Hydraulic servo system command shaping using iterative learning control

Author(s):  
S. Daley ◽  
J. Hatonen ◽  
D.H. Owens
2019 ◽  
Vol XVI (2) ◽  
pp. 31-42
Author(s):  
Mansoor Zahoor Qadri ◽  
Ahsan Ali ◽  
Inam-ul-Hassan Sheikh

Accurate position control of an electro hydraulic servo system (EHSS) is a challenging task due to inherent system nonlinearities, parametric variations and un-modelled dynamics. Since feedback controllers alone cannot provide perfect tracking control, an integration of feedback and feed forward controller is required. A cascaded iterative learning control (ILC) technique for position control of EHSS is proposed in this paper. ILC is a feed forward controller which modifies the reference signal for a feedback fractional order proportional-integral-derivative (PID) controller by learning through current control input and previous error obtained through trails. Unlike other feed forward controllers, ILC works on signal instead of system which eliminates the need of complete knowledge of the system. As compared to other controllers, the proposed technique provides fast convergence without the need of reconfiguring the existing control loop. Simulation and experiments revealed the effectiveness of the proposed technique for EHSS. The obtained results indicated eight percent improvement in rise time and nearly twenty one percent improvement in the settling time.


2019 ◽  
Vol 25 (8) ◽  
pp. 1484-1491 ◽  
Author(s):  
Jing Huang ◽  
Zhenxiang Xu ◽  
Guoxiu Li ◽  
Cheng Qiu ◽  
Haitao Huang

Owing to the control system being repetitive and nonlinear, a time-varying pilot factor control algorithm based on iterative learning control is proposed. The convergence of the TPF-ILC control algorithm is mathematically proven and the sufficient conditions are given. Thereafter, the initial state issue of iterative learning is explored, which is the critical issue of iterative learning control. The convergence of the system’s control error and the initial state of every single period have been mathematically proved by using continuous and repetitive properties of the system, even if the initial states of every single iterative learning period are not strictly the same. At the end of this paper, the TPF-ILC algorithm is applied in a hydraulic servo control system, and experimental results indicate the effectiveness and practicability of the TPF-ILC algorithm.


Sign in / Sign up

Export Citation Format

Share Document