Assessment of Optimization and Regression Methods for Engine Optimization

  • Yu Shi
  • Hai-Wen Ge
  • Rolf D. Reitz


Engine optimization problems by nature are multi-objective problems, which involve simultaneously optimizing multiple design parameters. Based on the review of optimization methods in  Chap. 2, it was determined that multi-objective genetic algorithms (MOGA) are an appropriate optimization method. This chapter assesses the performance of different MOGAs for engine optimization problems. The assessment was conducted using three popular MOGAs [μ-GA (Coello Coello and Pulido 2001), NSGA II (Deb et al. 2002), ARMOGA (Sasaki and Obayashi 2005)] applied to a heavy-duty diesel engine operated at a high-load condition. In addition to this assessment, the niching technique of NSGA II was also evaluated. Convergence and diversity metrics of MOGAs were defined to complete the assessment of different niching techniques. Regression analysis was then conducted on the design datasets that were obtained from the optimizations with two niching strategies


Pareto Front Pareto Solution Radial Basis Function Bezier Curvature Soot Emission 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Copyright information

© Springer-Verlag London Limited 2011

Authors and Affiliations

  1. 1.Department of Chemical EngineeringMassachusetts Institute of TechnologyCambridgeUSA
  2. 2.Engine Research CenterUniversity of WisconsinMadisonUSA

Personalised recommendations