SAE TECHNICAL PAPER SERIES

2003-01-1269

The 2002 Ohio State University FutureTruck The BuckHybrid002

Mary Gilstrap, Guillaume Anceau, Chris Hubert, Matt Keener, Shawn MidlamMohler, Kevin Stockmeier, Jean-Marie Vespasien, Yann Guezennec, Frank Ohlemacher and Giorgio Rizzoni
The Ohio State University

2003 SAE World Congress Detroit, Michigan March 3-6, 2003

400 Commonwealth Drive, Warrendale, PA 15096-0001 U.S.A. Tel: (724) 776-4841 Fax: (724) 776-5760 Web: www.sae.org

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 Permissions 400 Commonwealth Drive Warrendale, PA 15096-0001-USA Email: [email protected] Fax: 724-772-4891 Tel: 724-772-4028

For multiple print copies contact: SAE Customer Service Tel: 877-606-7323 (inside USA and Canada) Tel: 724-776-4970 (outside USA) Fax: 724-776-1615 Email: [email protected] ISSN 0148-7191 Copyright 2003 SAE International Positions 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 of a proposed manuscript to: Secretary, Engineering Meetings Board, SAE. Printed in USA

2003-01-1269

The 2002 Ohio State University FutureTruck The BuckHybrid002

Mary Gilstrap, Guillaume Anceau, Chris Hubert, Matt Keener, Shawn Midlam-Mohler, Kevin Stockmeier, Jean-Marie Vespasien, Yann Guezennec, Frank Ohlemacher and Giorgio Rizzoni
The Ohio State University
Copyright 2003 SAE International

ABSTRACT This year, in the third year of FutureTruck competition, the Ohio State University team has taken the challenge to convert a 2002 Ford Explorer into a more fuel efficient and environmentally friendly SUV. This goal was achieved by use of a post-transmission, charge sustaining, parallel hybrid diesel-electric drivetrain. The main power source is a 2.5-liter, 103 kW advanced CIDI engine manufactured by VM Motori. A 55 kW Ecostar AC induction electric motor provides the supplemental power. The powertrain is managed by a state of the art supervisory control system which optimizes powertrain characteristics using advanced energy management and emission control algorithms. A unique driver interface implementing advanced telematics, and an interior designed specifically to reduce weight and be more environmentally friendly add to the utility of the vehicle as well as the consumer appeal. The BuckHybrid002 achieves 23 mpg city and 32 mpg highway, meets ULEV emissions standards and meets or exceeds all of the performance and customer demands embodied in the stock Explorer. OVERVIEW OF BUCKHYBRID002 The BuckHybrid002 is a charge-sustaining, parallel, post-transmission hybrid-electric vehicle with a relatively simple powertrain architecture, which facilitates packaging, manufacturability and provides operational flexibility and efficiency benefits. A layout of the overall vehicle schematic can be seen in Figure 1. In such a drivetrain, the electric motor (EM) is directly coupled to the driveshaft after the transmission and is capable of adding or subtracting torque at the wheel. This torque is added to that provided by the engine at the output of the multi-gear transmission. In such a configuration (neglecting mechanical efficiencies here

for simplicity), the kinematic relation powertrain operation is given below:

governing

Tw = ((t (n g ) TICE ) + b TEM ) f

where Tw is the torque at the wheel, ng is the selected (discrete) gear ratio, TICE and TEM are the ICE and EM torque, respectively, b is the EM speed ratio (fixed) and f is the final differential gear ratio.

Tice

t(ng)

f Diff

Tw

ICE

Transmission Tem

Belt

Wheels

E_MOTOR
Figure 1: Kinematic Layout of a Post Transmission Parallel HEV Drivetrain The design, optimization, testing and integration of the various facets of the BuckHybrid002 are described in the following sections. TEAM STRUCTURE The 2002 Ohio State FutureTruck team is comprised of both undergraduate and graduate students. Backgrounds and areas of interest are extremely varied. Some team members work regularly on their cars; others had very little prior experience working on cars before joining the team. Team members come primarily from mechanical and electrical engineering, with representatives also from other engineering degrees as well as industrial design. Specific areas of interest and

research include controls, diesel engines, exhaust aftertreatment, electric machines, and design. Organizing this diverse group to make the rather drastic changes in a very tight time frame proves to be one of the greatest challenges of the competition. The Ohio State team has met this challenge by dividing the team into a number of groups, including: battery management system, controls, electric motor, electrical systems, engine, simulation/modeling, transmission, vehicle/driver interface, and interior design. Each team has a leader, responsible for directing the individual members and ensuring that schedules are met. These individual team leaders then form a core team, which meets every week to review progress and identify critical paths. Additionally, the core team helps schedule work that overlaps into several different areas, as most of the projects do. Also important to any successful project is identification of a timeline and effective use of this timeline to establish milestones, deadlines, and lead times. Shown below in Table 1 is a highly simplified version of the BuckHybrid002 timetable. For team use, a complete list of even relatively small-scale projects was made, and all critical paths were identified, allowing adherence to the schedule and successful completion of the vehicle. The BuckHybrid002 actually participated in the Tour de Sol competition, which forced the vehicle to be ready 4 weeks before the FutureTruck competition. Table 1: BuckHybrid002 Project Timetable Task Vehicle Modeling Design Component Selection Component Procurement Component Arrival Component Testing Assembly Control Development Control Tuning Vehicle Testing Su '01 Fa '01 Wi '02 Sp '02

potential availability for procurement, cost, time, complexity and packaging constraints.

The process was initiated by a clear identification of the targets to be met or exceeded for performance, fuel economy and emissions. The major design goals are summarized below in Table 2. The next step was to identify potential best-in-class component technologies and vehicle/powertrain architectures and evaluate all these in light of the multiple objectives. Pros and cons of various technologies, as well as upfront considerations of availability, suitability, cost, and prior experience were debated at length by members of the core team. Table 2: OSU BuckHybrid002 Design Goals Vehicle Characteristic Weight Fuel Economy Emissions Eighth Mile Time Towing Modes of operation Telematics/DIS BuckHybrid002 Design Goal 5000 lbs 30 mpg ULEV 10.5 sec 3000 lbs Hybrid, Engine Only Internet connectivity, vehicle systems monitoring, user friendly interface

Simulation at various levels of complexity was brought into the picture early on to perform preliminary evaluation of powertrain requirements to meet performance targets and lead relatively quickly to a rough sizing of components. A diversity of powertrain architecture: pre- and posttransmission parallel, series, etc were considered early on, particularly with respect to the rough sizing of individual components to meet performance targets with different powertrain configurations. Based on extensive discussions and heated debate, preliminary simulations, past experience, and ease of implementation, a posttransmission parallel HEV architecture was chosen (see Figure 1). Once these basic design decisions were made, the focus shifted to the individual technology/components selection, supported by more detailed simulations and serious considerations of component availability, cost and packaging. The bulleted list of factors seen above was considered for each major component.

VEHICLE DESIGN PROCESS

The vehicle design process was comprised of three tightly interconnected elements: vehicle performance simulation, architecture selection and component selection. The design and optimization process is highly interactive and necessitates multiple iterations. Furthermore, this optimization process is constrained primarily by the following factors:

4 3

7 8

10

11

12

1. 2. 3. 4. 5. 6.

Battery Box 1 7 Batteries 8 Gallon Removable Fuel Tank Battery Box 2 5 Batteries Battery Box 3 3 Batteries Electric Motor/Belt Box Battery Box 4 7 Batteries

7. Inverter 8. Battery Box 5 5 Batteries 9. Transfer Case 10. Catalyzed Diesel Particulate Filter 11. Manual Transmission 12. VM-Motori Diesel Engine

Figure 2: BuckHybrid002 Packaging Diagram

POWERTRAIN OVERVIEW Shown above is the overall layout of the BuckHybrid002, with all major components included. As described above, each component was carefully selected, as outlined in the following sections. POWERTRAIN - MECHANICAL INTERNAL COMBUSTION ENGINE The first, and most important, decision to be made when creating a low-emission, fuel-efficient vehicle is that of the engine. Three engines were initially considered, a flex-fuel 3.0L V6 Ford Ranger SI engine running on E-85, a 2.4L I5 Fiat CIDI engine, and a 2.5L I4 VM-Motori CIDI engine. The 3.0L V6 Ranger engine run on E-85 produces very low emissions and low Green House Gases. However, the overall torque, fuel efficiency, the teams lack of experience with ethanol, and the lack of a throttle by wire (which facilitates supervisory HEV control) made this engine unacceptable. The 2.4L I5 Fiat diesel produced adequate torque and was much more efficient than the 3.0L, but the length of the engine made implementation impossible. The 2.5L I4 VM-Motori diesel produced similar torque and efficiency as that of the Fiat, with an overall shorter length and it produced significantly lower emissions.

The engine selected for the 2002 OSU FutureTruck was the 2.5L VM engine. The current generation of high speed, Diesel engines developed for passenger cars offer exceptional performance, economy with low noise and significantly improved emissions. The following table outlines some of the key features of the engine. Table 3: ICE Features Engine Parameter Value Manufacturer Block Configuration Fuel Injection System Max Power Max Torque Minimum BSFC Maximum Efficiency Max Speed Weight Additional Features VM Motori 2.5 liter, in-line four Common rail, direct injection 103 kW @ 4000 RPM 333 N.m @ 2000 RPM 195 g/kWh 44% 4800 RPM 220 kg Turbocharged Intercooled Cooled EGR Dual Overhead Cams 16 valves Twin balance shafts

ENGINE EFFICIENCY The diesel engine offers exceptional fuel efficiency. The peak efficiency of the engine is 44% and occurs at 2250 rpm and 140 Nm of torque. As the diesel engine does not suffer from the parasitic pumping losses of a gasoline engine, it maintains good efficiency even at low loads. The following contour plot shows the efficiency map of the 2.5L engine used in the BuckHybrid002. It is obvious why the diesel engine offers such exceptional fuel economy potential even before considerations of hybridization.
Engine Brake Efficiency
350

affected slightly, lowering the value by 1-3%. Given the high torque output of the engine, this reduction in torque was more than offset by the other benefits of using biodiesel. REGULATED EMISSIONS It is critical to access the effect of biodiesel on emissions. Given the difficulty of controlling NOx emissions in a diesel engine, it was important to ensure that the fuel selected did not create elevated levels of NOx emissions. As the literature cites increases in NOx emissions associated with biodiesel blends, a careful evaluation was warranted. The following figure shows the engine-out emissions of B20 and B50 relative to the reference fuel, which is 100% low sulfur diesel. The results are a composite of all of the operating points tested in the test schedule. All of the test points were weighted equally to determine the composite change.
Effect of Biodiesel Blend on Emissions 0.2 0.175 0.15 0.125 0.1 0.075 0.05 0.025 0 HC CO Nox Emission Species

250 Engine Torque (N-m)

200

Peak Efficiency = 44%

5 0.3

150
0.4
0.35

0. 35
0.35
0.35

300

100
0.3

0.3
0.25
0 5 .2
0.25

Change Relative to Low-Sulfur Diesel

0.3

B20 B50

0.2

50

0.2

0 1000

1500

2000

2500 Engine Speed (rpm)

3000

3500

4000

Figure 3: Engine Brake Efficiency FUEL SELECTION Three fuels were investigated for use in this years FutureTruck, including low sulfur diesel, 20% biodiesel blended with low sulfur diesel (B20), and 50% biodiesel blended with low sulfur diesel (B50). The use of low sulfur diesel is critical for the application of advanced diesel aftertreatment devices as it can accumulate on active catalyst sites blocking catalytic activity. Low sulfur diesel also has yields lower sulfur compound emissions, which have been shown to have adverse environmental impact in the form of acid rain (although unregulated). To compare the three available fuels, a test schedule was run with each fuel on an engine dynamometer while emissions, torque, and other important engine parameters were measured. The test schedule consisted of twenty points clustered in the most frequented region of the torque speed plane. These test points served to evaluate the effect of each of the fuels on emissions and torque production. ENGINE PERFORMANCE Engine performance using biodiesel blends was very satisfactory. The engine behaved virtually identically with low sulfur diesel, B20, and B50. The peak torque capability of the engine was

CO2

Figure 4: Effects of Biodiesel Blend on Emissions The figure clearly shows that NOx emissions are largely unaffected by choice of fuel blend. CO and HC emissions both show elevated levels, but not a significant increase. Diesel engines generally have exceptionally low CO and HC emissions, so, even a modest increase in these emissions is acceptable. In addition, CO is more readily oxidized by a catalytic converter than HC due to its low molecular weight when compared with heavy hydrocarbons. In light of this, it would seem that B50 would be the better choice. GREENHOUSE GAS EMISSIONS The figure also indicates that no net increase or decrease in tailpipe CO2 emissions was observed. As CO2 emissions are directly related to fuel consumption, this indicates that the fuel efficiency of the engine is unchanged by the choice of fuel blend. Although CO2 emissions are not the only greenhouse gas under consideration, CO2 is by far the dominant contributor to global warming potential. To truly calculate the emission impacts of various fuels, one must also take into account the upstream emissions of each of the fuels. As Biodiesel is made from biomass

feedstock, it serves as a CO2 sink. The leftover biomass after the harvest serves to sequester carbon in soil. This shows up as a net negative CO2 credit for each gallon of Biodiesel consumed. This CO2 credit is subtracted from the vehicles tailpipe CO2 emissions to get a total cycle contribution to CO2 emissions. As B50 contains the largest amount of Biodiesel, it yields the best overall CO2 emissions, and therefore the lowest greenhouse potential. FINAL FUEL SELECTION For the reasons cited above, the choice of fuel blend was clearly B50. The following list outlines the key reasons: Minimal impact on engine performance Minimal impact on NOx emissions Minor increase in HC emissions Negligible impact on fuel economy Negative CO2 credit

emissions. Careful metering of the water is necessary to prevent poor engine operation; elevated CO, HC, and smoke emissions; and excessively increased fuel economy. Metering the water is controlled via the supervisory controller based on the engine operating point. HYBRID CONTROL STRATEGY The hybrid control strategy is also used to mitigate NOx emissions. The strategy is outlined extensively in the controls section. The basic concept is, by inclusion of NOx formation into the equivalent fuel usage, it is possible to pick an optimal torque split which uses the electric motor to enable the operating point of the engine to stay away from regions of high NOx emissions. CATALYZED DIESEL PARTICULATE FILTER A catalyzed, wallflow, diesel particulate filter (CDPF) is used primarily to control HC, CO, and PM emissions. The catalyst also has a slight NOx reduction capacity due to a De-NOx reaction. An additional feature of the CDPF is that it eliminates the need for a muffler, which saves cost, weight, and space. Catalyzing the particulate filter has two important effects. The first is that the catalyst acts as an oxidation catalyst, eliminating the need for an additional catalyst. The second is that the catalytic coating lowers the soot ignition temperature of the filter to a level achieved during normal driving conditions. In this way, the particulate filter is kept free of soot build up. This is critical to keep pressure losses low to maintain low parasitic losses on the engine. This also eliminates the need for active regeneration methods, which add substantial cost and complexity to the system. The effect of the catalytic coating for reducing emissions is shown in the following figure. The figure shows the catalyst conversion efficiency for CO, HC, and NOx as a function of temperature. The tests were conducted on the BuckHybrid002 engine running at 1500 rpm with step increases in load torque. This speed is typical of cruising conditions at constant speed.
Catalyst Conversion Efficiency
100% 95% 90% 85% 80% 75% 70% 65% 60% 55% 50% 150 10% 9% 8% 7% 6% HC_eff 5% CO_eff 4% 3% Nox_eff 2% 1% 0% 350 400

EMISSIONS CONTROL INTRODUCTION The exceptional efficiency of a modern Diesel engine is countered by a very difficult emissions problem. Managing diesel NOx emissions and particulate matter is a significant challenge. In addition, the low temperatures encountered in diesel exhaust pose a challenge for maintaining high enough catalyst temperatures to promote catalysis. Technologies such as NOx adsorbers and urea-SCR can be effective at controlling NOx; however they require complicated control strategies and hardware to implement successfully. These technologies were investigated, but it was deemed that a fully functional and reliable system could not be produced for the first years competition. These technologies will be revisited for use on the 2003 OSU FutureTruck. The following subsections outline the emissions control system of the 2002 OSU FutureTruck. COOLED EGR Cooled exhaust gas recirculation (EGR) is the first step in reducing the NOx emissions in this engine. The engine is factory equipped and calibrated for the cooled EGR. Adding exhaust gas to the intake air has the effect of lowering the combustion temperature as exhaust gas has a higher heat capacity. Since NOx formation is largely a thermal phenomenon, a reduction in combustion temperature can reduce NOx formation. The use of cooled EGR is of benefit because it has a lower temperature and a higher density, this allows for more exhaust gas to be added to the intake air. WATER INJECTION Water injection is used to reduce NOx emissions from the engine. Water is injected into the intake air which in turn is carried into the cylinder. The water in the cylinder increases the heat capacity, which lowers peak temperatures and thus reduces NOx

200

250 300 Temperature

Figure 5: Catalyst Conversion Efficiency

NOx Efficiency

HC and CO Efficiency

From the chart, it is clear that the catalyst is very effective at oxidizing HC and CO. Due to the high platinum loading of the catalyst, the 50% light off temperature of the catalyst is very low. So low that it was not possible to run the engine at that low of an exhaust temperature. Test results indicate that the catalyst temperature will be high enough to ensure >97% CO conversion and >86% HC conversion efficiency several seconds after the engine is started. There is also a slight NOx reduction effect of the catalyst. This is known as a De-NOx reaction. It is only effective at low temperatures, and involves the selective reduction of NOx molecules using the available hydrocarbons in the exhaust as a reducing agent. At high temperatures, this reaction does not occur because the hydrocarbons are oxidized too rapidly on the platinum catalyst sites. PASSIVE THERMAL MANAGEMENT As diesel engines have relatively cool exhausts, retaining the available heat is critical. To accomplish this, insulating materials were applied wherever possible to the exhaust system up to, and including the catalytic converter. This allowed the catalytic converter to operate at as high of a temperature as possible, warm up faster, and thus operate at high conversion efficiencies. This also aids in allowing the particulate filter to be passively regenerated. TRANSMISSION Several factors were weighed in the process of making the decision of whether to use an automatic or a manual transmission. A manual transmission offers higher mechanical efficiency and is lighter, but most consumers prefer vehicles with automatic transmissions. Automatic transmissions also make it possible to optimize the vehicle, since the shifting will be out of the hands of the consumer; although controlling the transmission so that the operation is smooth is a difficult goal to achieve. As fuel efficiency played a more dominant role in the decision process this year, a 5-speed manual transmission was selected, with the clear intent to replace it or upgrade it to an automated manual transmission in year two. Specifically, the M5ODR4 transmission was chosen for its ability to mount in the stock location and attach to the stock transfer case. In addition, the control strategy activates up-shift and downshift lights on the dash to help optimize fuel economy.

power/weight ratio is not best-in-class, it has the advantage of being already in house, very rugged and liquid-cooled (a significant advantage for the hot climate of the FutureTruck competition), as well as CANcompatible for ease of control integration. The Ecostar (Ballard) electric motor boasts a 55 kW peak power rating and a peak torque of 200 Nm. The motor can run and produce usable power over a wide range of speeds, with its top speed at 12000 rpm.

Figure 6: Ecostar Electric Motor

Figure 7: Electric Motor Coupling ELECTRIC MOTOR COUPLING A synchronous belt and pulleys couple the driveline and the electric motor with the desired ratio of approximately 2.9:1. Figure 8 depicts the motor power vs. speed curve which results from this ratio. Motor power initially rises with speed, then, after reaching its peak, falls with further increases in speed. Larger ratios provide more low-end power, while smaller ratios give better high-end performance. The 2.9:1 ratio represents a balance between these two effects, giving good performance over the entire operating range.

POWERTRAIN ELECTRICAL
ELECTRIC MOTOR After considering multiple candidates for characteristics, availability and price, it was decided to use an electric machine used in a previous design, an Ecostar (Ballard) AC induction machine, see Figure 6. While its

Motor power vs vehicle speed 80:28 coupling 80

70

60

drive shaft connects the output of the pulley box to the rear differential. Here, the comparatively large excursions of the rear axle clearly suggested the use of the drive shaft with U-joints on each end. Sliding, splined driveshafts accommodate any axial motion. The front driveline is unchanged, so that the full functionality of the transfer case is preserved. Because the rear driveline is always engaged, the engine can supply additional power to the electric machine for regeneration of electrical power anytime the engine is motoring the vehicle. There is also the added benefit of continuously supplying power to the wheels, thereby allowing for smoother transmission shifting. One drawback of placing the engine and the motor on opposite sides of the transmission is that the vehicle must be moving in order to generate any electricity. Thus one could not use the engine and charge the batteries while at rest. Such a powertrain layout has been dubbed a double shaft parallel hybrid. BATTERIES Several batteries types were considered for use in the 2002 Ohio State FutureTruck. Advanced batteries, such as Nickel-cadmium and Nickel-metal-hydride, are very appealing because of their weight savings over traditional lead-acid batteries. However, due to time constraints in creating suitable package and to allow for proper testing, sealed lead acid batteries were selected for the 2002 truck, with the intention of switching to a more advanced battery in year two. The advanced lead acid batteries available today have excellent power density, are very economical, and widely used in EVs. Specifically, Hawker Genesis batteries were selected; this type of battery has been used before with great success. To achieve the 140 to 395V needed to power the inverter, a single series strand of 27 batteries was used, giving a nominal voltage of 324V. Since this year the truck does not have an all-electric mode, it was not necessary to have batteries with an extended capacity. With this in mind, and the space and weight constraints of the vehicle, simple simulations were run and 16 amp-hour cells were chosen. BATTERY BOXES The battery pack is split into five separate boxes. Two boxes contain 7 batteries, two contain 5, and one contains 3, to make up 27 total batteries. Each box is constructed with a frame made of 6061 aluminum. PVC plastic is used as an insulator between the batteries and the frame. A plastic spacer separates each battery and there is an air gap at the top of each box to ensure proper ventilation. The boxes are spread throughout the underside of the vehicle; see Figure 2, to distribute the added weight as much as possible. Each battery box

Motor Power (hp)

50

40

30

20

10

10

20

30

40 50 Vehicle Speed (mph)

60

70

80

Figure 8: EM Power versus Speed The belt system uses Gates brand Poly Chain GT2 series high-tension synchronous belts. The pulley connected to the electric motor has 28 teeth, while the pulley connected to the drive shaft has 80 teeth. Poly Chain GT2 belt has a poly-urethane lining, allowing the 8mm pitch belt used in Ohio State's design to transfer the large torques produced by the motor, despite the comparatively small diameter (2.8 in) of the 30 tooth pulley. This results in easier motor mounting through the use of smaller equipment. Use of the synchronous belting provides both high efficiency operation and compliance to isolate both the driver and the hardware to any form of disturbances in the driveline-motor synchronization. Figure 7 shows a CAD model of the coupling and motor, with the front cover of the motor removed. As can be seen, the enclosure of the belt box uses thick aluminum plate (the entire assembly is enclosed in thick aluminum.) Meshing of the synchronous teeth can lead to substantial NVH problems particularly at the high surface speeds seen during highway cruising. The very thick walls of the belt box result in a rigid structure whose natural frequencies should not be excited by the meshing frequency of the belt teeth, even during high speed operation. The motor and pulley box are connected to the truck body. To compensate for any motion of the transfer case output or the motor mounting relative the frame of the truck and thus to one another, Ohio State has opted to use a small driveshaft, with universal joints on each end. Use of a flexible coupling was investigated; however, the large torques and high speeds that the coupling must support limited the selection to couplings such as disk couplings that could only support a minimum amount of misalignment. In order to ensure reliability, the Ohio State FutureTruck team decided upon the use of the U-joints and a sliding yoke. Another

also contains its own temperature transducer to monitor for overheating. Exhaust fans are mounted on top of each box to scavenge both heat and gases (if any), and are powered through a DC/DC converter and are energized whenever the high voltage is energized.

bus. The TIM/motor pair is very efficient over a large range of speeds and torques and is capable of fourquadrant operation.

Figure 10: High Voltage Electric Schematic SOLAR POWER UNIT (SPU) The truck features two 32 W, Unisolar flexible solar panels that supplement the alternator in maintaining charge in the twelve-volt battery. These solar panels are extremely durable, and yield easily to mounting on the roof of a vehicle due to their flexibility. With direct, full sun, the roof-mounted solar panels are capable of supplying four amps of current to charge the battery. At the onset, four amps of current are not that much compared to the several tens of amps needed to run a vehicle. However, an analysis of energy flows and losses indicates that it is indeed significant. Consider the average chemical to mechanical conversion efficiency of the engine to be 25% and that the alternator efficiency is 50%. With these values, the solar collector offsets roughly 500 W of fuel power by lessening the load of the alternator. This translates into roughly 50 mL of fuel per hour of full sun. Continuing this example further, consider a vehicle that is used to drive to work every day. The vehicle receives the equivalent of five hours of full sun and uses four liters of fuel per day. During the day, the solar cells could provide the equivalent of 250 mL of fuel to the battery. This would directly offset parasitic fuel consumed for providing the energy via the alternator. This would reduce fuel consumption by 6%. Although this example is clearly fictitious, it demonstrates that a small amount free energy can be used to leverage a substantial fuel savings.

Figure 9: Battery Box

HIGH VOLTAGE ELECTRONICS & CIRCUITRY The high voltage electrical system consists of three main sections, connected in a chain. First are the battery packs, described above. They provide a nominal supply of 324 V. This supply is provided to the high voltage junction box, the second component in the chain. This junction box contains multiple internal switches, relays, sensors, and converters for its own use, including the Ground Fault Sensor, as well as the relay control by the Emergency Disconnect Switch (EDS), inertia switch and the Manual Isolation Switch (MIS). The current and voltage sensors are connected to the E-meter and to the CAN bus via microcontroller and ADC, to be used for state of charge estimations. The junction box serves to provide power to all of the internal sensors and relays described above, as well as to the inverter/controller of the electric motor. The electric motor and components are the third component of the high voltage system. The Traction Inverter Module (TIM) of the electric motor connects to the main leads of the junction box. The TIM controls the power sent to the electric motor and a 3phase, variable frequency AC inverter with high speed IGBT. The TIM contains isolation relays, this isolating the high voltage system from the rest of the vehicle, other than the small group of observation signals sent from the sensors to the controls system via the CAN

POWERTRAIN SUMMARY
Seen below in Table 4 is a summary of the final design decisions, and major components, as well as basic specifications.

Table 4: Component Summary COMPONENT MANUFACTURER MODEL / SPECIFICATIONS R2516F 2.5 L ,103 kW peak, CIDI TRANSMISSION MOTOR Ford Ecostar M5OD-R4 5 speed manual F8Y8-14B280AC, AC Induction, 55 kW peak F8YF-14B298AH, High Speed IGBT G16EP, PbAcid 16 Ahr, 12V Figure 11: Schematic of the hardware architecture The choice of the hardware was also important. The supervisory controller is made up of the ETAS ES-200 development board for Motorola MPC555 (32-bit microprocessor, CAN-compatible). This is an important improvement in comparison to the previous ES-1000 for two main reasons: the ETAS ES-200s small size and the very low electric power requirement. With the exception of the TIM, none of the sub-systems are CAN-compatible. Therefore, in order to control them via the CAN-bus, three CAN-translators were developed in house. The decision was made to build three different modules for packaging and wiring purposes, the small modules are located as near as possible of the control target and there is only one wire coming out of it to the supervisory controller: the CAN bus. The engine module, the driver module and the battery module convert the commands received from the supervisory controller via CAN into signals understandable by the subsystems. At the same time all the analog and digital feedbacks signals are acquired, encoded and send back to the supervisory level through the CAN bus. These modules are based on a DIP module C164CI (16-bit controller, CAN-compatible) and some signal conditioning electronics. ENERGY MANAGEMENT STRATEGY The core of the energy management strategy is based on the Equivalent Consumption Management Strategy (ECMS) technique. The aim of this strategy is a heuristic formulation of the use of the electric machine in terms of equivalent fuel consumption. This approach allows formulation of a cost function leading to a simple instantaneous optimization (of the total fuel consumption) suitable for real time control.

ENGINE

VM-Motori

INVERTER

Ecostar

BATTERY

Hawker Energy

CONTROL STRATEGY
The goal of the BuckHybrid002s control strategy is to optimize the behavior of the selected powertrain while remaining transparent to the driver. To optimize the overall powertrain, each individual component must be optimized and controlled to obtain the best behavior. Therefore, to mimic the actual optimization process, a hierarchical organization has been chosen to build the control system architecture. In this section the overall approach and hardware organization of the control system are first described and then the software implementation of the control strategy at the supervisory level are discussed. OVERALL APPROACH Ohio State has chosen to adopt a network-based control architecture. A supervisory level controller takes the globally optimized decisions and sends commands to dedicated low-level controllers for each sub-system (TIM: Traction Inverter Module, engine ECU, etc). This organization permits a separation of each of the subsystems affording flexibility and reliability. Taking into account all of these considerations the CAN (Controller Area Network) protocol and architecture were chosen. This decision was made in accordance with the design goals to reduce wiring complexity and increase flexibility and expandability, knowing that some other capabilities can (and will) be added without modifying the whole architecture.

m f _ total = m f _ ice + m f _ em

Furthermore, using a diesel engine will lead to better fuel economy results, but on the other hand some emissions characteristics (especially NOx) may be degraded. To help minimize this effect, NOx emission penalties have also been included in the equivalent fuel consumption. The minimization of this cost function decides the optimal torque split between the ICE and EM. This combination yields the following cost function:

There are several ways to shape the penalty function for the SOC. After optimization, the following normalized penalty function was selected:

SOC norm =

SOC SOC low SOC high SOC low

J = (1 NOx ) *
Where:

m f _ total m f _ total _ max

+ NOx *

m NOx _ total m NOx _ total _ max

m NOx _ total is the total (or equivalent) mass flow rate of

NOx emission due to the operation of both ICE and EM.

m NOx _ total = m NOx _ ice + m NOx _ em m NOx _ ice is estimated directly from the NOx emission
map of the engine.

m NOx _ em is the equivalent NOx emission due to EM

operation (at present) that must be accounted by the extra fuel usage due to the ICE operation cost (in the future), which is required to sustain the battery SOC. NOx is a weighting factor (0NOx1) whose value depends on how much importance (or penalty) one wants to give to NOx emissions. In order to implement a charge-sustaining control strategy, battery state of charge (SOC) must at some point be considered before applying the torque request to the ICE and EM. Previously, this has been done by calculating the optimal torque split based on fuel consumption and then shifting the optimal torque demand based on SOC. This year, however, a more ideal method will be implemented. Battery SOC will be taken into account during the optimization process, as a factor in deciding the optimal torque. This is done by including SOC in the equivalent fuel consumption as a multiplicative penalty function on the EM equivalent fuel consumption as shown below:

Figure 12: Penalty Function for SOC Based on this SOC penalty function, two examples of the optimal electric machine torque can be seen below in Figures 13 and 14. From the figures, the direct influence of SOC on the optimal torque split can be seen. This methodology lends itself to performing offline optimization and having optimal maps. Functions of torque requests, speed, gear selected and state of charge are programmed in the supervisory controller for fast, real-time implementation.

m f _ em = SFC * Pem * EC * penalty _ SOC

Where:

SFC is the specific fuel consumption

Figure 13: Optimal EM torque; SOC=0.6; 1st gear

Pem is the power of the electric machine, and

EC

is the efficiency of the electric path.

OPERATING MODE SELECTION Through a switch on the Driver Information System (DIS), the driver may select one of these three operating modes: 1. Normal mode (Hybrid) which is the default mode, 2. Conventional mode (ICE only), 3. Limp home mode (electric machine only). Note: The Limp home mode is an emergency operating mode that the driver must select only when the ICE fails. In this mode, some accessories such as the power steering and vacuum assisted brakes will not be available. VEHICLE MODELING Performances of the BuckHybrid002 were evaluated using VP-SIM, a vehicle simulator extensively developed at Ohio State University over the last few years. VP-SIM is a Matlab/Simulink, modular, scalable vehicle simulator. It is based on an original unified representation of energy and power flow in all types of vehicles, including conventional, hybrid electric and fuel cell vehicles. A model of the BuckHybrid002 is built using a scalable component library and then energy/power flows in the vehicle powertrain during many types of driving cycles (including actual measured ones) and acceleration tests are simulated. Dynamic performance (acceleration, top speed, towing capability, gradeability) and energetics (urban and highway fuel efficiency) are evaluated. Moreover VP-SIM is a forward simulator (with respect to physical causality); hence control algorithms developed and validated in simulation are directly usable for real-time control of the vehicle. Figures 15 through 17 present some typical simulation results for a Federal Urban Driving Schedule (FUDS).

Figure 14: Optimal EM torque; SOC=0.75; 1st gear STATE OF CHARGE ESTIMATION In order to use the SOC penalty as explained above, the energy management strategy must estimate the battery SOC. The OSU BuckHybrid002 uses the current integration technique to estimate the battery SOC. Basically, the idea is to keep track of the energy flow (in and out of the battery), from the beginning of the trip. Therefore, at each instant of time t, the battery SOC can be estimated using the following formula:

SOC (t ) = SOC 0
Where:

batt
Qnom

=t0

I ( )d

SOC 0 is the initial SOC (= SOC(t0) )

batt is

the battery pack efficiency (while charging or

discharging).

Qnom is the nominal charge of the battery (in Coulomb).

I is the current (in A) out of the battery (positive if discharging, negative if charging).
The initial value is estimated measuring the open circuit voltage (OCV) and using an OCV-SOC chart established by experiments and this value is automatically reinitialized every time the vehicle is stopped for several minutes.

Figure 15: ICE Operating Points; FUDS driving cycle

Figure 16: EM Operating Points; FUDS driving cycle

Figure 19: EM Operating Points; FHDS driving cycle

Figure 17: Battery SOC; FUDS driving cycle The previous graphs show the operating points for both internal combustion engine and electric machine. A second typical driving cycle is the Federal Highway Driving Schedule (FHDS); the simulation gives the following operating points for the internal combustion engine and the electric motor.

Figure 20: Battery SOC; FHDS driving cycle The simulation results are summarized below in Table 5. FUDS and FHDS data are presented as well as real road tests which are based on on-road data collected in 2000. TABLE 5: Fuel Economy Simulation Results Cycle FUDS Real Road Test (city) 24.3 FHDS Real Road Test (highway) 28.5

Fuel consumption (mpg)

22.8

32.3

TELEMATICS
The BuckHybrid002 features an advanced telematics system integrating hardware provided by Cisco, as well as some additional custom hardware. The telematics

Figure 18: ICE Operating Points; FHDS driving cycle

system focuses on several features: a Human Machine Interface (HMI) function allowing intelligent interaction between the driver and the vehicle's hybrid powertrain (extensive vehicle/powertrain monitoring and operational mode selection), a data logging function (internal and external to the vehicle), wireless communication and Internet connectivity, and enhanced consumer features (GPS navigation, PDA, digital multimedia, etc). To seamlessly perform all these functions, the telematics system interfaces simultaneously with the vehicle CAN network, the in-vehicle Ethernet, and a driver-visible display as well as a wireless Ethernet connection. A schematic of the telematics system architecture is given in Figure 21 below.

from the core telematics computer and informs the vehicle control strategy of the selection via appropriate periodic CAN messages. The core telematics computer drives the telematics system and provides most of the system's driver-visible functionality. This computer is a custom PC-104 style 400MHz AMD K-6 system running Linux and featuring a 32MB flash memory device and a 20GB hard disk with a journaling file system. Linux makes the computer robust against operating system crashes; the journaling file system makes the computer robust against power failure data corruption. This computer interacts with the driver via an 8" LCD display and can display detailed vehicle information, select the hybrid powertrain's operating mode, help the driver navigate via GPS, display Internet web sites, play digital audio, and provide Personal Data Assistant (PDA) applications. Efforts will be made to optimize the graphical interface so as to minimize driver distraction and maximize safety. An 802.11 wireless Ethernet broadband uplink to the global Internet allows a wide variety of Internet applications to be delivered to the vehicle, potentially greatly improving the consumer experience. Movie tickets could be purchased online while in transit to the theater. The driver could locate gas stations and other businesses in unfamiliar cities via the Internet. The uplink also allows the transmission of vehicle telemetry data outside the vehicle (in this case a competition server).

802.11 Receiver

Router
Land Based Ethernet

802.11 Xmitter

Touch screen LCD

PC-104 P (Pentium with disk on chip, Linux OS, 20 GB HD)

HUB

CISCO Gryphon Box

GPS

SUSPENSION
Cell Phone/ Modem

Figure 21: Schematic Representation of Telematics Hardware Architecture As shown in Figure 21, the physical center of the telematics system is the Darkstar mobile router. The router connects the other telematics components via a TCP/IP-over-Ethernet network; all communication between telematics devices passes through this network. Because a CAN bus connects the vehicle powertrain components, the telematics system uses the Gryphon device as a TCP/IP-to-CAN gateway. The Gryphon runs custom software which observes the vehicle powertrain data constantly exchanged over the CAN bus and forwards relevant information to the core telematics computer over the telematics network. This software also receives operational mode selection commands

For safety and consumer acceptability reasons, it is important to maintain similar suspension characteristics to the stock vehicle. Due to the increased load of battery and electric motor weight, the stock springs will no longer suffice. Not only would this load decrease ride height, but because of the larger initial compression of the springs, the spring rate and ride quality will also be affected. To compensate for this, new springs which maintain the stock dimensions have been procured. The custom wound springs maintain the stock load-deflection curve, with a larger initial spring stiffness to compensate for the additional weight. This allows for the stock ride quality to be completely maintained.

INTERIOR
Lear Corporation, a tier-one supplier of automotive interiors, is working with this years team to provide an advanced interior. The focus is on weight savings and reducing the environmental impact of all materials used, while maintaining or improving the safety, aesthetics and creature comforts.

SEATS Due to the short time frame, the stock seat frame and track are maintained, so as to assure meeting the FMVSS standards. However, the stock seat foam has been replaced with new seat foam technology. With this new technology, the foam densities are optimized to reduce weight without effecting the seat performance. The stock seat foam is manufactured by using a molded polyurethane process. This new technology uses slab foam. This slab foam is produced by a VPF (Variable Pressure Foaming) process. It has both physical and environmental advantages. This continuous platform cutting process has been utilized to create shape and contour in the multi layered / multi density foam. The foam is a unique combination of three processing techniques. By utilizing this new technology, the weight of the foam pad is reduced by approximately 15%. The total weight savings due to this new foam is 7.5 pounds.

CARPET AND DASH INSULATOR The stock carpeting and dash insulation have also been replaced by lighter weight versions. Both of these parts are important to passenger comfort, as they are significant sound dampers. The carpet system and dash insulator for the OSU Future Truck are made from a new acoustical material from Lear Corporation called SonoTec EF. It is a lightweight alternative to the use mass barriers in carpet backings and dash insulators that have traditionally been used to block sound transmission into the passenger cabin. The material used is a dual density, thermoformable fiber construction that is used in place of the carpet underpad in a floor system and in place of the decoupler in the dash insulator. The material is molded to the exact shape of the floor sheet metal and the dash sheet metal, assuring a perfect fit and highest levels of acoustical performance. The dual density construction provides improved sound absorption for the carpet and dash insulator, which represent just under 50% of the total interior sound absorption in the vehicle. Benefits of this material include significant weight savings and equal or better acoustical performance across all frequencies at all truck operating conditions. In this particular application for the Explorer, the weight savings represent 16.9 pounds in the original carpet and 5.2 pounds in the original dash insulator. SUMMARY Below is summarized the total weight savings yielded by the new interior components. TABLE 6: Summary of Interior Components Approximate Weight Savings PART Seat foam Figure 22: Rendering of New Seats Dash insulator 5.2 16.9 29.6 WEIGHT SAVINGS (LBS) 7.5

The seats have also been reupholstered. The seat material is compostable and recyclable. Shown above in Figure 22 is a rendering of the seat design. The outer material is TPU, made by Ultraleather HP. TPU is a leather alternative. It is more durable than regular vinyl, PVC free and the process used to make TPU is friendlier to the environment than vinyl. Plus, TPU is recyclable, unlike vinyl. Also, it looks and feels as appealing as the stock leather. The inner material is a wool-ramie blend that is completely compostable. This material is designed specially to be a renewable source.

Carpet TOTAL

DFMEA AND SAFETY

One of the primary objectives in the design of the BuckHybrid002 was to ensure its durability, safety and construction quality. All design objectives were analyzed to ensure safety and reliability.

MECHANICAL SYSTEM In addition to functionality and packaging constraints, one of the primary considerations in the mechanical system design of the BuckHybrid002 was safety. The crash worthiness of the vehicle was considered from the standpoint of weight addition, component locations (impact on CG), crush zones, vehicle stability and modifications to existing parts. To counteract the mass increase (roughly 10% of GVW, primarily due to the addition of the batteries), the 27 batteries were split into 5 separate pack distributed in between the frame rails and the two axles. This overall packaging (see Figure 2) of all the heavy components (battery packs, electric motor, inverter, EM coupling, fuel tank) led to a fairly even weight distribution (longitudinal and lateral), and a net lowering of the vehicle center of gravity (CG). This factor, the stiffening of the frame rail chassis by multiple added cross-beams, combined with the stiffening of the suspension springs to overcome the weight addition and maintain ride height close to stock, led to an enhanced vehicle lateral stability. To avoid a failure mode that would result in an unsafe condition, the battery boxes, electric motor mount and coupling, and other critical components were analyzed for excessive stress. No component was fastened to the truck without the appropriate grade fastener or fastenerlocking device. All welds to the vehicle were carefully examined both during initial assembly and after approximately 1500 miles of travel on actual roads during the Tour de Sol 2002 and local testing. Finally, the absence of any high voltage component in the engine compartment and frontal crash zone, as well as within 18 from the back bumper (rear crush zone), or outside of the chassis frame rails ensures that major electrical hazard due to a crash would not significantly affect the electrical integrity of the high voltage system, except in the event of extreme severity crashes where the integrity of the frame rails (and hence of the passenger compartment) is severely compromised. The use of automotive production CIDI engine, traction electric motor and inverter (liquid cooled) and transmission/transfer case minimizes the failure modes to be expected in this prototype. Only two mechanical assemblies were custom made (electric motor coupling and battery boxes) and hence are more prone to failures, as any prototype regardless how carefully designed and realized. To ensure the durability of the driveline, all rotating shafts used high quality bearings and were machined with splines and heat-treated. The driveshafts were custom-manufactured professionally outside using industry standard practice. All other mechanical components are OEM or aftermarket parts, hence subject to the reliability and failure mode of quality automotive parts used for in the consumer market.

ELECTRICAL SYSTEM The high voltage bus on the BuckHybrid002 includes many safety features. An emergency disconnect system allows all high voltage to be terminated at the closest point possible to the battery pack. Proper fuses are used in the battery pack to ensure it will not be overloaded. Temperature monitoring in all 5 battery packs in performed at all time when the high voltage is energized. Furthermore, a separate battery temperature sensor is required by the battery charger with proper interlock requiring a valid temperature sensor and measurement to allow any charging. Ground fault protection, manual isolation switch and emergency (electrical) disconnect, and inertial switch on both terminals of the battery pack ensure high voltage protection under proper vehicle operation, vehicle maintenance, crash, or electrical ground leak. Battery pack ventilation (individual fans on each pack) are powered through a DC/DC converter as soon as the packs are energized, hence ensuring scavenging of heat and gases at all times the high voltage pack is energized (including charging). CONTROL SYSTEMS The supervisory vehicle controller has many levels of operation that are intended to be highly fault-tolerant. In the event that a control fails for either software or hardware reasons, control will be diverted to another system. For example, if the supervisory controller has a software error, the low level control modules will automatically perform the necessary control tasks to maintain vehicle operation. The control architecture is such that the driver can manually control all systems if the need arises. The inputs to the controllers are operated within a certain range of voltages that allow for indication of failure. Some critical signals such vehicle speed, battery current and voltage have build-in redundancy measurement paths and control code to deal with unacceptable discrepancies to perform fault detection and identification, corrective action if possible, default the vehicle to a limp-home mode, or (worse case scenario) completely disable the vehicle in safe manner. The loss of any critical signal will result in the vehicle defaulting to a limp-home mode under which the drive motor function will be disabled, and the engine will operate in a conventional mode. Furthermore, a manual by-pass switch can be activated by the user to circumvent the entire control layer, and feed the throttle command directly to the engine ECU as in a conventional automobile, while disabling all other systems. The ECU of the engine itself, being a EUROIII certified ECU, has build-in limp-home mode and OnBoard Diagnostics to comply with current regulations, hence adding an underlying safety net below control layer and analogous to current production automobile.

INTENDED MARKET
Given the current popularity of SUVs, the intended market for the BuckHybrid002 has grown considerably. No longer are SUVs desirable to customers who will utilize its off-road and towing capability extensively, but it has come to serve the role of a people mover as well. Customers expect brisk acceleration, decent fuel economy, pleasing aesthetics, and a comfortable ride. The BuckHybrid002 is intended for the consumer who wants the benefits of a conventional SUV, but is also concerned with fuel economy and/or, environmental impact. These benefits include four wheel drive, high ground clearance, exceptional safety, towing capability, and ample cargo room. The BuckHybrid002 improves upon these attributes by simultaneously offering higher fuel economy, improving greenhouse gas emissions, as well as the many other features outlined in this document.

different vehicle packages are assembled today. Naturally, the hybrid Explorer would require additional components, such as the high voltage system, supervisory controller, and additional wiring harnesses. But as the majority of the systems are identical, the base platform could be shared and manufacturing costs kept at a minimum.

COST ANALYSIS
The estimated cost of the BuckHybrid002 to produce on a large scale is $37,000. For this estimate, the diesel engine and manual transmission were assumed to have nearly equivalent value as the stock items that they replaced. With the exception of the high voltage system and electric motor, the vehicle is simply a conventional diesel vehicle with a production diesel engine powertrain. This should keep assembly costs to a level comparable with the stock vehicle. One cost that is difficult to quantify in such an estimate is the added engineering expense necessary to develop such a vehicle. Development and testing of the hybrid powertrain and control strategy would be a significant cost that is not present in a conventional vehicle. This is especially challenging given the novelty of production hybrid electric vehicles. Without a prior base of engineering knowledge to work from, research and development costs could be significant. These costs are not included in the estimate. The following table illustrates the added cost associated with the major components. Table 7: Major Component Cost

MANUFACTURABILITY
Where possible, all components were selected from available stock components. By minimizing the number of custom-made components, one ensures the manufacturability of the vehicle. All of the major driveline components are production items, with the exception of the electric motor belt coupling. In a production vehicle, the belt drive would no doubt be replaced with a gear assembly of some sort. As such assemblies are very common, it poses no manufacturability problem. One problem faced when integrating a new powertrain into an existing platform is packaging. Many of the design decisions were dictated by the existing platforms geometry. If the vehicle was being designed from the ground up, consolidating components as well as making the most of the available space could realize many improvements in manufacturability. With the exception of the high voltage system and the electric motor, the BuckHybrid002 has an identical layout of the stock Explorer. The engine is situated in the conventional north-south orientation used in rear wheel drive vehicles. Thus, installing the diesel engine is an identical process to the assembly of the stock vehicle. One deviation from a stock vehicle assembly is the battery enclosures, electric motor, and inverter. All of these items would be assembled on the vehicle frame from above, prior to attaching the cabin. This poses no unnecessary difficulty as many items are installed in this manner. One can envision a single production line that could produce both a conventional Explorer, as well as the hybrid Explorer. This is much the same way that

Prototype Cost
Base Vehicle Electric System Motor $30,000 $7,000 $15,000 $2,000 $4,000 $58,000

Production Cost
$30,000 $2,500 $2,500 $1,000 $1,000 $37,000

Additional Control Hardware High Voltage System Telematics System Total Hybrid Package

CONCLUSION
A mixing of technology and innovation create the Ohio State University FutureTruck. A state-of-the-art diesel engine, an advanced AC induction electric motor, and a lead acid battery pack provide ample power to meet the needs and wants of potential Explorer drivers. The hybrid powertrain, having a combined rating of 220hp, excels either out on the open road or off the beaten path. Due to the totally transparent control strategy, there will be no difference for the consumer in operation between the BuckHybrid002 and a normal Explorer. An updated and environmentally friendly interior, along with a user friendly HMI system create a pleasant driving atmosphere, providing comfort and class. Add to this that the BuckHybrid002 still achieves a combined fuel economy of 29 mpg and meets ULEV emissions standards -- it is not your typical SUV.

On a more personal note, the authors would like to acknowledge the many students and researchers who have contributed time and knowledge to so many facets of the project, culminating in the completion of the BuckHybrid002, ready on time for the Tour de Sol 2002 and the FutureTruck 2002 competition. REFERENCES 1. Rizzoni, G., Guezennec, Y., Brahma, A., Wei, X., and Miller, T., VP-SIM: A Unified Approach to Energy and Power Flow Modeling Simulation and Analysis of Hybrid Vehicles, SAE Paper 2000-01-1565, Proc. 2000 Future Car Congress, Paper 2000-01-1565, April 2000, Arlington, Va. 2. G. Rizzoni, L. Guzzella, B. Baumann. Modeling and design optimization of hybrid vehicles. IEEE/ASME transactions on Mechatronics, September 1999. 3. Brahma, A., Guezennec, Y., Paganelli, G., Rizzoni,G., Yurkovich, S., A hardware- and architecture-independent supervisory control strategy for hybrid-electric drivetrains, Proc. 4th Stuttgart Symp., Feb 2001. 4. Paganelli, G., Guerra, T.M., Delprat, S., Santin, J.J., Delhom, M., Combes, E., Simulation and Assessment of Power Control Strategies for a Parallel Hybrid Car", J. Automobile Engineering, also in Proc. of the Institution of Mechanical Engineers IMechE, SAE International, 214, pp. 705-718, 2000. 5. M. Hopka, A. Brahma, S. Dilmi, G. Ercole, C. Hubert, S. Huseman, H. Kim, G. Paganelli, M. Tateno, The 2000 Ohio State University FutureTruck, SAE Special Publication SP-1617, 2001. 6. Rizzoni, G., Brahma, A., Ohio State University Competes in FutureTruck, ETAS Real Times pub., 1-2001 7. Paganelli, G., Ercole, G., Brahma, A., Guezennec, Y., Rizzoni, G., "A General Formulation for the Instantaneous Control of the Power Split in ChargeSustaining Hybrid Electric Vehicles", in Proc. AVEC 2000, 5th International Symposium on Advanced Vehicle Control, Ann Arbor, MI, August 2000. 8. Paganelli, G., Ercole, G., Brahma, A., Guezennec, Y., Rizzoni, G., General Supervisory Control Policy for the Energy Optimization of Charge-Sustaining Hybrid Electric Vehicles, to be published in JSAE Review, Oct. 2001. 9. Paganelli, G., Tateno, M., Brahma, A., Rizzoni, G., Guezennec, Y., Control Development for a hybridelectric sport-utility vehicle: strategy, implementation and field test results, to be published in Proc. ACC 2001

ACKNOWLEDGMENTS
The Ohio State BuckHybrid002 team would like to acknowledge the support of many sponsors, without whom this major endeavor would never have been possible. First of all, we would like to thank the organizers of the FutureTruck competition, the US Department of Energy and Argonne National Laboratory for providing an outstanding level of support and Ford Motor Company for providing the competition vehicles, the logistical and technical support to all teams. We would also like to thank all the other official competition sponsors who are making this extraordinary event possible for all the schools and provide us access to cutting edge technology. In addition, the Ohio State team has relied heavily on sponsorship from many companies. Specifically, we would like to acknowledge VM Motori, ETAS, Inc., Ballard (formerly Ecostar), Lear Corporation, Aquamist, Degussa, Griffin Industries, and countless others would have made this project possible through their generous donations. We also want to thank our local Ford dealer, Graham Ford of Columbus, for their enthusiastic sponsorship, patience and relentless efforts to supply us with un-findable, yet existing parts. Also, a special Thank You goes to Honda Manufacturing of America, a long term supporter of the Ohio State College of Engineering, for a particularly generous and timely cash donation towards our BuckHybrid002 effort. Last, but not least, the team is very thankful to the OSU Center for Automotive Research and Intelligent Transportation for the use of its facilities, research equipment and machine shop and to the College of Engineering for being so supportive of student project activities for many years and specifically, for their financial and student support towards FutureTruck.

CONTACTS Further information about the OSU BuckHybrid002 can be found at the team web page at http://turbo.eng.ohiostate.edu/future. Information about the Ohio State University GATE program and Center for Automotive Research can be found at http://car.eng.ohio-state.edu or through Prof. or Yann Giorgio Rizzoni ([email protected]) Guezennec ([email protected]).