Application of discrete derivative method with a new frequency mapping technique for adaptive-notch-filter based vibration control in industrial servo systems

Author(s):  
Tae-Il Kim ◽  
Ji-Seok Han ◽  
Ji-Ho Park ◽  
Tae-Ho Oh ◽  
Jee-hyung Lee ◽  
...  
2017 ◽  
Vol 15 (2) ◽  
pp. 668-679 ◽  
Author(s):  
Jong-Min Yoon ◽  
Wook Bahn ◽  
Tae-Il Kim ◽  
Ji-Seok Han ◽  
Sang-Hoon Lee ◽  
...  

This paper describes the vibration control of a flexible manipulator using Filtered-x LMS algorithm. In this study adaptive notch filter is applied to the vibration control of a flexible manipulator model. The adaptive notch filter is designed to estimate multiple vibration mode frequencies of the flexible manipulator and to minimize the effect of vibration. The filtered-x LMS algorithm is applied to make the root strain error and the system input as close to minimization as possible. In the process, the adaptive notch filter learns to eliminate the resonant vibration frequencies of the system. The experimental results show that this presented adaptive notch filter system can suppress the vibration of the flexible manipulator and track the desired joint angles.


2004 ◽  
Vol 2004.41 (0) ◽  
pp. 181-182
Author(s):  
Seiji YOKOE ◽  
Yoshio IWATA ◽  
Hidenori SATO ◽  
Toshihiko KOMATSUZAKI

2003 ◽  
Vol 125 (4) ◽  
pp. 585-594 ◽  
Author(s):  
Yuan H. Guan ◽  
W. Steve Shepard, ◽  
Teik C. Lim

A direct hybrid adaptive approach based on the Lyapunov stability theorem is proposed for performing active vibration control of a rotational gear pair subject to multiple-harmonic, transmission error disturbances. The analysis applies a reduced single-degree-of-freedom gear pair model of the elastic mesh mode with time-varying tooth mesh stiffness. It is assumed that the resultant actuation force for suppressing the rotational vibration of the gear pair can be directly applied along the tooth contact line-of-action by employing a set of suitably configured inertial actuators. The proposed controller simultaneously adapts both the feedback and feed-forward gains, and only requires knowledge of the fundamental gear mesh frequency that is given by the product of the instantaneous gear rotational speed and the number of gear teeth. The analysis indicates that the proposed controller is insensitive to the gear mesh frequency estimation errors, and the resulting vibration control is more effective than those provided by the adaptive notch filter and filtered-x LMS algorithms. The control theory also incorporates dynamic normalization and leakage enhancements in order to optimize performance and improve robustness. Finally, the salient features are demonstrated in several numerical examples.


Sign in / Sign up

Export Citation Format

Share Document