Vibration reduction using biologically inspired topology optimization method optimal stiffeners distribution on an acoustically excited plate | www.esteco.com

Vibration reduction using biologically inspired topology optimization method optimal stiffeners distribution on an acoustically excited plate

Author(s): 
Enrico Sabbatini, Gian Marco Revel and Marcelo H. Kobayashi (Universita Politecnica delle Marche, University of Hawaii)

CHALLENGE - This paper presents the development of a biologically inspired method for topology optimization and its application to a vibration suppression problem. The problem at hand is the optimization of the vibrational performance of a panel used to shield a vibroacoustic source. The motor is usually installed in the car door and a panel positioned in front of it shields noise. This panel reduces the direct noise transmission in the car cabin. The structure of this panel has to be optimized to reduce the vibration induced from the motor to the panel.

SOLUTION - The multi-objective topology optimization aims to reduce both the vibration amplitude and mass of the plate. Experimental tests are performed for baseline plate model validation and identification of acoustic excitation distribution. A set of solutions are designed by the proposed method and numerically compared with traditional optimization approaches, showing improved performances. Finally, the robustness of the solutions to uncertainty in branch widths is demonstrated. Finite element analysis is employed to assess the performance of different solutions for the reinforcement structures. The model of the steel plate was implemented in COMSOL using shell elements. The optimization algorithm is based on the Sobol design of experiment approach implemented in the software modeFRONTIER. The algorithm chosen is MOGA II. The MatLab environment, used for the implementation of the optimization algorithm, allows the complete control of the input-output variables and is interfaced with COMSOL FE toolbox.

BENEFITS - The proposed method showed improved performance in comparison to traditional optimization methods, with an advantage also in mass savings. Improvements can be in the order of 10–15% (up to over 20%) for both objectives. The results reported in this work demonstrate that, with expected extensions and improvements, this holistic methodology can play a key role in the development of advanced prototyping of dynamic systems.

LOG IN TO DOWNLOAD