Research On Sliding Mode Variable Structure Control Of Structural Vibration

The contents in the thesis belong to the field of structure control in seismic engineering, which consists of the researches of control implement and control strategy. Although some meaningful progresses in this field have been made and many control implements have been successfully applied to practical engineering, there are still some problems at the field of structure control strategy. The effects of some complicated factors, such as nonlinear of controlled system, intense time-change, uncertainty of structural parameters, discretization of continuous-time controlled system, complexness and variability of structure conformation, and time delay in control, bring the research of structure control strategy many serious difficulties. Therefore, it is meaningful to seek structure control algorithms which are robust to uncertainty of structural parameters, and to study the influence of theirs on control of structure vibration under the condition of system discretization, structural decentration and time delay in control.Based on the review of past research work, a state-of-art with respect to vibration control of structure, is present firstly. Then, new research on the application of sliding mode variable structure control(SMVSC) methods to vibration control of structure, is explored in this thesis with different arrangement of ideas. From the point of view of the oriented control implements, the sustain ability of SMVSC algorithms to active mass damper(AMD) and tuned liquid column damper(TLCD), is studied in the thesis; From the point of view of the type of controlled structures, adjacent buildings, eccentric buildings and offshore platforms are studied and analyzed; From the point of view of time type of control system, discrete-time systems which are relevant to practical engineering, are adopted in the paper; From the point of view of the improvement of SMVSC algorithms themselves, not only the general SMVSC method, but also fuzzy-based SMVSC and BP networks-based SMVSC are applied to structure control; From the point of view of the characteristic of control analysis, the influence of time delay in control on control quality, is discussed. Along the above-mentioned angles of study, the major contents in the thesis are summarized as follows:A discrete-time SMVSC method for adjacent buildings with time delay in control is developed . Firstly, the discretization theory of continuous system without time delay and with time delay, is summarized, respectively. Secondly, continuous state equation of motion for the adjacent building is set up, which is also discretized and transformed into standard discrete form.Lastly, the effectiveness of the SMVSC algorithm is verified by a simulation example and the influence of time delay in control on structure control is analyzed.A discrete-time SMVSC method for AMD-eccentric buildings with time delay in control is developed . Firstly, not only mechanic model but also continuous state equation of motion is set up, which is also discretized and transformed into standard discrete form on the basis of the deduction result of two-input discrete system . And then, a simulation example is used to validate the SMVSC method. Furthermore, the influence of time delay in control on structure control is also discussed.A discrete-time fuzzy-based variable structure control method for TLCD-eccentric buildings is developed. Firstly, the numerical model of a multi-story eccentric building-TLCD system is given and its continuous state equation is set up and discretized. And then, the theory of fuzzy variable structure control(FVSC) is simply discussed and the semi-active control strategy is established and implemented by FVSC. Lastly, a numerical simulated example proves that the FVSC algorithm is an effective method for controlling the translational response and rotational response of the eccentric buildings, furthermore, the SMVSC method possesses better robustness to uncertainty of structure parameters.A discrete-time ANN(artificial neural networks)-based variable structure control (ANNVSC) method for TLCD- platform system is developed . Firstly, a passive TLCD test is performed on the basis of modal analysis of a practical platform and some important characteristics of passive TLCD are gained. And then, based on the test, a semi-actively tuned liquid column dampenTLCD) based on discrete-time neural networks sliding mode law is proposed to vibration control of fixed offshore platforms excited by earthquake. Lastly, by a numerical simulation, the control effects of semi-active TLCD based on ANNVSC are compared with those of passive TLCD . Superiority of the former is proved, i.e. by a little energy cost, the former can possess better control ability than the latter. Moreover, the robustness of the ANNVSC algorithm can ensure system performance and stability.The research of Sliding Mode Variable Structure Control of Structural Vibration is a new project. The fruits of these fields will advance the overall development of structure control technology.