45
are modeled with little attention to the expected level of accuracy of the simulation both
because of rough approximations used in the analytical methods and a lack of
appreciation of the errors inherent in the validation data itself. A DOE effort to
benchmark certain system analyses and subsystem models against the best available data
would be of use to all.
because of rough approximations used in the analytical methods and a lack of
appreciation of the errors inherent in the validation data itself. A DOE effort to
benchmark certain system analyses and subsystem models against the best available data
would be of use to all.
It was pointed out that the DOE labs have extensive experience in producing basic
data for use in code validation. The national labs have done this sort of work for a number
of industries over the years (e.g., basic thermal-hydraulic data for devices such as heat
exchangers, steam generators, coal and gas combustion devices, etc.). It would be useful
to perform a similar role for areas of interest to the heavy-vehicle industry, e.g., fan
performance, radiator heat transfer, refrigeration systems, etc. In addition, the DOE labs
have had extensive experience in creating flexible overall system models that are not tied
to the limitation of a specific design.
of industries over the years (e.g., basic thermal-hydraulic data for devices such as heat
exchangers, steam generators, coal and gas combustion devices, etc.). It would be useful
to perform a similar role for areas of interest to the heavy-vehicle industry, e.g., fan
performance, radiator heat transfer, refrigeration systems, etc. In addition, the DOE labs
have had extensive experience in creating flexible overall system models that are not tied
to the limitation of a specific design.
There was extensive discussion of the unique nature of the heavy-vehicle industry
and how little control the OEMs often have over the specification of subsystems in their
products. Using the requirements specified by a large fleet operator as an example, it was
noted that the manufacturer of a Class 7-8 truck may be told what engines to use, what
cooling systems are to be installed, etc. Thus, the OEM has much less control over
optimization of the total system than, say, an auto manufacturer would have. Also,
because individual suppliers are often producing packaged "black-box" subsystems for
integration into a final product, a buy-in would be needed from each supplier to have
them transfer needed system modeling data to some central location (in some standard
format) before any true integrated system model would be possible.
products. Using the requirements specified by a large fleet operator as an example, it was
noted that the manufacturer of a Class 7-8 truck may be told what engines to use, what
cooling systems are to be installed, etc. Thus, the OEM has much less control over
optimization of the total system than, say, an auto manufacturer would have. Also,
because individual suppliers are often producing packaged "black-box" subsystems for
integration into a final product, a buy-in would be needed from each supplier to have
them transfer needed system modeling data to some central location (in some standard
format) before any true integrated system model would be possible.
It was noted that whatever DOE might choose to develop, the real test will be its
use by industry. That will be the only test of value. It is DOE's role to try to understand
the overall economic needs of the heavy-vehicle industry and to create a program that is
useful to that group while, at the same time, increasing fuel efficiency and reducing
harmful emissions from the next generation of truck designs those being DOE's
primary goals.
the overall economic needs of the heavy-vehicle industry and to create a program that is
useful to that group while, at the same time, increasing fuel efficiency and reducing
harmful emissions from the next generation of truck designs those being DOE's
primary goals.
In many of the discussions, it was stated that all of the subsystems of a Class 7-8
truck are interconnected, and this drives the need for a systems code to model such
interactions. While this is true to a degree, it is essential to realize that not all parts of the
system are closely coupled (e.g., the cab interior air system is only weakly connected to
other systems) and, in deciding modeling priorities, it would be valuable to concentrate
on closely coupled systems first (e.g., the relationship between underhood cooling
simulation and fan modeling).
interactions. While this is true to a degree, it is essential to realize that not all parts of the
system are closely coupled (e.g., the cab interior air system is only weakly connected to
other systems) and, in deciding modeling priorities, it would be valuable to concentrate
on closely coupled systems first (e.g., the relationship between underhood cooling
simulation and fan modeling).
There were questions of what would drive OEM interest in improved analysis and
modeling capability. The best answer seems to be through the possibility of reducing the
testing of vehicles. The creation of "virtual vehicles" would allow one to test perhaps
testing of vehicles. The creation of "virtual vehicles" would allow one to test perhaps