微切削精度很容易因切削參數的選擇不當或刀具的穩定度等而受到影響。切削參數選擇不佳會引起異常切削振動,造成工件表面粗糙度不佳。傾斜或撓性變形的刀具則會造成刀紋寬度變化和切削路徑偏移。因此一套有效的切削異常及誤差量測方法對於改善微切削的精度是很重要的。然而,微型工件因待測尺寸與量測空間的限制,無法有效檢測與補償精微工具機的加工誤差。本研究發展出以機械視覺為主可於機上進行的刀具振動異常檢測及非等高2D輪廓誤差檢測與補償系統,並透過靈敏度分析了解各誤差源對檢測精度的影響,進行誤差補償改善率驗證。振動異常檢測系統主要是利用灰階切片、AREN法、影像投影法、Canny Edge Method對微型工件進行粗糙比、刀紋深度異常、刀紋寬度、刀紋偏移檢測,結果可做為日後參數調整之依據。非等高2D輪廓誤差檢測與補償系統是利用Canny Edge Method、座標轉換、誤差辨識法及影像匹配原理,可於機上直接量測微型工件輪廓誤差及循跡誤差並補償。最後利用研究室已開發之刀具定位系統、體積誤差檢測及補償系統與研究中非等高2D輪廓誤差檢測及補償系統三項系統進行實機整合實驗,實驗結果證明該系統確實可有效提升微型工具機的加工精度。
The accuracy of a micro machining is very sensitive to the cutting parameters, stability of the micro cutter. Imperfect selection of cutting parameters easily results in abnormal vibrations causing worse surface roughness and non-uniform tool mark depths. A tilted or deflected cutter will cause non-uniform width of tool mark and deviation of cutting path. Thus, an effective measurement method to identify the abnormalities and errors is very important for improving the accuracy of a micro machining process. However, because of the limit of working space, it is very difficult to on-line measure the machining errors from a machined miniature component, and use the results for error compensation to further improve the machining accuracy. Several machine-vision-based measurement methods that can on-machine inspect the abnormal cutting vibration and the non-iso-height contour error were developed in this study. The influence of major errors on measurement accuracy was explored through error sensitively analysis. On grey level comparison method, AREN method, image projection method, and Canny Edge method were used to develop the measurement method. Micro machining experiments were conducted to verify the effectiveness and accuracy improvement efficiency of the developed measurement methods. The experimental results have shown that the measurement method can effectively improve the machining accuracy of a micro machining process.