Designing an internal combustion engine involves compromising among multiple performance metrics and targets with multiple control and noise factors. The main challenges are in determining the critical performance metrics, finding the optimal compromise between these metrics, and correctly represent the most important control and noise factors through CAE modeling and optimization. This paper presents a methodology for practical application of robustness and performance optimization using a CAE model. The key element of the methodology is a concept of surrogate noise. With this concept, the multiple noise factors affecting the system performance are represented through a limited number of noise factors for CAE modeling. The other part of the methodology is to substitute complicated and computationally time intensive CAE modeling with a cheap-to-compute Gaussian Kriging model through Optimal Sampling and Design of Experiment. The final part of the methodology is performing multi-criteria robustness and performance optimization as well as performance and robustness confirmation of the optimal design point. The proposed methodology has been applied to a practical problem of designing the IC engine main bearing system. The results of the analysis have provided practical recommendations and directions to drive the main bearing system design. In this paper, the methodology is demonstrated through the presentation of a simplified form of this investigation.

This content is only available via PDF.
You do not currently have access to this content.