Development of Fuel Cell System for Long Cruising Lange Autonomous Underwater Vehicle

Development of Fuel Cell System for Long Cruising Lange Autonomous Underwater Vehicle 3.4 Tadahiro Hyakudome, Takeshi Nakatani, Hiroshi Yoshida, Marine Technology and Engineering Center, JAMSTEC Toshihiro Tani, Hideki Ito, Koki Sugihara Mitubishi Heavy Industries, Ltd. Abstract The use of Autonomous Underwater Vehicles (AUVs) is now widespread among the underwater observation and survey. Because AUVs are not restricted by an umbilical cable, we think that AUVs have the most suitable form to collect ocean data or work depending on subjects of survey efficiently in the same way as the sea lives with various forms. For example, form to collect ocean data while cruising in a huge range, form to collect the topography data of the seafloor and under the bottom of the sea while cruising with following seafloor, form to work such as the collection of the rock of the seafloor or the setting of the sensor on the seafloor, form to collect ocean data slowly and carefully without moving the investigation spot against the current and so on. Underwater power source is one of very important elements to operate electrically driven underwater vehicles. Because there are no energy supply places in underwater. Many research and development about the underwater power source are carried out in all over the world. When the research and development of the power source, following things need to be considered: small and light weight, put in a pressure vessel or resist against water pressure, work against low water temperature, without vibration and noise, reliability and ease of maintenance. The power source has heaviest weight with the components of the underwater vehicles. The paper reported that concepts of the body design and model experimental results show specification of fuel cell system and several experimental results. Keywords Autonomous Vehicle, Shape, Power Source scientific and explore applications requiring detailed investigations. As for the investigation of the ocean earthquake and the survey of the ocean resource, the fields are very huge. For example, the northeastern Pacific Ocean off the coast earthquake that occurred on March 11, 2011, the fields of investigation are huge area such as 500km long and 200km wide. Figure 1 shows a concept of submarine earthquake investigation cruising by AUV. Several days are necessary to survey such a huge range. The investigation of the ocean often depends on the weather condition. In many cases, it is limited to a cruising schedule of support vessel. Therefore, AUVs do not depend on weather condition, and ability to cruise the underwater is expected for a long term. From the viewpoint of continuity of data to acquire, AUVs are wished to survey a wide range by one underwater cruise. Power source to guarantee that a vehicle operates at long time is necessary. The power source for this purpose becomes large. For such a reason, large payload is necessary for long cruising range type AUVs. In addition, a large payload is necessary for AUVs to equip several observation devices such as side scan sonar, multi beam echo sounder and so on, and to acquire several kinds of data at the same time. However, it is nonsense that even if the vehicle has large payload, and fluid drag becomes large. It is desirable for the vehicle to cruise long range as possible by limited energy. For such a reason, it is necessary that long cruising range type AUVs have large payload and low fluid drag form. I. INTRODUCTION Deep sea investigation becomes important for study about ocean earthquake, search for ocean resources and so on. Recently many underwater vehicles have begun to use for scientific and explore applications. These applications require stable posture or precise maneuvering of the vehicle for detailed investigations. Autonomous Underwater Vehicles (AUVs) are considered and chosen as suitable tool for conducting surveys concerning these needs. Many AUVs, such as Autosub, Thesus, REMUS and Hugin are being developed in European, North American countries. AUVs can do comprehensive surveys because the vehicle does not have to be connected to the support vessel by a tether cable. AUVs can move freely, because they do not need cable for power supply and communications. So, the vehicles are able to keep stable posture and maneuver precisely. AUVs can be used for 978-1-5090-5716-0/16/$31.00 2016 IEEE 165

187.5mm. The slenderness ratio d/l of the model of body part was 0.15. The drag coefficient Cd of the only body part was surmised 0.07. The vehicle model is equipped with front wings, rear elevators, vertical tail wing and thrusters then a drag coefficient comes larger than body part only, but the form was surmised low fluid resistance. There is a request to change cruising depth or altitude while the body keeps zero degree at pitch angle depending on observation. The front wing is necessary to satisfy this request. The thrusters are located to become outside the frontal projected area of the body to improve thrust efficiency. II. Fig. 1 Image of operation LONG RANGE CRUISING TYPE FORM A. Concept Tunas and dolphins can swim first in the ocean, whales can wander about huge range in the ocean, jellyfishes drift in the ocean, rays skim the seafloor and so on, sea lives got the most suitable form to let each life prosper. The use of Autonomous Underwater Vehicles (AUVs) is now widespread among the underwater observation and survey. Because AUVs are not restricted by a umbilical cable, we think that AUVs have the most suitable form to collect ocean data or work depending on subjects of survey efficiently in the same way as the sea lives with various forms. For example, form to collect ocean data while cruising in a huge range, form to collect the topography data of the seafloor and under the bottom of the sea while cruising with following seafloor, form to work such as the collection of the rock of the seafloor or the setting of the sensor on the seafloor, form to collect ocean data slowly and carefully without moving the investigation spot against the current and so on. C. Water Tank Test Water tank tests were took placed to measure fluid force for deriving a drag coefficient. Figure 2 shows the experimental set-up of measurement of drag coefficient of the model. The water tank test was took placed at Ocean and Underwater Engineering Tank of Research Institute for Applied Mechanics in Kyushu University. Specification of the tank is follows, 65m long, 5m width and 7m depth. The water tank has towed platform upon it. The model was installed in the water tank by this towed platform. The vehicle model has vertical tail wing only upper part, so the model was attached upside down for measuring vertical tail wing effect. Resistances when the model was towed at various speed was measured. A six degreeof freedom pylon load cell was put on the model. The drag coefficient was obtained by making fluid drag dimensionless at the frontal projected area. Figure 3 shows an experimental result. This figure shows the variation drag coefficient (Cdx) as a function velocity (U m/sec). In the figure, blue circles indicate experimental value, and blue line indicates estimated value. Drag coefficient close to constant with an increase in velocity. As a result, the long cruising range type form of the model has low fluid drag with Cd=0.13. B. Model In the case of streamlined revolution bodies, when a length of the body is L and a biggest diameter of the body is d, then d/l is called the slenderness ratio. A drag coefficient comes to a minimum when the slenderness ratio is from 0.2 and 0.25. And the thickness-length ratio of sea lives that swimming in the sea at high speed is around 0.25[6], [7], [8]. Here, the thicknesslength ratio of the blue whale is surmised 0.21. Therefore, the form of the blue whale which can cruise fast in spite of large body was adopted to a long cruising range type AUV. A model of long cruising range type vehicle was designed in consideration of the stated above and overall balance of the vehicle. Figure 2 shows a dimension of the mode. The model was assumed 1/8 scale of an actual vehicle. The length of the model is 1260.5mm, the maximum diameter of the model is 978-1-5090-5716-0/16/$31.00 2016 IEEE 166

Fixture Towed Platform Model Fig. 2 Dimension of the long cruising type model Water Fig. 3 The model in the water tank Fig.4 Result of measurement of drag coefficient III. POWER SOURCE A. Concept Underwater power source is one of very important elements. The power source has heaviest weight with the components of the underwater vehicles. When the power source becomes big 978-1-5090-5716-0/16/$31.00 2016 IEEE 167

in proportion to the scale up of the body, maneuverability and energy efficiency worsen of the vehicle. Therefore, it is important that the power source is small and light weight. Usually, many underwater vehicles used rechargeable battery such that the lithium-ion battery or the silver-zinc battery for power source. However, the case of long term operation type or long cruising range autonomous underwater vehicle, it needs many electric power supplies in proportion to operation time. To extend operating time, the battery must become heavier and lager. But it makes worse the maneuverability and energy efficiency of the vehicle. The power source need to work against low water temperature. The power source work without vibration and noise. Because low vibration and low noise environment is important not to interfere acoustic equipment such as observation devices or communication devices. The development of fuel cell power system for underwater power sources was planned in JAMSTEC since 1991. The fuel cell is a kind of electric generator using the chemical reaction of hydrogen and oxygen. It is able to generate electricity directly from chemical reaction without any intermediate steps. Among various fuel cell systems, solid Polymer Electrolyte Fuel Cell (PEFC) system is most suitable for power source of underwater vehicle. Because of the PEFC system generates electricity at efficiency over 50% and at low temperature about 60 degrees Celsius. And though the fuel cell is a kind of electric generator, mechanical noise and vibration is very small because the system need not using turbine. An AUV URASIMA adopts a closed cycle PEFC with metal hydride system as its power source to enable long range cruising. The vehicle achieved to cruise of 317km distance as power source by the fuel cell system. B. HEML Fuel Cell System Based on knowledge provided from such a result, the fuel cell and the gas storage system for next generation cruise type AUV are researched. The next generation cruising type AUV aims for cruise distance of several thousand kilometers. Since 2007, we started to develop a new concept fuel cell system for underwater power source. We decided development objectives to improve performance while solving the technical problems of the second generation fuel cell. There are start-up in short time, highly efficient system, no need blower for circulation of hydrogen and oxygen, not let hydrogen leak out, over 600 hours continuous run time, small system. Our technical goal is to achieve the system generation efficiency of over 60%. For this purpose, the fuel cell system need to have the following features: high performance cell, blower less, humidifier less, leak less, and a few minute start-up time. The system named after its features HEML (High-Efficiency Multi-Less) fuel cell system. Principle of HEML Fuel Cell system is shown in figure 5. One example of the statement about left image is shown in the table in figure 5. Only oxygen flow line is shown in figure to simplify explanation. The HEML Fuel Cell consist of two stack, the stack A and stack B, 4 valves (#1 to #4) and 2 liquid separators. In left image case, stack A is in the upper stream from oxygen source of supply. At first, #1 and #3 valves are opened and #2 and #4 are closed, oxygen gas flows from stack A through stack B. At this time, oxygen gas is used 50 % in stack A to generate electricity, and reactant water is drained, and humidified remaining oxygen gas is exhausted and supplied to stack B. Then reactant water is stored in stack B. As results, all gas is consumed. If the system keeps this condition, reactant water is over flow in stack B. Then after a few minutes, valve #1 and #3 is closed and valve #2 and #4 is opened. Then oxygen gas flows from stack B through stack as right image in the figure 5. The HEML Fuel Cell system can generate electricity continuously by doing such valve operation. C. 400W class HEML Fuel Cell System Test The HEML Fuel Cell system of 400W class in land test container is shown in figure 6. The specification of the system is listed in table 1. In the figure, Left side is reaction and generation part, right side is electric devices part. The time trend of generated voltage at endurance test is shown in figure 7. Load current was 20A constant. At this test, the HEML Fuel Cell system succeeded in 600 hours continuous generation. Output voltages of each cell were approximately constant. Operating pressure was also stable. After land test, we carried out sea trial. The towed vehicle was equipped with a main pressure vessel for the HEML Fuel Cell system, two 10 liters hydrogen and oxygen gas tanks, a pressure vessel for communication devices, a pressure vessel for battery, Synthetic Aperture Sonar (Max. 200W) and Hybrid ph-co2 sensor (Max. 4.5W) as load for the HEML Fuel Cell system. Maximum depth was 180m. The time trend of output voltage and current is shown in figure 8. The top graph is stack voltage. The bottom graph is current. The output voltage changed in response to load changes. During sea trial output voltage of the HEML Fuel Cell system had been very stable. D. 2kW class HEML Fuel Cell System Test After sea trial, we started develop 2kW class HEML Fuel Cell system to improve power capacity. The specification is listed in table 2. We confirmed the performance of the HEML Fuel Cell system such as Current versus Voltage curves (I-V curves). The I-V curve at land test shown in figure 8. Operating load ranges were 0A to 42A, and maximum power was 2,000W. The time trend of generated voltage and output power at endurance test is shown in figure 9. At this test, the 2kW HEML Fuel Cell system succeeded in 100hours continuous generation. Blue line denotes stack voltage and brown line denotes DC/DC converted Output voltage. Orange line denotes stack total power and DC/DC converted output power. Output voltage o were approximately constant. 978-1-5090-5716-0/16/$31.00 2016 IEEE 168

Switching Stack #1 #2 #3 #4 Reactant gas Reactant water Surplus gas Fig.7Time Trend of Stack Voltage at Endurance Test directly drained exhausted supplied O X O X humidified stored 100% B at 60 * in the stack consumed This table shows statement of left figure. * Outlet gas temperature at the upper stream A Table 2 2,000W class HEML Fuel Cell Specification Fig.5 Operating Principle of the HEML Fuel Cell System Size L.D.530 x 1,500L mm Output 2,000W Voltage 120V Temperature 60 degrees Celsius Efficiency Over 55% Gas Anode: Hydrogen Cathode Oxygen Reaction & Generation Part Electric Deice Part Fig.6 The HEML Fuel Cell System in Display Container Table 1 400W class HEML Fuel Cell Specification Size L.D. 600 x 900L mm Output 400W Voltage 24V Temperature 60 degrees Celsius Efficiency 60 % Gas Anode: Hydrogen Fig.8 I-V Characteristic of 2kW HEML FC Cathode Oxygen Fig.9 Time trend of output at 100hours running test on land 978-1-5090-5716-0/16/$31.00 2016 IEEE 169

Concluding Remarks In this paper, the results of tank test of cruising type form of the AUV, and test of new concept fuel cell system. The AUV shape and power source are the key function that enable long cruising range AUV operation for survey. The test results were fairy good and showed its possibility of practical use. REFERENCES [1] Aoki, T., et al, 2008, Advanced Technologies for Cruising AUV URASHIMA. Int. Journal of Offshore and Polar Engineering, Vol. 19, No.2, pp.81-90. [2] Kristensen, J. S., et al, 1998 Autosub : an autonomous unmanned submersible for ocean data collection, Electronics & Communication Engineering Journal,Vol.7 issue 3 pp.105-114. [3] Ferguson, J. S. (1998). The Theseus autonomous underwater vehicle. Two successful missions, Underwater Technology 98 Proc. 1998 Symposium pp.109-114. [4] Stokey, R. P. et al. (2005). Development of the REMUS 600 autonomous underwater vehicle, OCEANS 2005 Proc. of MTS/IEEE, Vol.2 pp. 1301-1304. [5] Jalving, K., et al, 2004, A toolbox of aiding techniques for the HUGIN AUV integrated inertial navigation system. Modeling, Identification and Control, Vol. 25, no 3, pp.173-190. [6] JSME, 2007, Mechanical Engineers Handbook, The Japan Society of Mechanical Engineers. [7] Kato, K., 1982, Guidance of the aircraft dynamics, Tokyo Univ. [8] Neaman, J. N., 1977, Marine hydrodynamics, The MIT Press, Cambridge, MA, USA. [9] Silvestre, C., Pascoal, A., 2006, Depth control of the INFANTE AUV using gain-scheduled reduced order output feedback, Control Engineering Practice, www.sciencedirect.com. [10] Collar, P. G. et al. (1995). Autosub : an autonomous unmanned submersible for ocean data collection, Electronics & Communication Engineering Journal,Vol.7 issue 3 pp.105-114. [11] Kristensen, J. et al. (1998). Hugin-an untethered underwater vehicle for seabed surveying, OCEANS 98, Conf. Proc. Vol.1 pp.118-123. [12] Stokey, R. P. et al. (2005). Development of the REMUS 600 autonomous underwater vehicle, OCEANS 2005 Proc. of MTS/IEEE, Vol.2 pp. 1301-1304. [13] Zerr, B. et al. (2005). Sidesacan sonar image processing for AUV navigation, OCEANS 2005 - Europe Proc. of MTS/IEEE, Vol.1 pp. 124-130. [14] Robert, A. et al. (2006). Advance in AUV remote-sensing technology for imaging deepwater geohazards, Geoscienceworld, The Leading, Vol.25 no.12 pp. 1478-1483. [15] Kirkwood, W. J. et al. (2004). Mapping payload development for MBARI s Dorado-class AUVs, OCEANS 04 Proc. of MTS/IEEE TECHNO-OCEAN 04, Vol.3 pp. 1580-1585. 978-1-5090-5716-0/16/$31.00 2016 IEEE 170