SAE Global Supplier Marketplace
Login / MySAE  |  Sign Up!
SAE Home Industries
Search: Advanced Search
Other SAE Magazines Sites

SAE Off-Highway Engineering Online SAE Off-Highway Engineering Online Magazine eMedia Advertising Info Contact Us

Technical Innovations
ERC creating next-generation engine models


Figure 1. The cycle at 6.0 CA° after TDC.


Figure 2. The cycle at 7.0 CA° after TDC, with the fuel colored by drop size.

Almost all University of Wisconsin Engine Research Center (ERC) studies rely on models to solve equations that describe spray and combustion processes. The current models, while continuously evolving, can trace their roots to work that is more than a decade old. Researchers and graduate students at the ERC are running numerical simulations of basic fluid and combustion processes to develop and test the next generation of engine models. "We use the results to test engine model ideas," said Christopher Rutland, Associate Professor of Mechanical Engineering at UW-Madison and a principal investigator for the ERC. "We are improving the models so they can be used to address issues such as emissions and fuel economy."

One project focuses on the turbulent mixing and combustion of fuel and oxidizers. Direct numerical simulations (DNS) are used to study turbulent reacting shear-layers, which are the prototypes of many internal-combustion engine flow situations. In DNS calculations, no turbulence models are used and all relevant scales are resolved on the computational grid. This permits the study of basic physical processes that occur in turbulence-flame interactions. Emphasis is placed on investigating the production, transport, and dissipation of turbulent stresses and kinetic energy.

A problem encountered with DNS is the large number and complexity of the simulation results, with 15 to 20 million nodes, each with a dozen or more pieces of information. This is much more than the 200,000 to 1 million nodes of information common in computational fluid dynamics (CFD) calculations. The ERC uses CEI's EnSight software to analyze, visualize, and communicate these massive calculations in order to compare them to existing models.

With EnSight, researchers get a clear, animated 3-D picture of large blocks of information concerning flow, mixing, and combustion processes. The results can be displayed using a large variety of tools that help develop new insight and understanding.

Figures 1 and 2 are visualizations from a test cycle of a single-cylinder version of the Caterpillar 3400 series diesel engine. Contour colors indicate the fuel vapor that has already mixed with the air. The colors indicate the degree of mixing, with red indicating a fuel-rich mixture. The beginning of fuel injection is 0.5 crank angles before top dead center (TDC).

Researchers are still exploring different factors in their DNS studies, and the turbulent mixing and reacting shear-layer project is expected to take another six months to a year to finish. "This is an evolutionary thing," said Rutland. "We develop new understanding of the basic processes that help us develop new ideas for models. These are compared to the DNS results, but then we have to put them into the engine applications and see if they have an impact. Through this process we can take these models into the next generation."

For more information, circle 424.

SAE Off-Highway Engineering April 2000

©2008 SAE International. All rights reserved.