DSP-based motion control for minimizing residual stage vibration
Motors continue to play an important role in our lives especially in the industry. As the demand for speed and precision increases, vibration generated by motion is no longer tolerable. Literature review yields three main ways of controlling these vibrations. They are the Passive method, the Acti...
Saved in:
Main Author: | |
---|---|
Other Authors: | |
Format: | Final Year Project |
Language: | English |
Published: |
2009
|
Subjects: | |
Online Access: | http://hdl.handle.net/10356/16774 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Nanyang Technological University |
Language: | English |
Summary: | Motors continue to play an important role in our lives especially in the industry. As the demand for speed and precision increases, vibration generated by motion is no longer tolerable. Literature review yields three main ways of controlling these vibrations. They are the Passive method, the Active method and method to minimizing vibration produced by motion. This project explored the third method by using three different motion profiles to move a linear voice coil actuator. The two commonly used motion profiles are the trapezoidal and the S-Curve profiles. One order higher was explored resulting in a third profile called the Smooth-S motion profile.
Most motor controllers in the market do not consider the vibration aspects of a motion system and they do not allow user to modify their controls. Therefore the main objective of this project was to implement a DSP-based motion control for minimizing residual stage vibration. Digital Signal Processor or DSP is good for mathematical control calculations in real time. A conditioning circuit was designed to convert the PWM output from the DSP into an analogue voltage. Circuits were added to prevent the actuators from swinging to the extremes when the program breaks or halts.
PID control was implemented and tuned. The model of the linear voice coil with its supporting circuitries was approximated using 2 methods of Frequency Response extraction. One was by using a linear potentiometer and the other was completely done by the DSP. The second method was better. The purpose of finding the system’s model was to improve the control system and also find the PID constants through mathematical method rather than by trial. However, due to time constrain, the second part was not pursued.
Lastly, an experiment was done with all three motion profiles and the step command on an elevated stage. The Smooth-S motion profile was found to be the best in minimizing the generation of vibration to the stage. |
---|