Design Optimization for Vibration Reduction of Viscoelastic Damped Structures Using Genetic Algorithms

Abstract

Due to the large number of design variables that can be present in complex systems incorporating visco-elastic damping, this work examines the application of genetic algorithms in optimizing the response of these structures. To demonstrate the applicability of genetic algorithms (GAs), the approach is applied to a simple viscoelastically damped constrained-layer beam. To that end, a finite element model (FEM) derived by Zapfe, which was based on Rao's formulation, was used for a beam with constrained-layer damping. Then, a genetic algorithm is applied to simultaneously determine the thicknesses of the viscoelastic damping layer and the constraining layer that provide the best response. While the targeted response is ultimately at the discretion of the designer, a few different choices for the fitness function are shown along with their corresponding impact on the vibratory response. By integrating the FEM code within the GA routine, it is easier to include the frequency-dependence of both the shear modulus and the loss factors for the viscoelastic layer. Examples are provided to demonstrate the capabilities of the method. It is shown that while a multi-mode optimization target provides significant reductions, the response for that configuration is inferior to the response when only single-mode reduction is considered. The results also reveal that the optimum configuration has a lower response level than when a thick layer of damping material is used. By demonstrating the applicability of GA for a simple beam structure, the approach can be extended to more complex damped structures.

Dynamics of Structures, Frames, and Plates with Viscoelastic Dampers or Layers: A Literature Review

“…Typically, in evolutionary approaches, such as the genetic algorithm (GA) [378][379][380], the artificial bee colony algorithm (ABCA) [376], or particle swarm optimization (PSO) [71,381], different starting points are selected and then the best solution is searched for sequentially. Such a search without calculating the gradient can lead to local optima, especially when there is little variation in the value of the cost function in the domain of design variables.…”

A Simplified Finite Element Approach for Modeling of Multilayer Plates

“…Since the multilayer plate is meshed in many plate nite elements, the sti ness matrix K and mass matrix M can be obtained by integrating the sti ness matrix KK and mass matrix MM of each plate nite element in the plate structure. By exciting a given force vector, the response can be obtained according to the Lagrange formula [21]:…”

Active Control Method Based on Equivalent Stiffness and Damping Coefficient for an Electromagnetic Isolation System

“…Chen et al adopted neural network control methods to attenuate external disturbance and isolate vibration in an isolation platform [20]. Xie et al employed genetic algorithm to isolate vibration of viscoelastic damped structures and verify the applicability of genetic algorithm to vibration isolation of a simple beam structure [21].…”