A novel sequential approximate optimization approach using data mining for engineering design optimization

Abstract

For most engineering design optimization problems, it is hard or even impossible to find the global optimum owing to the unaffordable computational cost. To overcome this difficulty, a sequential approximate optimization (SAO) approach that integrates the hybrid optimization algorithm with data mining and surrogate models is proposed to find the global optimum in engineering design optimization. The surrogate model is used to replace expensive simulation analysis and the data mining is applied to obtain the reduced search space. Thus, the efficiency of finding and quality of the global optimum will be increased by reducing the search space using data mining. The validity and efficiency of the proposed SAO approach are examined by studying typical numerical examples.

Keywords:

• engineering design optimization
• sequential approximate optimization
• data mining
• surrogate model

Acknowledgements

Authors would like to thank everybody for their encouragement and support.

Disclosure statement

No potential conflict of interest was reported by the authors.

ORCID

Pengcheng Ye http://orcid.org/0000-0002-9898-441X

Additional information

Funding

The grant support from National Science Foundation [CMMI-51375389 and 51279165] are greatly acknowledged.