Reinforcement learning for vibration suppression of an unknown system
The article presents a novel approach to vibration suppression that can be used in cases when system under consideration is unknown. The approach is based on a reinforcement-learning of artificial neural network fed with positions and velocities of the system components. The learning routine is designed on the basis of a 1+1 evolutionary scheme with concept of persistence of efficient steps. The performance of the four variants of the algorithm is evaluated on the basis of a simulated 3-DOF system.
KeywordsVibration suppression Artificial Neural Network Reinforcement Learning Unknown system Evolutionary approach
Unable to display preview. Download preview PDF.
The work presented in this paper was supported by the National Science Centre in Poland under the research project no. 2016/21/D/ST8/01678.
- 1.Z.-D. Xu, Y.-Q. Guo, J.-T. Zhu, and F.-H. Xu, Intelligent Vibration Control in Civil Engineering Structures. Academic Press, 1st ed., 2016.Google Scholar
- 3.M. D. Symans and M. C. Constantinou, “Semi-active control systems for seismic protection of structures: A state-of-the-art review,” Engineering Structures, 1999.Google Scholar
- 4.T. T. Soong and B. F. Spencer, “Active, semi-active and hybrid control of structures,” Bulletin of the New Zealand Society for Earthquake Engineering, 2000.Google Scholar
- 5.J. Zhang and P. N. Roschke, “Active control of a tall structure excited by wind,” Journal of Wind Engineering and Industrial Aerodynamics, 1999.Google Scholar
- 6.I. D. Landau, A. Constantinescu, and D. Rey, “Adaptive narrow band disturbance rejection applied to an active suspension - An internal model principle approach,” Automatica, 2005.Google Scholar
- 10.S. H. Zak, Systems and control. Oxford University Press, 2002.Google Scholar
- 11.S. Haykin, Neural Networks A Comprehensive Foundation. Pearson Prentice Hall, 2001.Google Scholar