為了改善元件於銑削加工時因機械之動態異常現象,因而影響其之精度及表面光滑度,因此針對近來研究室已發展的銑削異常診斷系統進行線上切削時的實驗驗證,以大量切削所獲取的訊號加以整理,逐一分析,以提升系統的可靠度及精確之門檻值。 為了確認所收集到的訊號是對資料分析有幫助的,於事前對於相關之研究設備再加以強化,並規劃了銑前實驗所需步驟,將銑削時以不同之R.P.M配合每分鐘進給率、計算切屑厚度、刀刃通過頻率(TPF)、試誤法等作大量切削來取得所需之訊號。 將取得之X軸Y軸的位移訊號,利用快速傅立葉轉換後取得順滑層及頻譜分析圖並應用小波轉換來得到微小訊號動態,以作為發生顫振、刃口積屑時判斷之依據。
Abstract In order to improve the dynamic abnormal phenomenon of the components that effect the precision and superficial evenness when processed by milling. The aim of the thesis is to collect the on line cutting of the milling abnormal system experiment confirmed information that developed by the Laboratory. The signals gained from the massive cutting were reorganized and analyzed item by item to promote the threshold value of margin of reliability and precision. In order to confirm the collected signals are helpful for information analysis, strengthening the correlative research equipment at first and scheming the steps of the experiment before milling. When milling started, using different R.M.P with the feeding rate per minute, counting the thickness of chip, TPF and try and error method. Transforming the collected the displacement signals of X axis and Y axis into. After using the fast Fourier transformation, the along slippery layer and spectral analysis chart could be got, and using wavelet transformation to get small dynamic signals to be the j