Review of the Research Program of the Partnership for a New Generation of Vehicles: Third Report (1997)

Systems analysis, along with vehicle, subsystem, and component modeling, is believed to be absolutely necessary to ensure that the best technologies are selected and that overall vehicle performance is optimized. The PNGV Technical Roadmap emphasizes this point in Section III-A and clearly states (PNGV, 1996):

A vehicle systems model can be very powerful when comparing the relative performances of selected vehicle configurations. A good example is the optimization of hybrid electric vehicle (HEV) performance. The hybrid vehicle has two very significant advantages that contribute to fuel economy: (1) the possibility of recovering some fraction of the braking energy and (2) the ability to run the selected power plant in a restricted, more efficient load and speed range. Conversely, the disadvantage of the hybrid vehicle is that it is a much more complex and costly system than a conventional vehicle with an internal combustion engine. These advantages and disadvantages must be considered in any trade-off analysis.

Systems analysis provides the tools to determine how fuel economy, cost, and emissions can be optimized within the overall vehicle performance requirements. This information can then be translated into performance requirements for

energy storage. System analysis, along with effective modeling tools, is the only way to ensure optimization of vehicle performance. Systems analysis also provides the opportunity to study trade-offs and make appropriate design compromises, and it leads to the specification of necessary characteristics for the interacting energy conversion, energy storage, and control technologies embodied in a vehicle. It is also extremely important for setting research, development, and engineering targets.

In its second report, the committee made the following recommendations (NRC, 1996):

• The PNGV should assess the impact on the overall program schedule of the delay in implementing systems analysis and vehicle engineering tasks and the need for remedial action. Priority projects must be identified and implemented by the technical teams as soon as possible.

• The PNGV should formalize the subsystem evaluation and selection process without delay, and performance criteria should be provided to the PNGV technology teams. The systems analysis must be an interactive process that continually receives new information, updates models, and provides updated results from optimizations and tradeoff studies to system, subsystem, and component designers.

• Overall vehicle system and subsystem analysis driving component developments should be under the control of USCAR technical director.

The committee voiced a strong concern that the systems analysis effort had been significantly delayed 12 to 18 months, primarily because of a lack of funding. The committee believed this would jeopardize the downselect process scheduled by the end of 1997.

The PNGV systems analysis team, led by the USCAR, finalized a contract with TASC and Southwest Research Institute (SWRI) in January 1996, and the effort outlined in the PNGV Technical Roadmap has been aggressively pursued during the past year. The statement of work specified initial system studies to identify, quantify, and rank a selected set of alternative vehicle configurations. This requires developing analytic methods and tools for comprehensive analysis of the identified new vehicle technologies and performing trade-off studies to select the final preferred vehicle configuration. The systems analysis team, along with the vehicle engineering team, identified nine vehicle configurations for benchmarking (see Table 3-1).

A rudimentary vehicle model has been created, and the team is currently developing models for the many subsystems and components that are being evaluated. These models are key elements in assuring that the 1997 technology downselect process can be accomplished with high confidence in the accuracy of

predicted technology performance. The current modeling status varies, from highly accurate models for technologies like internal combustion engines that are well known to the automotive industry, to very generic, simplistic models for technologies that are not well developed, such as fuel cells. Some subsystems being evaluated clearly will not be ready for inclusion in the concept vehicle architecture.

All models must be extensively validated, and this is being accomplished by taking known performance characteristics from several sources. The automotive companies and SWRI have provided this validation, where possible. The process is complicated by proprietary vehicle simulation capability that includes the effects of cold starts, environmental conditions, crashworthiness, vehicle duty cycle, emissions, and fuel economy. Scaling and sizing of the subsystems and components still represent a significant challenge, and this must be accomplished for each vehicle configuration. Cost and reliability models are very weak. Control strategies are only starting to be developed, with the University of Michigan and Oakland University providing support.

Only very gross model benchmarking for the conventional and series hybrid-vehicle configurations has been accomplished to date. One of the major issues is how to achieve effective interaction with the other PNGV technical teams, especially the vehicle engineering team. Vehicle requirements are the responsibility of the vehicle engineering team. These requirements must be constantly matured and refined. The first significant workshop involving the technical teams was held in September 1996. The workshop was a major milestone, and it should start the effective dialogue needed to develop representative and accurate subsystems and component models. Templates for each subsystem must be

developed in cooperation with the technical teams, and this will require considerable time and funds to achieve the accuracy necessary to ensure high-fidelity vehicle trade-off studies. It appears to the committee that effective interfaces have not been established with the subsystem teams. The tools being developed must be presented to the technical teams for a critique, which will lead to a more effective use of the models. In this case, the systems analysis team must determine users' real needs.

A significant problem regarding funding for 1997 was outlined to the committee. During 1996, $2.4 million was spent with contractors. A Phase 2 and Phase 3 program are planned for 1997 and 1998. Each phase will cost approximately $2.5 million. The 1997 funding will address the technology downselect, which is critical to meeting overall program timing. A source for this funding has not been identified. During the review of the systems analysis part of the PNGV program, the committee concluded that the Phase 1 expenditure of $1.8 million was excessive. The PNGV program managers should review the planned Phase 2 and Phase 3 proposals to ensure that the work planned is consistent with the proposed expenditures. The importance of this work requires that the PNGV strive to maximize what can be accomplished with limited budgets. An additional issue is the inadequate funding provided by the Department of Energy to the national laboratories to support systems analysis. Considerable know-how and developed technology are available at the national laboratories. These assets could be major contributors to achieving successful subsystem and vehicle models.

None of the funding issues outlined above has been addressed. Without focused management action and resource allocation, this vital systems analysis effort will be interrupted or fall further behind schedule. Achieving the original planned milestone of completing the trade-off studies by the end of 1997 to allow selection of the preferred vehicle concept appears to be a major challenge. The loss of effectively more than a year in starting the system analysis effort has resulted in very late development of the required vehicle, subsystem, and component models; thus providing little timely support or guidance to the technical teams. Requirements based on the model studies should have been communicated early in the program; however, this is only beginning now. Serious problems could result from the fact that technology teams may not be developing technology centered on the PNGV vehicle design requirements.

The committee believes that the system analysis part of the program has now been established with a credible foundation. This is based on the detailed review of the technical approach and a demonstration of the PNGV systems analysis capabilities that are under development. Timing, however, is going to be a major challenge, along with providing the necessary funding.

The committee regards the following as significant issues:

• All analytic models are very rudimentary at this time. Validity and fidelity must be established. Lack of validation data is hindering the efforts of the systems analysis team. Unfortunately, a lot of these data are considered proprietary by potential providers. Program timing is thereby being threatened.

• Cost and reliability models, which are critical to evaluating designs, are insufficient and are significantly behind schedule.

• Attention to date has been focused on creating a systems analysis tool, and the necessary models. Little effort has been made to understand how the technical teams, especially the vehicle engineering team, will use the tool in the studies necessary for the technology downselect process. Interaction with the technical teams is minimal and significantly behind schedule.

• Participation by the vehicle engineering team has been minimal. This affects the accurate establishment of vehicle requirements.

• No funding source for 1997 and 1998 has been identified.

• Anticipated support from the national laboratories has been seriously curtailed by lack of government funding. This affects the realization of validation data for fuel cells, batteries, flywheels, and ultracapacitors.

Based on its review of the current status of the PNGV program, the committee makes the following recommendations.

Recommendation. The managers of the PNGV should conduct a program review with the leadership of the vehicle engineering team and the systems analysis team to assess the capability of the existing projects to meet program needs for the technology downselect process. A corrective action plan should be formulated and implemented as a matter of some urgency. This will ensure that systems studies are designed and implemented to provide the necessary optimization and trade-off information to make the best downselect choices. It would be logical for the systems analysis team to place a priority on creating those models that appear to be sure contenders for the technology selection process in 1997.

Recommendation. The managers of the PNGV should define and obtain the necessary resources for conducting systems analyses in 1997 and 1998. As part of this process, a detailed assessment of the 1996 $1.8 million TASC/SWRI contract should be conducted. The selected contractor should provide detailed work plans and expenditures for each task along with expected results.

Recommendation. The USCAR management should make a concerted effort to overcome the barrier of proprietary rights, at least to the extent necessary to ensure validation of the systems analysis models.

NRC (National Research Council). 1996. Review of the Research of the Partnership for a New Generation of Vehicles, Second Report. Board on Energy and Environmental Systems and Transportation Research Board. Washington, DC.: National Academy Press.

PNGV (Partnership for a New Generation of Vehicles). 1996. Technical Roadmap (draft). Dearborn, Michigan: PNGV.

Viergutz, O. 1996. PNGV Systems Analysis Team Status. Presented to the Standing Committee to Review the Research Program of the PNGV at the Dearborn Inn, Michigan, September 10, 1996 .