Study of Axial Thermal Error Compensation for Machine Center by a Multiple-Regression Modeling Analysis Method
| Autor: | Wu Ching-Wei, 吳慶偉 |
|---|---|
| Rok vydání: | 2015 |
| Druh dokumentu: | 學位論文 ; thesis |
| Popis: | 103 High Speed Machining (HSM) technology has become the most important application in the metal cutting industry. The thermal error is the most important issue in high speed machining. Linear encoders are applied to reduce the ball screw thermal deformation problem in high precision metal cutting, however, this approach increases the production cost. Most current researches are focused on thermal error prediction using temperature data. Thermal error ompensation produces better cost ratio compared to the linear encoder and also improves the product quality. Worktable and linear encoder deformation was not considered in the present thermal compensation research method. A new technique is developed in this work to compensate for thermal errors. Two laser measurement systems are applied to acquire worktable and linear encoder thermal expansion. The heat sources are identified using thermal cameras and temperature sensors. The temperature sensors are applied to detect the real time temperature. A mathematical model is built using multivariable regression analysis based on the sensed temperature variation and the thermal expansion. In order to select the best combination of predictive variable data, the multi-collinear problem is obviated by calculating the correlation coefficient. The coefficient of determination (R2) and Mean Square Error (MSE) are determined as optimal indicators of a composite index that identifies the best combination of predictive variables. The experimental results show that the three axial thermal error can be improved by 80.93%. |
| Databáze: | Networked Digital Library of Theses & Dissertations |
| Externí odkaz: |
|