Abstract
In order to reduce the adverse effect of backlash nonlinearity on dual-motor driving servo system, the system model is firstly described. Then a novel adjustable parameters PID control algorithm combining fuzzy RBF neural network control with PID control is proposed for the first time, and analyzed its stability. In the simulation analysis, the system track step signal, ramp signal and sine signal, respectively. Simultaneously, different interference signal were applied to system output when system tracking different signal. Simulation results show that the proposed control algorithm has smaller tracking error and better anti-interference ability than that of mere PID control, because PID parameters can automatically be adjusted as the system tracking error changes. In the end, white noise interference signal are applied to system output in the experiment analysis. The experiment results also validate the higher tracking accuracy and stronger robustness of the proposed control algorithm than that of PID control.
Additional information
Funding
Notes on contributors
![](/cms/asset/dc95a2b4-e5eb-4f93-bc5d-4088fefa6327/tijr_a_1644207_ilg0001.gif)
Haibo Zhao
Haibo Zhao was born in Tongcheng city of Anhui province of China and received his master’s degree in the School of Automation in 2006 from Nanjing University of Science and Technology, China. He is a vice professor in the electrical engineering department of Tongling University, China. His main research interests include intelligent control and intelligent system.
![](/cms/asset/592c2a18-38cb-4a74-be6a-fb51d6e6aba8/tijr_a_1644207_ilg0002.gif)
Chengguang Wang
Chengguang Wang was born in Yuncheng city of Shandong province of China and received his master’s degree in the School of Automation in 2006 from Nanjing University of Science and Technology, China. He is a senior engineer in Sichuan Institute of Aerospace System Engineering, China. His main research interests include the design, simulation and debugging of high precision digital servo system. Email: [email protected]