Applications of the Optimal Unknown Input Observer for CNC Machine Contouring

• Main Authors: Yow-Choung Houng, 洪祐崇
• Other Authors: Pau-Lo Hsu
• Format: Others
• Language: zh-TW
• Published: 1994
• Online Access: http://ndltd.ncl.edu.tw/handle/09770508289806617899

碩士 === 國立交通大學 === 控制工程系 === 82 === In precise manufacturing industries, the accuracy of machin- ing process is the most important requirement. However, In prac- tice, the cutting torque and friction act as the external distu- rbance which decreases the accuracy of machining process badly. Therefore it is crucial to reduce the influence of external disturbance in the precise motion control. In recent two decades, major studies focus on deriving the necessary and sufficient conditions for the existence of unknown input observer (UIO). Few of them have been used for real appli- cations. In this reaaacher, by appling the 2-delay output method to the genernal-structured UIO, a stabilized UIO (S-UIO) is der- ived for non-minimum phase system to estimate both states and unknown inputs. When plant process and measurements are disturb- ed by noises, an algorithm for constructing aaaan optimal S-UIO is derived to minimize the variance of estimation errors. In the present experiments, an external load is added to the X-Y position table with four different controller''s structures (P, CCC, ZPETC, CCC+ZPETC). The Optimal S-UIO estimates disturb- ance for compensating torque so that the effect of external disturbance can be reduced. Experimental results show that both the contouring and tracking errors have been successfully degraded and prove that the proposed method is suitable for motion control.