Refine Your Search

Search Results

Viewing 1 to 3 of 3
Journal Article

A Machine Learning-Genetic Algorithm (ML-GA) Approach for Rapid Optimization Using High-Performance Computing

2018-04-03
2018-01-0190
A Machine Learning-Genetic Algorithm (ML-GA) approach was developed to virtually discover optimum designs using training data generated from multi-dimensional simulations. Machine learning (ML) presents a pathway to transform complex physical processes that occur in a combustion engine into compact informational processes. In the present work, a total of over 2000 sector-mesh computational fluid dynamics (CFD) simulations of a heavy-duty engine were performed. These were run concurrently on a supercomputer to reduce overall turnaround time. The engine being optimized was run on a low-octane (RON70) gasoline fuel under partially premixed compression ignition (PPCI) mode. A total of nine input parameters were varied, and the CFD simulation cases were generated by randomly sampling points from this nine-dimensional input space. These input parameters included fuel injection strategy, injector design, and various in-cylinder flow and thermodynamic conditions at intake valve closure (IVC).
Journal Article

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

2019-01-15
2019-01-0001
A computational fluid dynamics (CFD) guided combustion system optimization was conducted for a heavy-duty diesel engine running with a gasoline fuel that has a research octane number (RON) of 80. The goal was to optimize the gasoline compression ignition (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 a 1-1.5 g/kW-hr window. The numerical model was developed using the multi-dimensional CFD software CONVERGE. A two-stage design of experiments (DoE) approach was employed with the first stage focusing on the piston bowl shape optimization and the second addressing refinement of the combustion recipe. For optimizing the piston bowl geometry, a software tool, CAESES, was utilized to automatically perturb key bowl design parameters. This led to the generation of 256 combustion chamber designs evaluated at several engine operating conditions.
Technical Paper

Combustion System Optimization of a Light-Duty GCI Engine Using CFD and Machine Learning

2020-04-14
2020-01-1313
In this study, the combustion system of a light-duty compression ignition engine running on a market gasoline fuel with Research Octane Number (RON) of 91 was optimized using computational fluid dynamics (CFD) and Machine Learning (ML). The focus of this study was to optimize the piston bowl geometry at two compression ratios (CR) (17 and 18:1) and this exercise was carried out at full-load conditions (22bar indicated mean effective pressure (IMEP). CAESES, a commercial software tool, was used to automatically perturb key bowl design parameters and CONVERGE software was utilized to perform all CFD simulations. 128 piston bowl designs were evaluated at each compression ratio. Subsequently, a Machine Learning-Grid Gradient Algorithm (ML-GGA) approach was developed to further optimize the piston bowl design. This extensive optimization exercise yielded significant improvements in the engine performance and emissions compared to the baseline piston bowl designs.
X