مهندسی عمران شریف (Feb 2018)
OPTIMAL VIBRATION CONTROL OF STRUCTURES USING WAVELET TRANSFORM AND COLONIAL COMPETITIVE ALGORITHM
Abstract
The active control of structures has a special place in the controlling of linear systems vibrations. Finding the optimum control force, considering the capacity limitation of active control equipment, is the main objective of the design of control systems and one of the most important subjects in active control of structures. Traditional methods, such as LQR, calculate control forces without considering the external loads, such as earthquakes, which make finding of the optimal solution difficult for optimal control problems. Metaheuristic optimization methods, such as colonial competitive algorithm, are powerful algorithms because of their special capabilities which can be used in solving complex problems, but they have not been used coherently in the active control of structures. In this study, an optimization-based control approach for the active control of building structures is proposed against artificial earthquake excitations using wavelet transform and the Colonial competitive algorithm. The 10\% in 50 years earthquake was generated and decomposed via wavelet transform into different frequency levels. Then, in each frequency band, the elements of the gain matrix were searched in the search space with the aim of minimizing the performance index by using the colonial competitive algorithm. Then, the gain matrices obtained from optimization process would be used to calculate control forces. Iterative process of the present method and does not need to solve the Riccati equation, considering external force effects in the calculation of the control forces. In order to evaluate the effectiveness of the proposed controller, vibration of Single-Degree and Multi-Degree of Freedom (SDOF and MDOF) examples against artificial earthquake with the proposed and LQR controller was controlled and compared. Numerical simulation based on the 10\% in 50 years uniform hazard artificial earthquake showed that the performance of the presented control algorithm is better than the LQR controller approach in finding optimal control forces.
Keywords
