| Summary: | 碩士 === 中原大學 === 機械工程研究所 === 101 === The main purpose of this research is to improve the previously on-line cutting CNC machine tools abnormal intelligent monitoring system, the system includes: (1) abnormal cutting diagnosis module; (2) bi-direction real-time controller information acquisition module. The original system has three types of abnormal cutting diagnosis algorithms for chatter, built-up-edge, and cutter breakage. This study adds new monitoring function for the overload cutting, and it can avoid tools or machine damage due to excessive cutting resistance. Because the cutter breakage diagnosis algorithm can not detect cutter breakage occurring at every initial cutting process, improved the algorithm was proposed in this study. The new algorithm can detect abnormality within 2 seconds and proceed the necessary control within 3 seconds. The system uses signal variation characteristics analysis and Fast Fourier Transfer instead of complex single transfer methods for signal analysis. Fanuc Open CNC Api Spec was used to develop a program which can real-time extract the machine state parameters from the CNC controller with abnormal cutting diagnosis module to improve the diagnosis accuracy. The other part of this study is to establish parameters optimization method for turning process to enhance the machining efficiency. The characteristic curves of feed to root mean square of vibration and the feed to surface roughness were first built with use of curve-fitting, the characteristic curve equations were then obtained for calculation of the allowable vibration. Considering the tool load and allowable vibration value, the program made with Matlab can solve and recommend the optimal cutting parameters for turning process. The experimental results showed that the machining efficiency can be increased two times by the optimization algorithm.
|
|---|
