摘要
研究了高层建筑在顺风向脉动风荷载激励下T—S模糊控制问题。受控结构为拟建于澳大利亚Melbourne的一幢钢筋混凝土结构办公楼,高306米,总共76层,控制设备为安装于结构最顶层的主动调谐质量阻尼器,主动控制力由T-S型模糊控制器求得。选取LQG控制的仿真结果为T-S模糊控制器的学习样本,采用减法聚类法和混合学习算法对T-S模型进行结构与参数辨识,通过改变结构刚度来检验模糊控制器的鲁棒性。仿真结果表明:T-S模糊控制器鲁棒性好,控制效果能够满足高层建筑风振舒适度的要求;与Mamdani型控制器相比,在线计算时间短,无计算时滞问题。
This paper studies the vibration control of tall buildings under along wind excitation by T-S fuzzy controller. The building under consideration is a concrete office tower proposed for the city of Melbourne in Australia that is 76-story and 306 meters high. The adopted control scheme consists of an Active Tuned Mass Damper (ATMD) where the control force is achieved by a T-S fuzzy controller. The subtraction clustering method and the hybrid-learning algorithm are adopted to train the structure and parameters of the T-S controller. The training data are the simulation results of a LQG controller. The robustness of the controller has been demonstrated through the variation of building stiffness. The results show a good performance by the T-S controller for all cases tested. Because the online computing time taken by T-S controller is much shorter than by Mamdani fuzzy controller, the T-S controller has no time delay problem.
出处
《系统仿真学报》
CAS
CSCD
2003年第4期587-589,共3页
Journal of System Simulation
关键词
高层建筑
T—S模糊控制
模糊控制器
风振反应
钢筋混凝土结构
tall building
wind-excited vibration
fuzzy control
subtraction clustering method
hybrid learning algorithm