SEQUENTIAL OPTIMAL CONTROL ALGORITHM FOR VIBRATION CONTROL OF SMART ISOLATED STRUCTURES UNDER MAJOR EARTHQUAKES

摘要

In all technologies of structural control developed, base isolation is regarded as the strategy of structural control which has simple concept, stable performance and relatively twopenny cost. To date, more than one thousand of practical engineering projects worldwide have been implemented with isolation system. However, base isolation is a kind of passive strategy. For the random vibration caused by earthquakes, Passive control systems may not always be efficient. So scholars all around the world attempt to combine active (Reinhorn et al. 1987, Kelly et al. 1987, Ribakov et al. 2001) or semi-active (Feng et al. 1993, Patten et al. 1994) devices with isolation system, resulting in so-called "smart isolated" system. In these control systems, smart isolated system with semi-active control devices do not need to input too much control energy and has a better stability of passive control system and adaptability of active control system, drawing world wide attention in recent years(Symans and Constantinou 1999, He and Feng 2000, Lu et al. 2002). Among all of control devices, MR damper becomes a kind of promising semi-active device because of its nature of quick responding, lower energy consumption, and bigger restoring force and so on. Spencer et al. (1997, 2001) in America have developed successfully the full-scale MR damper offering the maximum of 200KN early and processed the experimental study to a small-scale smart isolated structure (Ramllo et al 2000). Yoshida (Yoshida et al. 1999) in Japan early carried out experimental research to the full-scale smart isolated structure with MR damper. Ou Jin-ping et al. (1999) in china earliest made small-scale MR damper and used it to the model experiment of single degree smart isolated structure (Yang et al. 2002). At same time of the experimental study of control devices, semi-active control algorithm makes a great progress and representative control algorithms have Fuzzy Logic Control(Nagarajaiah 1994, Symans and Kelly 1999, He and Li 2003), sliding Mode Control (Yang and Wu 1996) and LQG Control(Yoshida et al. 1994). Optimal control theory is one of the most useful tools, of which the control level of concerned response can be adjusted based on the predefined control objective function and corresponding weighing matrix, and enables the decision makers determine ideal control force according to their own requirement.