CFD-Guided Combustion System Optimization of a Gasoline Range Fuel in a Heavy-Duty Compression Ignition Engine Using Automatic Piston Geometry Generation and a Supercomputer 2019-01-0001
Improving the efficiency of the internal combustion engine in the commercial sector plays a dual role. Vehicle owners are continuously seeking lower total cost of ownership and the reduction of greenhouse gases contributes positively to the global climate change discussion. Simultaneously addressing criteria pollutants, however, has become critical when exploring high efficiency technologies. Gasoline Compression Ignition (GCI) has shown the potential to achieve high fuel efficiency with an improved NOx-Soot tradeoff compared to conventional diesel combustion. In this study, a computational fluid dynamics (CFD) guided combustion system optimization investigation was conducted for a Cummins ISX15 heavy-duty diesel engine running with a high reactivity gasoline fuel that has a research octane number (RON) of 80. The goal was to optimize the GCI combustion recipe (piston bowl geometry, injector spray pattern, in-cylinder swirl motion, and thermal boundary conditions) for improved fuel efficiency while maintaining engine-out NOx within 1 -1.5 g/kW-hr levels. The CFD model was developed using the multi-dimensional CFD software package, CONVERGE. A two-stage design of experiments (DoE) approach was used with the first DoE campaign focusing on the piston bowl shape optimization and the second one dealing with the further refinement of the combustion recipe. For the optimization of piston bowl geometry, a software package, CAESES, was utilized to automatically perturb various bowl design parameters, leading to 256 combustion chamber designs generated and evaluated at several key engine operating conditions. The second DOE campaign was conducted using CONVERGE CONGO, resulting in more optimized injector spray pattern, fuel injection strategy and in-cylinder swirl motion for the best performing piston bowl designs from the first DoE campaign. This comprehensive optimization study was performed on a world-leading supercomputer, Mira, to accelerate the development of an optimized fuel-efficiency focused design. Compared to the production combustion system, the combustion recipe developed from this study showed appreciably improved closed-cycle fuel efficiency across the key operating points investigated while meeting the engine-out NOx targets. Optimized piston bowl designs and injector spray patterns were predicted to provide enhanced in-cylinder air utilization and more rapid mixing-controlled combustion, thereby leading to a fuel efficiency improvement. In addition, tailoring the engine thermal boundary conditions toward leaner operation was also key to the improved fuel efficiency.
Yuanjiang Pei, Pinaki Pal, Yu Zhang, Michael Traver, David Cleary, Carsten Futterer, Mattia Brenner, Daniel Probst, Sibendu Som
Aramco Research Center - Detroit, Argonne National Laboratory, Friendship Systems, Convergent Science Inc.
International Powertrains, Fuels & Lubricants Meeting