TransForum Vol. 3, No. 3

DEBUT OF HYBRID ELECTRIC VEHICLES HEATS UP INTEREST IN UNDERHOOD THERMAL MANAGEMENT

Amidst a great deal of fanfare last year, the Toyota Prius joined the Honda Insight as the onlyhybrid electric vehicles (HEVs) mass-produced for the North American market. U.S. manufacturers have also unveiled prototype HEVs that should hit the market within the next few years.

These first-generation HEVs are the most technologically advanced vehicles ever made, combining a gasoline-powered engine with a battery and electric motor. They bring together the "best of both worlds": the extended driving range and easy refueling of the conventional automobile and the fuel economy and environmental benefits of an electric vehicle. They're also a harbinger of things to come. Automotive analysts now predict that carmakers will roll out as many as 30 more HEVs by 2010.

The technological complexity of these HEVs is of interest at Argonne, where researchers in thermal management have been working with industry partners on high-performance computer models that simulate complex fluid flows and heat transfer in the underhood systems of cars, including HEVs. The goal of this pioneering work is to improve underhood heat-load modeling, especially as it relates to emissions and fuel efficiency.

"Aerodynamics, packaging, and styling changes that reduce the size of the underhood compartment and front openings can have an adverse impact on the underhood thermal environments," says Adrian Tentner, director of Argonne's Intelligent Transportation Systems program. "The result can be higher emissions, reduced fuel efficiency, and damage to sensitive electronic components."

Automakers have been addressing these issues in conventional cars by using system analysis codes, primarily the PSAT and ADVISOR codes developed with U.S. Department of Energy (DOE) funding, and modern computational fluid dynamics (CFD) computer codes, such as the STAR-CD codes by Analysis and Design Application Company, of Melville, New York. The CFD codes are being used in tandem with experimental data to improve the performance of individual underhood components.

Despite the significant advances in underhood thermal management, hybrid cars remain largely uncharted territory. The addition of significant new thermally active components, such as the power electronics system, creates challenges for both system analysis codes and the component design CFD codes. "We need to better understand what happens under the hood of the hybrid as well as other, more advanced vehicle designs," notes Tentner. "There is little underhood experimental data about how these vehicles behave, so you cannot calibrate the thermal management systems code based on experience. What are needed are more detailed CFD codes that get this information for the systems analysis codes to help users get realistic, reliable results."

Under Tentner's leadership, Argonne recently completed a proof-of-concept project in which researchers investigated the merits of using high-fidelity CFD codes, along with the industry-accepted system analysis codes, to analyze underhood thermal phenomena in hybrid cars. At the end of the six-month effort, using the STAR-CD CFD code, researchers found that "yes, computational fluid dynamics can be used to model the geometry of hybrid engines and heat transfer in the underhood space, and provide the parameters needed to enhance the system analysis codes, resulting in improved evaluations of emissions and fuel efficiency."

Argonne researchers are planning to follow up their preliminary findings by extending the STAR-CD modeling capabilities to include hybrid-specific components and thermal underhood loads.

Interfacing the new modeling capabilities with the system codes would allow a more accurate analysis of integrated hybrid-vehicle propulsion designs.

Tentner is especially eager to take advantage of the experimental capabilities of Argonnne's Advanced Powertrain Research Facility; these capabilities are essential tools for validating various computations, an area in which Argonne researchers have traditionally excelled.

According to Tentner, "If we just run the computations, we can never be sure they are correct. We need to run a set of experiments on the vehicle and analyze the data. If we can obtain correct computational results, we will be able to replace future experiments with computer simulations of various situations. That will save time and money. With relatively little additional effort, the models can be extended to the analysis of other vehicles."