Metamodel Based Analysis and its Applications: A Review
Murugan Ramu, Prabhu Raja · 2013
Engineering analysis using computer based simulation is used extensively to predict the performance of a system. Such engineering analyses rely on running expensive and complex computer codes. Approximation methods are widely used to reduce the computational burden of engineering analysis. Statistical techniques such as design of experiments and Response Surface Methodology (RSM) are widely used to construct approximate models of these costly analysis codes which minimize the computational expense of running computer analyze. These models referred as metamodels, are then used in place of the actual analysis codes to reduce the computational burden of engineering analyses. Use of metamodels in the design and analysis of computer experiments has progressed remarkably in the past three decades. This paper reviews the state of the art of constructing metamodels and its evolutions over the past three decades.