Transported probability density function (tPDF) modeling for direct-injection internal combustion engines

E. H. Kung, Daniel Connell Haworth

Research output: Contribution to conferencePaperpeer-review

4 Scopus citations


Ongoing efforts in applying a "high-end" turbulent combustion model (a transported probability density function-tPDF-method) to direct-injection internal combustion engines are discussed. New numerical algorithm and physical modeling issues arise compared to more conventional modeling approaches. These include coupling between Eulerian finite-volume methods and Lagrangian Monte Carlo particle methods, liquid fuel spray/tPDF coupling, and heat transfer. Sensitivity studies are performed and quantitative comparisons are made between model results and experimental measurements in a diesel/PCCI engine. Marked differences are found between tPDF results that account explicitly for turbulence/chemistry interactions (TCI) and results obtained using models that do not account for TCI. Computed pressure and heat release profiles agree well with experimental measurements and respond correctly to variations in engine operating conditions. Computed CO and HC emissions show large deviations from experiment in some cases; further work is required in emissions modeling. With explicit accounting for TCI, other physical submodels that have been developed and calibrated to give acceptable results without consideration of TCI need to be revisited.

Original languageEnglish (US)
StatePublished - Dec 1 2008
Event2008 World Congress - Detroit, MI, United States
Duration: Apr 14 2008Apr 17 2008


Other2008 World Congress
Country/TerritoryUnited States
CityDetroit, MI

All Science Journal Classification (ASJC) codes

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


Dive into the research topics of 'Transported probability density function (tPDF) modeling for direct-injection internal combustion engines'. Together they form a unique fingerprint.

Cite this