Virtualizing combustion to cut aircraft engine emissions
Aircraft engines must meet the emission standards for nitrogen oxides (NOx) and carbon monoxide (CO) set by the U.S. Environmental Protection Agency. These standards are becoming more and more stringent, and engine companies are facing increasing challenges in designing and optimizing engines to comply. The combination of computational science and high-performance computing can provide a virtual tool to help meet these requirements.
Purdue’s Computational Energy & Propulsion Lab (CEPL) is working on developing accurate and efficient computational models for combustion processes. Ultimately, we aim to deliver important computational-modeling building blocks to the engine industry for the virtual tools it uses in design and optimization.
Aircraft pollution can cause serious health and environmental issues. Nitrogen oxides are produced when nitrogen (N2) and oxygen (O2) in air chemically react during the combustion process in aircraft engines under extremely high temperatures. CO results from incomplete combustion of carbon-containing engine fuels. These emissions are inevitable, governed by the laws of thermodynamics when hydrogen-carbon fuels are burned with air in combustion engines.
The challenge is to reduce the emissions without significantly compromising engine efficiency and performance. An ideal engine, with theoretically optimal performance, will never be built; rather, there will be continuous improvements toward this “optimal” design. Prototypes are built to narrow down specific designs in order to more closely align with the performance goal, in a time-consuming and expensive iterative process. The number of prototype engines that can be built is limited by cost — which minimizes the improvements that can be achieved.
What’s required is an efficient tool to fast-prototype and quickly identify feasible design options. That’s why virtual design tools based on computer models and algorithms are so attractive to engine companies. My lab’s work focuses on developing one building block of a virtual engine design tool: reliable and fast prediction of the engine combustion process.
The combustion process inside an engine is extremely complicated, involving thousands of chemical reactions. These reactions, coupled with the complex process of turbulence, make the combustion process extremely difficult to simulate by using computer models.
Taking a probabilistic approach to predicting combustion processes, we are accurately tracking the stochastic evolution of combustion quantities. We are building a generalized tool that can work for all scenarios and theoretically is applicable to all combustion problems. To date, we have developed sub-models in the framework for improved model consistency, accurate modeling of the molecular diffusion process, and an enhanced model for the small-scale mixing process. These models rely on computational calculations of probability to enhance model reliability and predictive capability.
We face a long journey before our upstream foundational work can flow down to commercial users like engine manufacturers. Two fundamental challenges remain: affordability and accuracy. We are collaborating with Sandia National Labs, the Argonne National Lab, and Honeywell Aerospace. We have received funding from the National Science Foundation, the American Chemical Society, the U.S. Department of Energy, and Honeywell.
Combustion engines will remain the chief energy supply for transportation, power plants, space exploration, and defense for decades to come. It is pivotal to minimize the impact of combustion pollution emissions while maintaining the energy supply that society demands.
Virtual design provides fast prototyping at low cost to continuously improve combustion engines with lower emissions. It will require years of effort by many researchers to develop all the building blocks that must be integrated into virtual design tools. We are proud to be in the vanguard of this important endeavor.
Haifeng Wang, PhD
Associate Professor
School of Aeronautics and Astronautics
College of Engineering, Purdue University
Related Links
