Cornering Error Control Design for Computer Numerical Control Machine Tools

碩士 === 國立臺北科技大學 === 機械工程系機電整合碩士班 === 107 === At present, most of the commercially available CNC machine tools are systems composed of multi-mass. When the drive transmits power to the load for processing, the response will be different during power transmission, and it will be affected by the contro...

Full description

Bibliographic Details
Main Authors: ZHANG,JING-XIANG, 張景翔
Other Authors: YEH,SYH-SHIUH
Format: Others
Language:zh-TW
Published: 2019
Online Access:http://ndltd.ncl.edu.tw/handle/7t2af5
Description
Summary:碩士 === 國立臺北科技大學 === 機械工程系機電整合碩士班 === 107 === At present, most of the commercially available CNC machine tools are systems composed of multi-mass. When the drive transmits power to the load for processing, the response will be different during power transmission, and it will be affected by the controller when performing position control. When the two errors are superimposed, the load position response of the machine tool will be different. Therefore, in this study, the position controller parameters are designed based on the theme of two-axis corner motion and implemented in a dual-quality experimental machine to suppress the corner error. Firstly, the two-mass system is identified by PSO-VCF. The root mean square error is calculated as the adaptive function using the actual velocity response and the simulated velocity response. The DC gain and anti-resonance peak of the two-mass system are used as the limiting conditions for system identification. The system identified by this method is confirmed to have a high degree of matching with the actual system. The controller parameter is designed by the model with a proportional integral speed controller, and the pole design position is given for calculation and the controller parameter map is drawn. Finally, the PSO-W is used to design the PIP position controller parameters and the S-curve velocity parameters to control the two-mass system in order to adjust the control parameters of the corner error and processing time. The time requirement was compared with the accuracy requirement, and the time requirement could be reduced by 17.38%. The required corner error could be reduced by 9.72 times.