A Shape Parameterization Method Using Principal Component Analysis in Applications to Parametric Shape Optimization

Abstract

This paper proposes a shape parameterization method using a principal component analysis (PCA) for shape optimization. The proposed method is used as a preprocessing tool of parametric optimization algorithms, such as genetic algorithms (GAs) or response surface methods (RSMs). When these parametric optimization algorithms are used, the number of parameters should be small while the design space represented by the parameters should be able to represent a variety of shapes. In order to define the parameters, PCA is applied to shapes. In many industrial fields, a large amount of data of shapes and their performance is accumulated. By applying PCA to these shapes included in a database, important features of the shapes are extracted. A design space is defined by basis vectors which are generated from the extracted features. The number of dimensions of the design space is decreased without omitting important features. In this paper, each shape is discretized by a set of points and PCA is applied to it. A shape discretization method is also proposed and numerical examples are provided.