Computational & Technology Resources
an online resource for computational,
engineering & technology publications
Civil-Comp Proceedings
ISSN 1759-3433
CCP: 93
Edited by:
Paper 166

Semi-Active Control of Structures using a Combined Genetic Algorithm - Neural Network - Fuzzy Controller

H. Ghaffarzadeh and V. Hamedi

Faculty of Civil Engineering, University of Tabriz, Iran

Full Bibliographic Reference for this paper
H. Ghaffarzadeh, V. Hamedi, "Semi-Active Control of Structures using a Combined Genetic Algorithm - Neural Network - Fuzzy Controller", in , (Editors), "Proceedings of the Tenth International Conference on Computational Structures Technology", Civil-Comp Press, Stirlingshire, UK, Paper 166, 2010. doi:10.4203/ccp.93.166
Keywords: semi-active control, genetic algorithm, fuzzy controller, neural network, FDSAB device.

Semi-active control systems are widely accepted as an effective way to improve seismic performance of the structures. Semi-active control devices can be adjusted to absorb or dissipate energy caused by natural hazards such as severe earthquake and strong wind. Semi-active systems, compared with other systems of control, combine the best features of both passive and active control systems. It has been demonstrated that semi-active control devices have the potential for improving the seismic behaviour of full scale structures [1]. Among many semi active control devices, semi-active friction systems with amplifying braces (FDSAB) are more attractive to use because of their mechanical simplicity, their small size and low operating power requirement [2].

One of the important factors in semi-active systems is the applied algorithm to determine scale change for structure dynamic parameters or control forces. The basic task is to determine a control strategy that uses the measured structural response to calculate an appropriate control signal to send to the actuator that will enhance structural safety and serviceability. Developing nonlinear control algorithms is an important part of using semi-active systems. Many significant algorithms have been proposed to enhance the implementation of semi-active control schemes for the vibration control of civil engineering structures in seismic zones.

In this paper, a combined algorithm is proposed, based on genetic algorithm-fuzzy controller (GA-FLC) and an artificial neural network (ANN), to control structures equipped with semi-active devices. A genetic algorithm technique was utilized to design an accurate fuzzy controller by optimization of the membership functions and fuzzy rules of the fuzzy controller. An expertly trained ANN model was used to reduce the time delay between the response measurement and the control action. The ANN was constructed based on the information of seismic structural response against an earthquake excitation which was obtained from several dynamic time history analyses of a structure.

The proposed control strategy was implemented for the seismic control of a diagonally braced frame using a friction damping system with amplifying braces. The efficiency of the proposed algorithm is demonstrated using the numerical simulation of a seven storey building subjected to earthquake excitation and was compared with a result which was obtained by controlling the frame using the classical control algorithm known as the regulator linear quadratic (LQR) method. The comparison of the controlled and uncontrolled seismic response of the example frame showed a significant response reduction in the controlled frame. The results indicated a better performance for the proposed strategy in comparison with the classical control LQR.

F.Y. Cheng, H. Jiang, K. Lou, "Innovative Systems for Seismic Response Control", CRC Press by Taylor & Francis Group, 2008.
J. Gluckh, Y. Ribakov, "Semi-active Friction System with Amplifying Braces for Control of MDOF Structures", The structural design of tall buildings, 10, 107-120, 2001. doi:10.1002/tal.168

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £145 +P&P)