A multi-fidelity RBF surrogate-based optimization framework for computationally expensive multi-modal problems with appl
- PDF / 3,140,918 Bytes
- 21 Pages / 595.224 x 790.955 pts Page_size
- 35 Downloads / 193 Views
RESEARCH PAPER
A multi-fidelity RBF surrogate-based optimization framework for computationally expensive multi-modal problems with application to capacity planning of manufacturing systems Jin Yi1 · Yichi Shen2 · Christine A. Shoemaker3 Received: 9 November 2019 / Revised: 14 February 2020 / Accepted: 12 March 2020 © The Author(s) 2020
Abstract This paper presents a multi-fidelity RBF (radial basis function) surrogate-based optimization framework (MRSO) for computationally expensive multi-modal optimization problems when multi-fidelity (high-fidelity (HF) and low-fidelity (LF)) models are available. The HF model is expensive and accurate while the LF model is cheaper to compute but less accurate. To exploit the correlation between the LF and HF models and improve algorithm efficiency, in MRSO, we first apply the DYCORS (dynamic coordinate search algorithm using response surface) algorithm to search on the LF model and then employ a potential area detection procedure to identify the promising points from the LF model. The promising points serve as the initial start points when we further search for the optimal solution based on the HF model. The performance of MRSO is compared with 6 other surrogate-based optimization methods (4 are using a single-fidelity surrogate and the rest 2 are using multi-fidelity surrogates). The comparisons are conducted on a multi-fidelity optimization test suite containing 10 problems with 10 and 30 dimensions. Besides the benchmark functions, we also apply the proposed algorithm to a practical and computationally expensive capacity planning problem in manufacturing systems which involves discrete event simulations. The experimental results demonstrate that MRSO outperforms all the compared methods. Keywords Surrogate-assisted optimization · Multi-fidelity surrogate · Radial basis function · Computationally expensive problems
1 Introduction Computationally expensive multi-modal black-box optimization problems arise in many science and engineering areas. For example, in the field of complex products design Responsible Editor: Nathalie Bartoli Electronic supplementary material The online version of this article (https://doi.org/10.1007/s00158-020-02575-7) contains supplementary material, which is available to authorized users. Christine A. Shoemaker
[email protected] 1
2
3
Environmental Research Institute, National University of Singapore, Singapore, Singapore Institute of Operation Research and Analytics, National University of Singapore, Singapore, Singapore Department of Industrial Systems Engineering and Management, National University of Singapore, Singapore, Singapore
optimization, the optimization analysis is driven by expensive computer simulation codes (e.g., finite element analysis (FEA) and computational fluid dynamics (CFD)), which are used to simulate the physical processes and to evaluate the design solutions (Simpson et al. 2004). Another example can be found in the hyper-parameter tuning process of machine learning algorithms. Each evaluation of the hyper-parame
Data Loading...
