An improved mode-pursing sampling method that balances global exploration and local exploitation based on kriging

Abstract

Mode-pursing sampling (MPS) is an efficient approach for solving expensive optimization problems. However, MPS does not actively seek sample points while considering the error information of the surrogate model. Hence, an improved MPS based on kriging (IMSK) is proposed in this article. This method employs the probability improvement function as a probabilistic distribution function and obtains multiple sampling points with a certain probability from a candidate set composed of the exploitation–exploration trade-off points in each iteration. In addition, two acceleration strategies are proposed to speed up the sequential optimization process. Several typical benchmark functions and an engineering problem are applied to test the performance of IMSK and compare it to that of different versions of MPS. The results show that IMSK can provide good optimized solutions at the same or lower computing costs. These properties may make IMSK suitable for addressing actual engineering optimization problems.

KEYWORDS:

• Mode-pursing sampling
• global optimization
• expensive black-box problems
• kriging model

Disclosure statement

No potential conflict of interest was reported by the authors.

Data availability statement

The data that support the findings of this study are available from the corresponding author upon reasonable request. The authors attest that all data for this study are included in the article.

Additional information

Funding

This work was supported by the Fundamental Research Funds for the Central University of Central South University [grant number 2019zzts267] and the State Key Laboratory of Heavy Duty AC Drive Electric Locomotive Systems Integration [grant number 2017ZJKF09].