Neural Network Active Control of Structures with Earthquake Excitation

  • Cho Hyun Cheol (Department of Electrical Engineering, University of Nevada-Reno) ;
  • Fadali M. Sami (Department of Electrical Engineering, University of Nevada-Reno) ;
  • Saiidi M. Saiid (Department of Civil Engineering, University of Nevada-Reno) ;
  • Lee Kwon Soon (Division of Electrical, Electronic, and Computer Engineering, Dong-A University)
  • Published : 2005.06.01

Abstract

This paper presents a new neural network control for nonlinear bridge systems with earthquake excitation. We design multi-layer neural network controllers with a single hidden layer. The selection of an optimal number of neurons in the hidden layer is an important design step for control performance. To select an optimal number of hidden neurons, we progressively add one hidden neuron and observe the change in a performance measure given by the weighted sum of the system error and the control force. The number of hidden neurons which minimizes the performance measure is selected for implementation. A neural network was trained for mitigating vibrations of bridge systems caused by El Centro earthquake. We applied the proposed control approach to a single-degree-of-freedom (SDOF) and a two-degree-of-freedom (TDOF) bridge system. We assessed the robustness of the control system using randomly generated earthquake excitations which were not used in training the neural network. Our results show that the neural network controller drastically mitigates the effect of the disturbance.

References

  1. T. T. Soong, Active Structural Control: Theory and Practice, Addison-Wesley, 1990
  2. K. Yoshida, S. Kang, and T. Kim, 'LQG control and $H_\infty$ control of vibration isolation for multidegree- of-freedom systems,' Proc. of First World Conference on Structural Control, CA, pp. TP4-43-52, 1994
  3. J.-J. E. Slotine and W. Li, Applied Nonlinear Control, Prentice Hall, 1991
  4. A. Tani, H. Kawamura, and S. Ryu, 'Intelligent fuzzy optimal control of building structures,' Journal of Engineering Structures, vol. 20, no. 3, pp. 184-192, 1998 https://doi.org/10.1016/S0141-0296(97)00077-1
  5. J. Ghaboussi and A. Joghataie, 'Active control structures using neural networks,' Journal of Engineering Mechanics, vol. 121, no. 4, pp. 555- 567, 1995 https://doi.org/10.1061/(ASCE)0733-9399(1995)121:4(555)
  6. D. E. Goldberg, Genetic Algorithms in Search, Optimization and Machine Learning, Addison- Wesley, 1989
  7. S. Haykin, Neural Networks: A Comprehensive Foundation, Prentice Hall, 1999
  8. S. Kung and J. Hwang, 'An algebraic projection analysis for optimal hidden units size and learning rates in backpropagation learning,' Technical Report, Princeton University, 1987
  9. J. Sietsma and R. Dow, 'Neural net pruning-why and how,' Proc. of IEEE International Conference on Neural Networks, vol. 1, pp. 325- 333, 1988
  10. R. P. Lippmann, 'An introduction to computing with neural nets,' IEEE ASSP Magazine, vol. 4, no. 2, pp. 4-22, 1987
  11. G. Mirchandani and W. Cao, 'On hidden nodes for neural nets,' IEEE Trans. on Circuits and Systems, vol. 36, no. 5, pp. 661-664, 1989 https://doi.org/10.1109/31.31313
  12. K. G. Mehrotra, C. K. Mohan, and S. Ranka, 'Bound on the number of samples needed for neural learning,' IEEE Trans. on Neural Networks, vol. 2, no. 6, pp. 548-558, 1991 https://doi.org/10.1109/72.97932
  13. H. C. Cho, 'A study on vibration control of building structure using neural network predictive control,' Master Thesis, Dong-A University, Korea, 1999
  14. M. Saerens and A. Soquet, 'Neural controller based on back-propagation algorithm,' IEE Proc.-F, vol. 138, no. 1, pp. 55-62, 1991
  15. M. Saiidi, 'Hysteresis models for reinforced concrete,' Journal of the Structural Division of ASCE, vol. 108, no. ST5, pp. 1077-1087, 1982
  16. M. S. Fadali and K. E. Zayyat, 'Disturbance rejection control of bridge response to earthquake excitation,' Earthquake Engineering and Structural Dynamics, vol. 25, pp. 291-302, 1996 https://doi.org/10.1002/(SICI)1096-9845(199603)25:3<291::AID-EQE555>3.0.CO;2-0