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

Jihad Badra, Fethi Khaled, Jaeheon Sim, Yuanjiang Pei, Yoann Viollet, Pinaki Pal, Carsten Futterer, Mattia Brenner, Sibendu Som, Aamir Farooq, Junseok Chang

Research output: Contribution to conferencePaperpeer-review

26 Scopus citations


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). This work was focused on optimizing the piston bowl geometry at two compression ratios (CR) (17 and 18:1) and this exercise was carried out at full-load conditions (20 bar indicated mean effective pressure, IMEP). First, a limited manual piston design optimization was performed for CR 17:1, where a couple of pistons were designed and tested. Thereafter, a CFD design of experiments (DoE) optimization was performed where CAESES, a commercial software tool, was used to automatically perturb key bowl design parameters and CONVERGE software was utilized to perform the CFD simulations. At each compression ratio, 128 piston bowl designs were evaluated. 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. Up to 15% savings in indicated specific fuel consumption (ISFC) were obtained. Similarly, the optimized piston bowl geometries produced significantly lower emissions compared to the baseline. Emissions reductions up to 90% were obtained from this optimization exercise. The performances of the optimized piston bowl geometries were further validated at different operating conditions at the high-load point and at part-load conditions (6 bar IMEP) and compared with those of the baseline designs. The dependence of the engine performance on the piston bowl geometry at part-loads was lower than that at high-loads because injections normally occurred earlier (-60 to-20 CAD after top dead center (aTDC)) where minimal interactions between the spray and piston were anticipated. The interactions between late injections (-3 to 3 CAD aTDC) and piston geometry at high-loads significantly affected, fuel-air mixing, droplet breakup, combustion and emissions. It was also observed that heat losses, dictated by the interactions between the flame and piston surface, significantly affected the performance of the engine.

Original languageEnglish (US)
StatePublished - Apr 14 2020
EventSAE 2020 World Congress Experience, WCX 2020 - Detroit, United States
Duration: Apr 21 2020Apr 23 2020


ConferenceSAE 2020 World Congress Experience, WCX 2020
Country/TerritoryUnited States

Bibliographical note

Publisher Copyright:
© 2020 SAE International; Argonne National Laboratory, operated by UChicago Argonne, LLC, for the U.S. Department of Energy.

ASJC Scopus subject areas

  • Automotive Engineering
  • Safety, Risk, Reliability and Quality
  • Pollution
  • Industrial and Manufacturing Engineering


Dive into the research topics of 'Combustion System Optimization of a Light-Duty GCI Engine Using CFD and Machine Learning'. Together they form a unique fingerprint.

Cite this