RBF-Elman Neural Network Control on Electro-Hydraulic Load Simulator

Abstract:

Due to load simulation system existing strong disturbance, parameters time-variation and nonlinear, there was low control precision, poor adaptive ability and robustness in traditional control algorithm. In order to improve load simulation performance, The RBF-Elman neural network-based adaptive control method is presented. In this way, the load simulator system is identified by the RBF-Elman neural network identifier, which provides model information (Jacobian matrix) to the PID controller and synchronous compensator in order to make it adaptive. Back-propagation algorithms are used to train neural network. The PID controller which is designed by requirement for steady can overcome the shortcoming of the neural network controller. Finally, the simulations confirm that this control scheme results in a quick response, robustness, and excellent ability against disturbance.

• Periodical: Applied Mechanics and Materials (Volumes 433-435)
• Pages: 1054-1060
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.433-435.1054
• Online since: October 2013
• Authors: Xiao Hua Wang*, Shuai Wu, Zong Xiao Jiao
• Keywords: Dynamic Tracking, Electro-Hydraulic Load Simulator, Extraneous Torque, Intelligent PID Control, Neural Network Identification

References

[1] Zongxia Jiao, JunxiaGao, Qing Hua, Shaoping Wang: Chinese Journal of Aeronautics Vol. 17(2004), pp.39-46.

[2] Qing Hua: School of Automation Science and Electric Engineering, Beihang University (2001). in Chinese.

[3] Y. H. Li: Journal of Dynamic Systems, Measurement and Control, Vol. 124(2002), pp.415-419.

[4] ChangnianLiu: The optimum theory of hydraulic servo systems. Beijing, Metallurgy Industry Press (1989). in Chinese.

[5] Zhanlin Wang: Hydraulic servo control. Beijing, Beijing University of Aeronautics and Astronautics Press (1987). in Chinese.

[6] LinaXu: Neural Network Control. Harbin, Harbin Institute of Technology Press(1999). inChinese.

[7] JinkunLiu: Matlab simulation of advanced PID control. Beijing, Publishing House of Electronics IndustryPress(2011). inChinese.

[8] GexinBi: StudyontheRecurrentNeuralNetworkDynamicalIdentificationandItsApplicationtoShip MotionControl. Dalian, School of Machinery, DalianMaritimeUniversity (2009). in Chinese.

[9] Zongxia Jiao, Qing Hua: Chinese Journal of Mechanical EngineeringVol. 39 (2003), pp.10-14. in Chinese.

[10] PhamD. T: Chinese JournalofSystemEngineeringVol. 2(1998), pp.90-97.

[11] Lin F. J, Shieh H. J, Huang P. K: IEEE trans. ultrasonics, ferroelectrics, and, frequency control Vol. 55 (2008), pp.183-198.

[12] Naixiao Zhang, PingfanYan, Yanda Li: Neural Network and Fuzzy Control. Beijing: Tsinghua University Press(1998). in Chinese.

[13] JinzhuPeng, Rickey Dubay: ISA TransationsVol. 50 (2011), pp.588-598.