Optimization of tuned mass dampers subject to non-stationary random excitation

This paper presents a stochastic approach for the optimum design of a single tuned mass damper (TMD) protecting a linear mechanical system. The optimum strategy is directed towards maximizing the effectiveness of the TMD in reducing the vibration level in the system to be protected. This performance is evaluated on the basis of the reduction of the root-mean-square displacement and acceleration of the main system. Several numerical analyses are carried out for assessing the influence of the system and the excitation characteristics of the optimum TMD. In particular, a comparison between the TMD optimum designs for a system excited by stationary and nonstationary filtered white noise is presented. Finally, the advantages of modelling the seismic excitation as non-stationary filtered white noise for the optimum TMD design are clearly stated.