Semi-Active Control of Dynamically Excited Structures Using Active Interaction Control

Citation

Zhang, Yunfeng (2001) Semi-Active Control of Dynamically Excited Structures Using Active Interaction Control. Dissertation (Ph.D.), California Institute of Technology. doi:10.7907/N9QZ-ZP30. https://resolver.caltech.edu/CaltechThesis:10242017-093657605

Abstract

This thesis presents a family of semi-active control algorithms termed Active Interaction Control (AIC) used for response control of dynamically excited structures. The AIC approach has been developed as a semi﷓active means of protecting building structures against large earthquakes. The AIC algorithms include the Active Interface Damping (AID), Optimal Connection Strategy (OCS), and newly developed Tuned Interaction Damping (TID) algorithms. All of the AIC algorithms are founded upon the same basic instantaneous optimal control strategy that involves minimization of an energy﷓based performance index at every time instant.

A typical AIC system consists of a primary structure targeted for vibration control, a number of auxiliary structures, and interaction elements that connect the auxiliary structures to the primary structure. Through actively modulating the operating states of the interaction elements according to pre-specified control logic, control forces favorable to the control strategy are reactively developed within the interaction elements and the vibration of the primary structure is thus restrained. The merits of this structural control approach include both high control performance and minimal external power requirement for the operation of the control devices. The latter is important during large earthquakes when power blackouts are likely to occur. Most encouraging is that with currently available technology this control approach can be readily implemented in real structures.

In this thesis, the cause for an over﷓attachment problem in the original OCS system is clarified and corresponding counter-measures are proposed. The OCS algorithm is reformulated within an energy framework and therefore all of the AIC control algorithms are unified under the same instantaneous optimal control strategy.

To implement the AIC algorithms into multi-degree-of-freedom systems, two approaches are formulated in this thesis: the Modal Control and Nodal Control approaches. The Modal Control approach directs the control effort to certain dominant response modes, and the Nodal Control approach directly controls the response quantities in physical space. It is found that the Modal Control approach is more efficient than the Nodal Control approach. The effectiveness of the AIC control algorithms is verified through numerical simulation results for three-story, nine-story and twenty-story steel-framed buildings. The statistical behavior of the AIC system is evaluated based on a Monte Carlo simulation.

Thesis Files