•  
  •  
 

Abstract

Pneumatic equipment in marine diesel engine and associated piping system often emit extremely high level of noises when releasing pressure. Concerning the crew’s hearing health, discovering an efficient noise silencing device becomes essential. A silencer fortified with extended tubes, perforated tubes, orifice plate, and penetrable resin inlet is presented in order to competently dampen the blown-up noises. A simplified objective function by means of FEM, Artificial Neural Networks (ANNs), and a Genetic Algorithm (GA) is established to enable the numerical calculation when using a finite element method. Three silencer designs, silencer A (with element A, a penetrable resin inlet), B (with element B, a shell composing of extended tubes, perforated tubes and orifice plate), and C (combined with element A and element B), are proposed in this study. The sound transmission of these silencers is analyzed using a finite element program (COMSOL). Additionally, silencer C is numerically optimized using FEM together with Artificial Neural Network and Genetic Algorithm. In order to investigate the efficiency of silencer’s noise abatement, three frequencies (500 Hz, 3000 Hz, and 5000 Hz) representing the whole frequency region are selected as the target frequencies during the Transmission Loss (TL) optimization. Consequently, the shape-optimized silencers within a space-constrained situation are easily accomplished.

Download
COinS