A New Adaptive Method Based on MRAC and Neural Network for Servo System
Based on neural networks (NNs), combined with model reference adaptive control (MRAC) and classical PID control, a new control scheme is proposed in this paper for a practical uncertain servo system. In this scheme, MRAC and NN controller play significant roles in reducing influences which arose from modeling error, unknown model dynamics, parameter variation and disturbance acted on the velocity-loop to adjust the system to track the nominal velocity-loop model, which is procured through practical experiment. Hence, in the velocity loop, the control system is reconstructed as a linear plant, so that we can use a PID controller in position loop to guarantee good tracking performance. Especially, as an innovation, a robust item is introduced to enhance the robustness of the system. According to Lyapunov approach, the stability of the proposed controller is proved. The simulation results demonstrate that the proposed strategy can achieve high tracking precision for real-time position closeloop system.
Neural Network MRAC Servo System Lyapunov stability
Hongjie Hu Yimin Wu Peng Kou Jiyang Liu
School of Automation Science and Electrical EngineeringBeihang University Beijing 100191, P.R. China Beijing Institute of Radio Metrology and MeasurementBeijing 100854, P.R. China School of Automation Science and Electrical Engineering Beihang University Beijing 100191, P.R. Chin
国际会议
哈尔滨
英文
201-206
2011-01-18（万方平台首次上网日期，不代表论文的发表时间）