地震作用下高层建筑结构的分散模糊迭代学习控制研究

Decentralized fuzzy iterative learning control of tall buildings under earthquake

针对地震作用下高层建筑结构的模糊半主动振动控制方法进行研究,引入分散控制的策略,将模糊逻辑理论与迭代学习控制算法原理相结合,提出分散模糊迭代学习控制DFILC(Decentralized Fuzzy Iterative LearningControl)算法,并应用于高层结构地震反应振动控制中。对某20层钢结构Benchmark结构模型进行数值计算与分析,结果表明,采用分散控制策略的DFILC算法可以有效地抑制结构地震反应,相对于集中控制的单一失效分散控制系统的可靠性更强且数据实时处理效率大为提高;同时,模糊逻辑理论的运用解决了模型的不确定性和外扰的非线性。

Methods for fuzzy semi-active vibration control of tall buildings under earthquakes is studied. By introducing the decentralized control strategy and combining the Fuzzy control and Iterative Learning Control （ILC）, an algorithm named Decentralized Fuzzy Iterative Learning Control（DFILC） has been presented for the decentralized seismic response control of tall buildings. A typical 20-storey seismically excited building is used for comparing the performance of the two control strategies. The results indicate that the DFILC strategy could effectively mitigate structural seismic response as well as the traditional centralized control strategy but is superior in enhancing the reliability and more efficient in the data real- time processing of the control system. It is also more suitable for the control system with model uncer- tainty and external disturbance nonlinearity thanks to the fuzzy logic theory.