Structural Design Optimization for Reducing Sound Radiation from a Vibrating Plate Using a Genetic Algorithm

Abstract

This paper presents a numerical method of optimum design of mechanical structures for reducing noise. This method couples an optimization technique based on a genetic algorithm (GA) with both a finite element code for determining structural vibrations and acoustic radiation analysis for determining sound pressure levels at the receiver point. Reducing the sound pressure level associated with the fundamental acoustic resonant mode is accomplished by adding masses to a vibrating plate coupled to an acoustic cavity which is excited by a sound source. The positions of the masses on the plate are adopted as the design variables. By using the presented method, the optimal positions that would not be easy to be found by any intuitive solution are determined so as to effectively reduce the sound pressure level. The numerical prediction of the effect of the optimization is compared with the experimental result.