Approximate optimization of high-speed train nose shape for reducing micropressure wave

The paper deals with the nose shape design of high-speed railways to minimize the maximum micropressure wave, which is known to be mainly affected by train speed, train-to-tunnel area ratio, slenderness and shape of train nose, etc. It is advantageous to develop a proper approximate metamodel for replacing the real analysis code in the context of approximate design optimization. The study has adopted a newly introduced regression technique; the central of the paper is to develop and examine the support vector machine (SVM) for use in the sequential approximate optimization process. In the sequential approximate optimization process, Owen's random orthogonal arrays and D-optimal design are used to generate training data for building approximate models. The paper describes how SVM works and how efficiently SVM is compared with an existing Kriging model. As a design result, the present study suggests an optimal nose shape that is an improvement over current design in terms of micropressure wave.