion. In addition, since it was measured under complicated phenomena due to multidirectional non-waves, it was confirmed that power generation characteristics could not be sufficiently evaluated. 1.3 Research Objectives As described above, according to the previous research, the energy conversion efficiency of the buoy-type wave power generator with a DE has yet to be precisely determined, particularly the power generation property of the DE with vertical direction movement. In this work, therefore, we aim to clarify the power generation property of the DE mounted on the square type floating buoy with vertical direction movement in the two-dimensional water tank.. Two-Dimensional Water Tank Experiment.1 Outline of Experiment The apparatus adopted in this work takes the form of a DE mounted on the top of the buoy. The piles that were installed on the buoy act as guide at the bottom of the tank (above the temporary bottom). Four guides were used in order to secure sufficient mooring force at the connection points on the four sides of the buoy. Furthermore, a rope mooring the DE passes through the interior of the buoy and is threaded through the valleys installed at the bottom of the tank, and the DE was pre-stretched by fixing the end of the rope to the top of the tank at the point of P. The incident wave was measured at a position of 6.35 m from the end of the ramp. The schematics are shown in Fig. 3. Wire mounting Pile plate Wave maker 300mm A A 600mm 590mm Incident wave Y Z Z Y X Fig m X Pulley Block 400mm 150mm Generator P Pulley Block Experiment schematic drawing.. Outline of Experiment 5,000mm Ground Plan 300mm Pile Mooring Wire 0mm 5,000mm A-A Sectional Plan 50mm 1,500mm 1,500mm 10,000mm Wave maker Incident wave 10,000mm In order to evaluate the power generation of the DE with vertical direction movement, a regular wave was adopted as the incident wave, which was generated by the piston-type plate. The depth of water was 0.4 m, where the center of gravity of the model was located. The wavelengths of both long waves and short waves were measured for the regular waves. The measured specifications of period and height of the wave are shown in Table 1. In this experiment, power generation was realized via the extension and contraction of DE, which was wire-connected to the buoy with vertical movement. The DE was installed at the center of gravity of the buoy, and was fixed to the upper portion of the buoy by screws. The specifications of experimental model and DE are shown in Table, and Photo 1 shows the photographs of the model installation.

4 18 Experimental Investigation on the Power Generation Performance of Dielectric Table 1 Wave conditions. Height [m] Period [sec] Table Specifications of the experimental model and dielectric elastomer. Buoy length[mm] Draft [mm] Buoy width [mm] Generator diameter [mm] 10.0 Buoy height [mm] Generator height [mm] Model total mass [kg] 6.35 Photo. 1 Model float with DE. Target 1 P 1 =(x 1, y 1 ) Target P =(x, y ) was taken as r. Details in Eqs. () and (3) are specified in the Fig. 4. L L y y L () L y1 y tan x x 1 x1 x tan y1 y 1 (3) Fig. 4 L L 1 Details of the target position in the floating model. 3. Experimental Analysis Method In this experiment, the power P I of the regular wave acting as incident wave is determined, which is calculated utilizing the water density, the amplitude of incident wave a, the buoy width B, the water depth h and the wave number k, as represented by the following Eq. (1). g a kh P I 1 tanhkhtb (1) 8 sinhkh The buoy displacement is determined with one CCD camera by measuring two marks placed at the center of gravity of the buoy mounted with a dielectric elastomer via a two-dimensional analysis system. Then, each motion component was resolved and calculated with Fourier analysis method. The Heave component was taken as 3, and the Pitch component P G 4. Results and Discussion Fig. 5 shows the power of the incident wave obtained from Eq. (1). The horizontal axis is the dimensionless wave number, and L is the width of the buoy model. It was found that the power of incident wave decreases as the dimensionless wave number increases. Furthermore, the amplitude of the incident wave was set to be 0.0 m during the model experiment. Fig. 6 shows the final electrical energy output of experimental model mounted with dielectric elastomer, and the horizontal axis is the dimensionless wave number like Fig. 5. The vertical axis is the energy per unit time converted from the electrical energy output in the experiment, i.e. the power. Although the power variation with respect to the wave number of the incident wave is not significant, it can be confirmed that the energy output increases with the dimensionless wave number. Fig. 7 shows the power generation (mj) of the DE versus experimental time. The vertical axis is the power generation, which represents the energy accumulation. The horizontal axis is the real time during experiment. Each plot corresponds to the incident

5 Experimental Investigation on the Power Generation Performance of Dielectric 183 Fig. 5 Power of the incident wave. Fig. 6 Power generation per unit time. Fig. 7 Cumulative power generation energy. wave with a different period. It is found that larger power generation can be achieved for the incident waves with a shorter period. Particularly, the power generation remarkably increases when the period of the incident wave is 0.7 s. The staircase change can be seen in each plot. This is due to the expansion and contraction of the DE corresponding to the response of the single vibration of the buoy with regular wave, and the rising stage corresponds to one cycle movement of the incident wave. Therefore, the shorter wave period results in more vibration cycles, are shown the time between 10 s and 0 s in Fig. 7. Although the power generation within one cycle does not show significant variation with the wave period, the difference can be seen when the data are exhibited as the energy output per unit time, the trend of which is shown in Fig. 6. In other words, the wave period does not remarkably affect the power generation within one cycle, while under the same conditions the power of energy output increases as the wave period decreases. For this reason, the power of energy output increased in the region with shorter wave periods, or at the side with larger dimensionless wave numbers, as seen in Fig. 6. Fig. 8 shows the final energy conversion efficiency of buoy model mounted with the DE. The vertical axis corresponds to the value of Fig. 5 divided by that of Fig. 6. The horizontal axis is the dimensionless wave number. It is found that the power generation efficiency increases steadily in the range of 0%-50%. In other words, the shorter wave period and larger wave number result in higher power generation efficiency. In this experiment, an energy conversion efficiency of 50% was achieved within 0.7 s of the incident wave, which verifies the outstanding power generation performance. However, the energy output characteristics are largely dependent on the wave number of the incident wave. As described above, under the same experimental conditions, the power generation energy is the same within one cycle, while the power generation energy per unit time is smaller for the wave with longer periods. Furthermore, as shown in Fig. 5, smaller dimensionless wave numbers, or a longer wave period of the incident wave, results in higher wave power, which leads to the significant dependence of the energy conversion efficiency on the wave number, or the wave period. Fig. 9 shows the response function of the Heave component of the buoy model. The horizontal axis represents the dimensionless wave number. Since the theoretical calculation in the figure is based on the analysis of three-dimensional cases, it is regarded as a

6 184 Experimental Investigation on the Power Generation Performance of Dielectric Fig. 8 Power generation efficiency of each incident wave power. Fig. 9 Response function of heave. Fig. 10 Power generation per unit amplitude. reference here. The model experiment was operated in a two-dimensional water tank. However, the experimental data matched well with the theoretical calculation because there is a gap between the buoy model and the side walls, and the wavelength of the incident wave adopted in this experiment is longer than the length of the buoy model. However, the differences between the theoretical calculation and experimental value become remarkable in the long-period region with small dimensionless wave numbers, which is because when the movement increases, the nonlinearity of the response of the DE becomes significant, and the maximum displacement tends to be suppressed low as the restoring force increases. The theoretical calculation assumes the force response characteristic of the DE as that of a linear spring in the frequency domain. As the characteristic of Heave motion of the model, the displacement increases at the long-period side with small dimensionless wave numbers, which tend to be constant in the region with kl >1.5. In this experiment, the square type floating buoy is adopted and the horizontal movement of the buoy is completely restrained. The buoy performs two degrees of freedom movements, i.e. Heave and Pitch. Since the expansion and contraction of the DE attached to the wire are largely determined by the Heave component, it can be argued that the motion characteristic of the Heave component represents the extent of expansion and contraction of the DE. In order to confirm the power generation with respect to the expansion and contraction of the DE, the power generation amount E within one vibration is converted into that for the Heave per unit amplitude, as shown in Fig. 10. Thus, since the influence of the wave period is included, the time required for the power generation within one cycle is different. The horizontal axis is the dimensionless wave number. A peak emerges at kl=1.5, and despite slight fluctuations, the power generation energy in the whole wave number range is overall constant. In other words, the variation with respect to change of the wave number is relatively small when compared with the response function of Heave and the power of energy output, as shown in Figs. 6 and 9, respectively. The energy output should hardly change if the extension and contraction of the DE are constant. This is the intrinsic property of the dielectric elastomer for power generation, which can be manifested even when it is attached to the buoy as one part of the mooring apparatus.

7 Experimental Investigation on the Power Generation Performance of Dielectric 185 However, as discussed above, the power generation is identical when the displacement is the same, which means that the superiority of short-period incident waves will present when converted into the power of energy output. In order to maintain high efficiency of the power generation over a wide range of incident wave periods, it is necessary to generate a larger Heave at the side of long period with larger wave power. Generally, in the case of buoy setting in the experiment, except for the intrinsic period, long period waves benefit the Heave of the buoy, while the displacement tends to decrease at the side of short period. The DE itself plays a role of a mooring device, and its force response characteristic determines the Heave motion and the amount of power generation. Therefore, in order to tune the movement characteristics of Heave, it is necessary to develop a buoy device with a minimum size according to the spring-characteristics of DE. In this experiment, the power generation can be maximized via the optimization of Heave characteristics. It is a great achievement to achieve a maximum efficiency of 50% under this condition. In this work, the simple shaped buoy is adopted; however, it is of vital importance to quantitatively clarify the specific characteristics of wave power generation utilizing DE. Chiba et al. have demonstrated high power generation efficiency (70-90%) as a result of the optimization of equipment shape and the design of floating bodies [6, 7]. Also in this work, the importance of the optimization of the equipment shape and the design of the floating body was confirmed. Even in actual sea areas, if the optimum shape and design are matched, it is thought that high output and high efficiency power generation corresponding to the waves unique to the area can be realized. 5. Conclusions In this work, the potential application of the DE mounted on the simple buoy was studied, and the conclusions are summarized below. (1) The power generation efficiency improves as the dimensionless wave number of the short-period incident wave increases. () In this work, energy conversion efficiency about 50% is realized even for a simple buoy without considering the shape of the buoy within 0.7 s of the incident wave. In the future, high power generation with high efficiency is likely to be realized via the optimization of equipment shape and the buoy design, as indicated by Chiba et al. with an efficiency as high as 90%. (3) However, the result of () shows that the dependence of power generation output characteristics on the wave number and wave period is remarkable. (4) The extension and contraction of the DE are determined by the motion characteristics of the Heave of the simple shaped buoy model. Furthermore, in order to have freedom in tuning the motion characteristics of Heave, we found that a device capable of setting the spring characteristics of the DE to the smallest possible size with respect to the size of the floating body is necessary. (5) By optimizing the shape of the equipment and matching it with the optimum floating body, it may be possible to realize high power with high efficiency in actual sea areas. References [1] U.S Department of Energy, Energy Efficiency and Renewable Energy (EERE), World Energy Council. Accessed March 1, [] Miyazaki, T., and Osawa, H Search Report of Wave Power Devices. In Proceedings of the 007 spring Conference of the Japan Society of Naval Architects and Ocean Engineers, [3] Chiba., S., Waki, M., Pelrine, R., and Kornbluh, R. 008, Innovative Power Generators for Energy Harvesting Using Electroactive Polymer Artificial Muscles. In Proceedings of the SPIE, [4] Chiba, S., Prahlad, H., Pelrine, R., Kornbluh, R., Stanford, S., and Eckerle, J Electro Power Generation Using Electro Active Polymers (EPAM). In Proceedings of the 15th Japan Institute of Energy Conference, [5] Chiba, S., Waki, M., Masuda, K., Ikoma, T., Osawa, H.,

8 186 Experimental Investigation on the Power Generation Performance of Dielectric and Suwa, Y Innovative Wave Power Generator Using Dielectric Elastomers Artificial Muscle. In Proceedings of the World Hydrogen Technologies Convention-011, Scotland. [6] Chiba, S., Waki, M., Wada, T., Hirakawa, Y., Matsuda, K., and Ikoma, T Consistent Ocean Wave Energy Harvesting Using Electroactive Polymer (Dielectric Elastomer) Artificial Muscle Generators. Applied Energy 104: 497. Doi: /j.apenergy [7] Chiba, S., et al Elastomer Transducers. Advance in Science and Technology 97: [8] Pelrine, R., and Chiba, S Review of Artificial Muscle Approaches. In Proceedings of the Third International Symposium on Micromachine and Human Science, 1-9. [9] Kornbluh, R., Pelrine, R., Pei, Q., and Chiba, S., et al High-Field Electrostriction of Elastrometric Polymer Dielectrics for Actuation. In Proceedings of the SPIE, CA. [10] Waki, M., Chiba, S., et al., 017. Development of Wave Generation Module for Small Ships Using Dielectric Elastomer. Journal of Material Science and Engineering B7 (7-8): Doi: /161-61/ [11] Huang, J., Shian, S., Suo, Z., and Clarke, D Maximizing the Energy Density of Dielectric Elastomer Generators Using Equi-Biaxial Loading. Advanced Functional Materials 3: [1] Lin, G., Chen, M., and Song, D In Proc. of the International Conference on Energy and Environment Technology (ICEET 09), [13] Brouchu, P. A., Li, H., Niu, X., and Pei, Q Factors Influencing the Performance of Dielectric Elastomer Energy Harvesters. In Proc. SPIE 764 (1): 76J. [14] Vertechy, R., Papini Rosati, G. P., and Fontana, M Reduced Model and Application of Inflating Circular Diaphragm Dielectric Elastomer Generators for Wave Energy Harvesting. Journal of Vibration and Acoustics, Transactions of the ASME 137: [15] Koh, S. J. A., Zhao, X., and Suo, Z Maximal Energy That Can Be Converted by a Dielectric Elastomer Generator. Applied Physics Letters 94 (6): [16] Bortot, E., and Gei, M Harvesting Energy with Load-Driven Dielectric Elastomer Annular Membranes Deforming Out-of-Plane. Extreme Mech. Lett. 5: [17] Moretti, G., Fontana, M., and Vertechy, R Parallelogram-Shaped Dielectric Elastomer Generators: Analytical Model and Experimental Validation. Journal of Intelligent Material Systems and Structures 6 (6): [18] Brochu, P., Yuan, W., Zhang, H., and Pei, Q Dielectric Elastomers for Direct Wind-to-Electricity Power Generation. In Proc. of ASME 009 Conference on Smart Materials, Adaptive Structures and Intelligent System. [19] Zhou, J., Jiang, L., and Khayat, R Dynamic Analysis of a Tunable Viscoelastic Dielectric Elastomer Oscillator under External Excitation. Smart Materials and Structures 5 (): [0] Chiba, S., et al Simple and Robust Direct Drive Water Power Generation System Using Dielectric Elastomers. Journal of Material Science and Engineering B7 (1-): Doi: / /