Title :
Co-simulation of neural networks PID control for ship steering hydraulic system
Author :
Lin, Fuquan ; Lin, Shaofen ; Zeng, Shengbin
Author_Institution :
Dept. of Marine Eng., Univ. of Jimei, Xiamen, China
Abstract :
For optimization of the dynamic performance of ship steering hydraulic system, a back propagation (BP) neural networks PID control algorithm was studied based on the characteristics of nonlinearity and time variation of the system. Then the control system model in MATLAB/Simulink was designed according to the algorithm, and a co-simulation between the control system and the hydraulic system was carried out via the interface of MSC.EASY5. The co-simulation results show that the PID control mode presented in this paper has the capabilities of self-studying and self-adapting. Results also reveal that the control mode can obtain better robustness and faster response than the conventional PID control. Conclusions are drawn that the application of neural networks PID control on complex and nonlinear hydraulic system control can bring about better performance.
Keywords :
backpropagation; hydraulic systems; neurocontrollers; nonlinear control systems; ships; steering systems; three-term control; MSC interface; Matlab; Simulink; backpropagation neural networks PID control algorithm; complex system control; nonlinear hydraulic system control; nonlinearity characteristic; self-adapting capability; self-studying capability; ship steering hydraulic system; time variation characteristic; Control system synthesis; Control systems; Hydraulic systems; MATLAB; Marine vehicles; Mathematical model; Neural networks; Nonlinear control systems; Nonlinear dynamical systems; Three-term control; Co-simulation; Hydraulic System; Neural Networks;
Conference_Titel :
Information and Automation (ICIA), 2010 IEEE International Conference on
Conference_Location :
Harbin
Print_ISBN :
978-1-4244-5701-4
DOI :
10.1109/ICINFA.2010.5512490
