汽车外文文献1

400 Commonwealth Drive, Warrendale, PA 15096-0001 U.S.A. Tel: (724) 776-4841 Fax: (724) 776-5760 Web: 2006-01-0218A Hybrid Powertrain Provided with anEmulated Fuel Cell System and a Battery Pack: Experimental ResultsMarco Santoro, Manlio Pasquali and Gianfranco PagniENEA Research Center “Casaccia”Luca SoleroUniversity of Rome “ROMA TRE”Reprinted From: Applications of Fuel Cells in Vehicles 2006(SP-2006)2006 SAE World CongressDetroit, MichiganApril 3-6, 2006The Engineering Meetings Board has approved this paper for publication. It has successfully completed SAE's peer review process under the supervision of the session organizer. This process requires a minimum of three (3) reviews by industry experts.All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior written permission of SAE.For permission and licensing requests contact:SAE Permissions400 Commonwealth DriveWarrendale, PA 15096-0001-USAEmail:permissions@Tel:724-772-4028Fax:724-776-3036For multiple print copies contact:SAE Customer ServiceT el:877-606-7323 (inside USA and Canada)T el:724-776-4970 (outside USA)Fax:724-776-0790Email:CustomerService@ISSN 0148-7191Copyright © 2006 SAE InternationalPositions and opinions advanced in this paper are those of the author(s) and not necessarily those of SAE. The author is solely responsible for the content of the paper. A process is available by which discussions will be printed with the paper if it is published in SAE Transactions.Persons wishing to submit papers to be considered for presentation or publication by SAE should send the manuscript or a 300 word abstract to Secretary, Engineering Meetings Board, SAE.Printed in USAABSTRACTThe increasing concern about polluting emissions of vehicles, the cost and the limited availability of petroleum are urging many researchers in the world to develop innovative and more energy-efficient traction powertrains. In order to improve the energy conversion efficiencies into the vehicle, auto manufacturers are looking at new energy-storage-system technologies and fuel converters with great interest. The University of Rome“ROMA TRE” and ENEA (the Italian National Agency for New Technologies, Energy and the Environment) have jointly developed and tested a Fuel Cell Emulator, able to feed a vehicle powertrain just like an actual fuel cell should do. Aim of this work was to substitute the fuel cell with a much more reliable device and test different powertrain layouts and control strategies. This approach allows comparing different powertrains by simply resizing their components. INTRODUCTIONComprehensive studies (e.g. [1]) have compared several powertrain layouts and technologies to a reference conventional vehicle - provided with gasoline engine and automatic transmission - over many driving schedules. The cycles with low power demand (low speed or stop-and-go operations) appear to be the most suitable for hybrid operations. These results are logical considering the sources of savings for hybrid vehicles: regenerative braking, no engine idling, and better powertrain efficiency at low power demands.When considering the introduction of advanced vehicles, a complete well-to-wheel evaluation must be performed to determine the potential impact of a technology on carbon dioxide and Green House Gases (GHGs) emissions. A well-to-wheels (WTW) analysis of a vehicle/fuel system covers all stages of the fuel cycle, from energy feedstock recovery (well) to energy delivery at the vehicle’s wheels. A WTW analysis is also referred to as a fuel-cycle analysis. Considering fuel economies for the different vehicle’s configurations on the U.S. Combined cycle (including FUDS and FHDS) it is clear that substantial gains can be achieved through dieselisation or hybridisation. The results considering well-to-pump (WTP), pump-to-wheels (PTW) and WTW powertrain efficiencies for the combined cycle imply that: • Dieselisation can increase the efficiency by more than 20%,• Hybridisation alone leads to an improvement of more than 30%,• Dieselisation and hybridisation lead to an improvement of more than 50%,• A gain of more than 150% can be obtained with the hybrid fuel cell. The hybrid fuel cell configuration combines high fuel-cell-system efficiency and regenerative braking to achieve the highest fuel economy.Considering the energy loss during the NEDC (New European Driving Cycle) cycle for each component for the configurations considered, the same analysis shows that the engine is, by far, the least efficient of the components (accounts for more than 75% of the total losses for the reference case). Fuel cell vehicles lose only half the energy of the best parallel case. Moreover, a hybrid fuel cell powertrain consumes less energy than a system containing only a fuel cell. The weight advantage of the fuel cell system is not sufficient to compensate for the loss in regenerative energy. Fuel cell vehicles supported by an energy storage system achieve the highest fuel economy. This is why regenerative braking energy may be stored and the fuel cell can often operate in high-efficiency regions.Other studies [2] pointed out that on a fixed time budget, vehicle miles travelled by vehicle vary inversely with the average driving speed. In other words, personal vehicles based in congested urban areas may accumulate fewer miles of driving per year than suburban-based vehicles. Thus, owners of hybrid vehicles living in congested areas may drive less than hybrid owners living in suburban area, nullifying the large fuel economy2006-01-0218A Hybrid Powertrain Provided with an Emulated Fuel CellSystem and a Battery Pack: Experimental ResultsMarco Santoro, Manlio Pasquali and Gianfranco PagniENEA Research Center “Casaccia”Luca SoleroUniversity of Rome “ROMA TRE”Copyright © 2006 SAE Internationaladvantage they hold over comparable conventional vehicles. Nonetheless, we believe that the extensive use of hybrid vehicles should dramatically improve the air quality in urban areas. This is thanks to the degrees of freedom these powertrains are provided with which allow to design low tailpipe-emission vehicles especially conceived for urban traffic.Because of their high efficiency and low emissions,fuel cell vehicles are undergoing extensive research and development. A clean vehicle, such as a fuel cell vehicle, does not mean that there are no emissions from a well-to-wheel perspective. When producing hydrogen from reforming at a station, fuel cell vehicles have a lower advantage in terms of efficiency and emissions. Emerging methods for generating hydrogen by exploiting renewable energy sources (e.g. [3]) will soon allow considering fuel-cell-propelled vehicles as actual Zero Emission Vehicles (ZEVs).AIM OF THE WORKAim of our work was the realization of a fuel cell emulator able to generate the voltage/current relationship the fuel cell would supply,and investigate different control approaches of a hybrid powertrain containing a battery pack. The emulator has to be a much more reliable “fuel converter” than the actual fuel cell. During testing, the emulator has been constrained to operate in steady state conditions, i.e. the fuel cell current rate has been continuously monitored and constrained.A major European automobile manufacturer has provided us with a Simulink model of a 60-kW PEM fuel cell (Fig. 1). The model is mostly empirical,relying on fuel cell component input/output relationships measured in the laboratory and quasi-static, using data collected in steady state tests. Most of these data are contained in look-up tables. By changing the number of elementary cells in the model, we have modified the size of the fuel cell, and we have therefore tested and compared the behavior of different-size fuel cells.We have modified at ENEA labs in Rome a battery cycler, i.e. an AC/DC converter that is able to charge and discharge a battery pack according to a chosen voltage/current law. Such modified battery charger can be driven from external signals, calculated by the modeldescribed above.Fig. 1: 60kW PEM fuel cellTHE POWERTRAINWe have shown in previous papers [4, 5] the availability and effectiveness of our Multi Input Power Electronic Converter (MIPEC). This device manages the bidirectional power flow from three different sources for feeding a traction drive (Fig. 2).Fig.2: MIPEC manages the power flowing from a BatteryStorage Unit (BSU), an UltraCapacitor tank (UC) and the Fuel Cell (FC) to the Traction Drive The powertrain does not contain any ultracapacitor and therefore just two inputs of MIPEC can work in this test.The fuel cell system has been emulated thanks to the battery cycler (3-phase AC/DC converter) and a suitable DC link capacitive filter. An x86 real time microprocessor platform (PC 104) regulates the output voltage of the FC emulator as function of the output DC current and the operating temperature.The microprocessor controls the emulator on the basis of the fuel cell model: at a given current generated by the emulator, the instantaneous output voltage can differ on dependence - this is evident at low-load currents - on the operating conditions (Fig. 3).Fuel Cell Voltage vs. Current102030405060708090100110120130140150Current [A]Fig.3:Minimum, mean and maximum fuel cell Voltagevs. mean supplied Current The powertrain is completed by a VRL A battery pack and a traction drive (induction motor with peak power of 30kW and continuous power of 15kW), coupled to afour-quadrant-operation dynamometer (Fig. 4).Fig. 4: 15kW continuous-, 30kW peak-power tractiondrive, mechanically coupled to dynamometerHOW THE TEST BENCH OPERATESFig. 5: Layout of the powertrain under testingReferring to Fig. 5, PC104 microprocessor and boards, with a time step of 0.1s xCalculate hydrogen consumption, fuel cell temperature and voltage in dependence on the current requestedx Drive the cycler by imposing the voltage the fuel cell would supplyx Calculate the batteries state of charge and their maximum allowable currentxCalculate and send to the MIPEC DSP the maximum current and the maximum current rate the fuel cell could meetx Acquire fuel-cell emulator current and voltage xAcquire batteries current and voltageThe MIPEC DSP xReceives information from the PC104 microprocessor concerning maximum fuel cell current, maximum battery current, maximum rate of the fuel cell currentxDrives MIPEC switching devices (i.e. calculates their duty cycles length) and thus determines the current flows from batteries and fuel cell.Torque and speed from the drive are acquired with a time step of 1s during testing.Main aim of the powertrain control strategy -this algorithm is implemented in the PC104 boards -is to limit the fuel cell power rate. This result can be gained by introducing an energy storage system which allows supporting the rapid change of power flows over the driving schedule. Moreover, the VRL A battery pack allows recovering energy during vehicle braking and coastdown.The adoption of a hybrid configuration allows decreasing the cost of the powertrain: a fuel cell of lower size - supported by an energy storage system - can feed a vehicle with the same performance that should be met by a vehicle provided with a full-size one.Varying the quantity of elementary cells in the model we have rescaled the fuel cell size and tested the behavior of powertrains propelled by a 7-, 15-, and 22-kW fuel cell,with the same battery pack at three different initial SOC values.Fig.6shows how the powertrain power flows are qualitatively managed. When the vehicle accelerates, the fuel cell meets the power request (Pload ) like a“sluggish” genset. Most of the power is supplied by the battery pack in this phase.At the end of the acceleration phase, the fuel cell can recharge the battery pack (Pbatt is negative). During braking or coastdown the kinetic energy is stored in the battery pack and the fuel cell is turned off.P load P batt P fcPtFig. 6: Power flow controlThe fuel cell model contains the maximum allowable fuel cell current and current rate (Figs. 7and 8,respectively)vs. the fuel cell stack temperature.10203040506070102030405060708090100110120temperature [°C]max current [A]Fig. 7: Maximum fuel cell current vs. temperature 01020304050607010152025303540temperature [°C]max di/dt [A/s]Fig. 8: Maximum fuel cell current rate vs. temperatureThis information is sent from the PC104 boards to the MIPEC DSP, and MIPEC monitors the FC current and the battery voltage. MIPEC splits the power request between fuel cell and batteries and is able to accurately constrain both the fuel cell current and its rate.The recharge current supplied by the fuel cell is constrained in dependence on the batteries state of charge (Fig. 9).Fig. 9: Maximum recharge current vs. SOCTESTING THE POWERTRAINThe powertrain has been tested over the urban section of the NEDC (Fig. 10), consecutively repeated 20 times.By this way, we were able to observe the system evolution at varying temperature,batteries state ofcharge, etc. etc.Battery socFig. 10: Urban segment (ECE 15) of the New EuropeanDriving Cycle Figs.11and 12 show the power flows in the vehicle over one NEDC urban section at cold and warm fuel cell,respectively. It is evident the fuel-cell power dependence on the operating temperature: the cold fuel cell is able to supply a maximum power of 8 kW (Fig. 11). The warm fuel cell can deliver a maximum power of nearly 12 kW (Fig.12).The batteries supply in both cases the power necessary to meet the driving cycle.The increase of hydrogen consumption is due to the fuel cell temperature increase. Points A and A 1,B and B 1,C and C 1, respectively show the hydrogen consumption in the same instant of two consecutive driving cycles (Fig. 13). A hydrogen vent (0.3g/s for 2s in the 7-kW fuel cell and 0.5g/s for 2s in the 22-kW fuel cell), depending on the past hydrogen consumption, is needed for expelling water from the distribution channels.101214time [s]Fig. 11: Power flows over the first NEDC urban section.The fuel cell is cold30403060308031003120314031603180-22468101214power [kW]P loadP battP fcMax pfc(T)time [s]Fig. 12: Power flowing over one NEDC urban section.The fuel cell is warmtime [s]Fig. 13: Instantaneous hydrogen consumption over twoconsecutive NEDC urban sectionsThe final SOC of batteries is nearly the initial one (Fig. 14).The total hydrogen consumption (Fig. 15) is therefore the actual energy consumption over the testing schedule. The SOC is calculated according to an algorithm that has been developed in collaboration with the University of Pisa (Italy). It takes into account the variation of batteries capacitance in dependence on the battery current and temperature.Fig. 14: SOC variation during testingtime [s]Fig. 15: Hydrogen consumptionPotenza [kW]P e r c e n t u a l e s u i C a m p i o n i T o t a l i A c q u i s i t i [%]Fig. 16: Fuel cell operating points distribution (7 kW,SOC = 0.6)Potenza [kW]P e r c e n t u a l e s u i C a m p i o n i T o t a l i A c q u i s i t i [%]Fig.17: Fuel cell operating points distribution (22 kW,SOC = 0.6) Figs.16 and 17 depict the distributions of the acquired fuel-cell emulator operating points and are referred to a fuel cell power of 7 kW and 22 kW, respectively. SOC is 0.6in both cases. The smaller fuel cell has to recharge more often the batteries, as they provide most of the acceleration power and are discharged. The fuel cell thus operates mostly in its high efficiency region,and although generates meanly more power than the 22-kW fuel cell, its hydrogen consumption is lower.The hydrogen consumption is 179g for the 7-kW fuel cell and 206g for the 22-kW one.CONCLUSIONThe development of a fuel cell emulator allows testing in a reliable and affordable way various powertrain configurations. By this way we are able to optimize experimentally the control strategies. The knowledge of the optimal control strategy allows realizing a vehicle provided with an actual fuel cell. In order to gain such results, a reliable emulator - whose performance has been compared with the reference fuel cell - is necessary.Moreover,this approach allows emulating any genset.The powertrain testing over the ECE 15 cycle allows to recognize the source of savings that the powertrain hybridization has generated,when the vehicle meets an urban schedule. The 22-kW powertrain is able to supply all the power the vehicle needs, and the batteries operate only until the fuel cell is cold. As soon as the fuel cell has reached its operating temperature, it is able to provide the needed power and to meet the dynamic behaviour the cycle requires.Fig. 17 shows that the 22-kW fuel cell mainly operates in the 0.5-1.5kW range. This power is nearly the power of the fuel cell accessory loads, and the fuel cell thus operates in low-efficiency regions. The 7-kW fuel cell cannot always meet the whole traction power and the batteries have to compensate this gap of power. The batteries must be recharged periodically by the fuel cell,e.g. when the vehicle is idling. This is why the 7-kW fuel cell mostly operates in high efficiency regions (Fig. 16).57.558.559.5These conclusions refer to the driving cycle here considered. If we suppose to analyze high-speed schedules (e.g. highway driving cycles), we will obtain completely different results. The optimal drivetrain sizing depends heavily from the driving cycle under testing. In each case, the current fuel-cell stack prices suggest to seriously consider hybrid drivetrains.REFERENCES1. A. Rousseau, P. Sharer “Comp aring Ap p les toAp p les: Well-to-Wheel Analysis of Current ICE and Fuel Cell Vehicle Technologies”, Proceedings SAE 2004 World Congress, March 8-11, 2004. Detroit, MI, U.S.A.2. D. Santini, et al. “Hybrid Electric Vehicle TechnologyAssessment: Methodology, Analytical Issue and Interim Results”, ANL Report ESD/02-1 (2001)3.http://www.enea.it/com/ingl/solar/index.html4. Santoro M., Puccetti A., Pasquali M., Solero L.,L idozzi A. “A Novel Multi-Inp ut Power Electronic Converter for Automotive A p p lications”, 11thInternational Power Electronics and Motion Control Conference EPE – PEMC, September 2-4, 2004, Riga, Latvia5. Di Napoli A., Crescimbini F., Solero L., L idozzi A.,Pede G., Santoro M., Pasquali M. “Multi Input Power Electronic Converter for Automotive Ap p lications”,AutoTechnology, Volume 4, December 2004, ISSN 1616-8216CONTACTMarco Santoro Ph.D., Electronic Engineer, currently joins the Department of Robotics at ENEA Research Center“Casaccia”, Rome, Italy. He can be reached at marco.santoro@casaccia.enea.it.DEFINITIONS, ACRONYMS, ABBREVIATIONS DSP:Digital Signal ProcessorENEA: Ente per le Nuove tecnologie, l’Energia e l’Ambiente (Italian National Agency for NewTechnologies, Energy and the Environment) FHDS: Federal Highway Driving ScheduleFUDS: Federal Urban Driving ScheduleGHG: Green House GasMIPEC: Multi Input Power Electronic ConverterNEDC: New European Driving CyclePEM: Proton Exchange MembranePTW: Pump To WheelsSOC: State Of ChargeVRLA: Valve Regulated Lead Acid (battery)WTP: Well To PumpWTW: Well To WheelsZEV: Zero Emission Vehicle。