Research on the Delay Suspension System Based on Different Learning Rules

Abstract:

Track irregularity is one of the most important aspects of the suspension control performance impact. Due to the delay phenomenon of suspension system, the traditional PID control is unable to accurately track irregularity of track changes. Based on the qualitative analysis to delay suspension system, through modifying conventional PID controller, the article establishes four kinds of Neural Network adaptive PID Controllers based on different learning rules. The control strategy combines by the simple structure of the PID Control, the self-organizing ability of the Neural Network Control and the self-learning ability of the learning rules, adjusts PID parameters on-line, adapts to the change of track irregularity, and has strong robustness. However, the different learning rules have different learning ability of the interferences of track irregularity for the delay phenomenon of the suspension system. Simulation result shows that the supervised Delta learning rules is confirmed more effectively to realize tracking compared with other learning rules, faster response speed, more simple structure, easier operation for the track interference and load disturbance of the delay suspension system. This provides a very good way to solve the interference problem of orbit of the delay suspension system.

• Periodical: Applied Mechanics and Materials (Volumes 380-384)
• Pages: 676-680
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.380-384.676
• Online since: August 2013
• Authors: Rong Rong Song, Zi Li Chen
• Keywords: Delay, Learning Rule, Neural Network PID Control, Suspension System, Track Irregularity

References

[1] Zhiqiang Long, Guohua Shi, Lianchun Wang: Suspension Influence Analysis of Track Irregularity of Maglev Train. Control and Decision Conference. (2010), pp.942-947.

DOI: 10.1109/ccdc.2010.5498085

[2] Hengkun Liu, Xiao Zhang and Wensen Chang: PID Control to Maglev Train System. Industrial and Information Systems. (2009) pp.341-343.

[3] Li Mei, Li Zhongxing, Shen Xufeng and Guo Jiwei: Study on PID Control for Semi-active Air suspension for Riding Comfort. Intelligent Systems(GCIS), 2010 Second WRI Global Congress on. Vol. 2 (2010), pp.47-50.

DOI: 10.1109/gcis.2010.82

[4] Daghooghi Z., Menhaj M.B., Zomorodian A and Akramizadeh A.: A real-time Control of Maglev System Using Neural Networks and Genetic Algorithms. Industrial Technology. (2012) pp.527-532.

DOI: 10.1109/icit.2012.6209992

[5] Jing Xu, Juntao Fei: Neural Network Predictive Control of vehicle Suspension. Information Science and Engineering(ICISE), 2010 2nd International Conference on. (2010) pp.1319-1322.

DOI: 10.1109/icise.2010.5691157

[6] Yizhou Hong, Huangqiu Zhu, Qinghai Wu, Jiaju Chen, Dehong Zhu. Dynamic Decoupling Control of AC-DC Hybrid Magnetic Bearing Based on Neural Network Inverse Method[C]. Electrical Machines and Systems, 2008. 2008: 3940-3944.

DOI: 10.1109/icems.2005.202871

[7] Wang Ruochen, Chen Longjiang, Hao Bin, and Zhang Xiaoliang. Fuzzy Dynamic Model and Neural Network Control of Semi-Active Suspension. Journal of Jiangsu University (NATURAL SCIENCE EDITION). (2009) Vol. 30 No. 1 pp.23-26 (in Chinese).

[8] Lairi, M. and Block, G. A Neural Network with Minimal Structure for Maglev System Modeling and Control. Intelligent Control/Intelligent Systems and Semiotics, Proceedings of the 1999 IEEE International Symposium on. (1999) pp.40-45.

DOI: 10.1109/isic.1999.796627

[9] Shiakolas P.S., Van Schenck S.R., Piyabongkarn D.: Magnetic Levitation Hardware-in-the-loop and Matlab-based Experiments for Reinforcement of Neural Network Control Concepts. Education, IEEE Transactions on. Vol. 47 (2004) pp.33-41.

DOI: 10.1109/te.2003.817616

[10] Yunde Xie, Wensen Chang: Electromagnetic (EMS) Modeling and Simulation of Maglev Train System in Vertical Direction . Journal of the China Railway Society. Vol. 4(1996), pp.47-54(in Chinese).

[11] Yonghua Tao: The new PID control and its application. Machinery Industry Press. 1998(in Chinese).