Abstract
The widespread use of computer experiments for design optimization has made the issue of reducing computational cost, improving accuracy, removing the “curse of dimensionality” and avoiding expensive function approximation becoming even more important. Metamodeling also known as surrogate modeling, can approximate the actual simulation model allowing for much faster execution time thus becoming a useful method to mitigate these problems. There are two (2) well-known metamodeling techniques which is kriging and radial basis function (RBF) discussed in this paper based on widely used algorithm tool from previous work in modern engineering design of optimization. An integral part of metamodeling is in the method to sample new data from the actual simulation model. Sampling new data for metamodeling requires finding the location (or value) of one or more new data such that the accuracy of the metamodel can be increased as much as possible after the sampling process. This paper discussed the challenges of adaptive sampling in metamodel and proposed an ensemble non-homogeneous method for best model voting to obtain new sample points.
Highlights
Computer simulation is ubiquitous in engineering design and optimization to solve the complex system
Typical types of surrogate modeling that recently used by the researcher is polynomial response surface (PRS), kriging or Gaussian process, radial basis function (RBF), support vector regression (SVR), multivariate spline (MARS), and artificial neural network (ANN)
Adaptive sampling has the potential to proliferate in research area optimization to refine the model and improve accuracy of metamodel
Summary
Computer simulation is ubiquitous in engineering design and optimization to solve the complex system. Another work published by the same author review kriging metamodel in experimental design and proposed “robust” optimization accounts for uncertainty in some simulation input with the Taguchi method. Unlike the comparison of six metamodels, comparative research between kriging and RBF metamodeling techniques for design optimization of variable stiffness composites indicates that both models are the most precise and robust model in design space exploration. Zhou (2016) in his research implement ensemble method to select suitable metamodel for objective-oriented sequential sampling proposed genetic algorithm (GA) to optimal weight. Based on a comprehensive review, the main objective of this paper to propose a new approach of implementing an ensemble method in adaptive sampling stage to enhance sampling technique in order to improve metamodel approximation. The contribution of this paper is: (i) proposed method of infill sampling criteria which include deterministic and metaheuristic method for new sample selection to improve accuracy of metamodel, (ii) employ non-homogenous ensemble method for model voting the best location of new sample
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
