針對一般CNC工具機,量測及補償機台的體積誤差可在不增加機台成本下提高機台加工精度,另一方面針對微型工件的製造需求,具高精度及微小加工範圍特性要求的Non-MEMS微切削技術也需要誤差量測與補償技術來進一步提升其加工性能。然由於精微工具機的加工空間限制,現行的工業量測技術並無法被滿足其應用需求。本研究分別對一般型工具機的體積誤差量測與補償方法以及對精微工具機加工誤差量測與補償方法進行研究。在一般型工具機應用上,結合Single Socket Method和Element free method,將誤差量測、預測和補償功能合併成一個完整系統,並佐以友善人機介面以方便產業應用;在精微工具機應用上,其誤差量測與補償方法係利用影像濾波、邊緣辨識、座標轉換和影像比對方法,可於機上直接量測微型工件的加工誤差,再利用NC程式自動辨識及編譯系統重新編譯誤差補償後的新NC加工程式碼,作為後續加工之用。本研究改良了過去無法檢測不等高微型加工輪廓的缺點,同時也提出新的加工輪廓相似度比較法,可有效節省檢測的時間。研究中最後也進行實機切削驗證,實驗結果證明所提出誤差補償系統可提升一般型CNC工具機精度62.7%;提升精微工具機加工精度達52.3%。
For general CNC machine tools, the volumetric error measurement and compensation can further enhance its machining accuracy without increasing manufacturing cost. According to the manufacturing demand of miniature parts, the non-MENS micro machining technology with characteristics of ultra-high accuracy and small workspace, also needs error measurement and compensation technology to improve its machining performance. However, due to the limitation of workspace of micro machine tools, the current measurement method currently used in industry cannot be applied. In this study, the volumetric error measurement and compensation method for general CNC machine tools and the machining error measurement and compensation method for micro machine tools were developed. For general CNC machine tools, with use of Single Socket Method and Element free method a complete system with friendly human-machine interface that can perform error measurement, prediction, and compensation was developed. For micro machine tools, with use of image filtering, edge identification, coordinates transformation, and image comparison method, an error measurement system which can on-machine inspect the machining error directly from the workpiece was developed, and an automatic NC-code identification/rewriting system was also developed for generating the error-compensated NC codes for production. The proposed method resolves the drawback of the previous method developed by our lab which cannot measure the errors of contours with different heights. Besides, a new image comparison method which provides better computation efficiency was also proposed. The experimental results have shown that the proposed method can further improve machining accuracy 62.7% for a CNC machine tool, and 52.3% for a micro machining tool.