基于磁流变阻尼器的结构振动优化控制

磁流變 (MR)阻尼器利用 MR流提供可控性是當今最新的半主動控制裝置 ,工程應用前景十分廣闊。但由于 MR阻尼器所固有的高度非線性動特性 ,使得描述其阻尼力—輸入電壓關系的逆向動特性的數學模型很難得到 ,而逆向動特性模型對實現整個控制策略又是至關重要的。本項研究針對這一問題提出運用神經網絡技術建立MR阻尼器的神經網絡模型來模擬其逆向動特性 ,并設計與之相適應的控制系統 ,建立起基于 MR阻尼器的結構振動控制的有效分析方法 ,通過數值仿真結果探討所提出的結構控制策略的有效性.A Magnetorhrological (MR) damper that uses MR fluids to provide controllable characteristics is one of the most promising semiactive control devices for civil structural vibration control. However, due to their highly nonlinear dynamic behaviour, it is very difficult to obtain an inverse MR damping mathematical model that has an explicit relationship between the desired damping force and command signal (voltage) . This force voltage relationship is required especially for structural vibration control design using MR dampers. In this study, we explore such a possibility via the neural network (NN) technique. Recurrent NN models are constructed to emulate the inverse dynamics properties of the MR damper. It is attemted to design a corresponding control system, and develop an effective dynamic analysis method for structures treated with MR dampers. Numerical simulations are also presented to illustrate the effectiveness of the proposed control strategy.