Projeto ótimo robusto de um absorvedor dinâmico de vibrações

Abstract

This work aims at the development and application of an alternative methodology for the robust optimum design of a dynamic vibration absorber, which is a widely used passive vibration control technique. The purpose of this works is to improve the dynamic performance of the device under parametric uncertainties, within in the primary system, that are modeled as random variables with normal distribution. For the accomplishment of this work, the structure of a degree of freedom was built, which after is tuned to dynamic vibration absorber. The structure parameters were identified through inverse problem techniques. The Monte Carlo Simulation combined with the Latin Hypercube sampling is used as a stochastic solver to evaluate the structure uncertain dynamic response. The sensitivity analysis is applied in order to determine the most preponderant uncertain parameters in response variability. The robust optimization is implemented ally with the Monte Carlo Simulation and solved using a multiobjective genetic algorithm. The results demonstrate the significant influence of uncertainties on the mechanical system dynamics. Moreover, the optimal design permits to obtained improvements, proving the efficiency of the proposed methodology and presenting itself as a viable alternative for passive improvement of uncertain mechanical systems.