MRダンパを設置した実物大建築構造模型のニューラルネットワークによる制振

概要

論文の詳細を見る
Semi-active vibration control has attracted a great deal of attention taking advantage of vibration reduction effect, control stability and effective energy consumption. An MR damper is known as one of semi-active control devices. In this paper, experiments for seismic response reduction were conducted on a full-scale building structural model equipped with an MR damper. It is difficult to deal with time variation of properties of the MR damper and nonlinearity of the structure. It is another problem that optimal damping may be varied according to shape and magnitude of a wave inputted to the structure. In order to deal with these problems, neural networks were applied to the vibration control system. As a result of the experiments, it is shown that the displacement of each floor of the structure was reduced by neural network control system as much as optimal damping, simultaneously with suppressing the acceleration to some extent.
一般社団法人日本建築学会の論文
2006-03-20