現今CNC工具機切削刀具之磨耗掌握,對於加工精度與品質乃扮演重要角色,而刀刃磨耗之檢測卻仍普遍使用目視、觸摸、主軸阻抗或刀尖震動來認知。多數業者常藉由累積工時去追蹤刀具壽命,並以不超越表定壽命來提早停用,造成刀片的浪費。由於刀片的壽命常因加工型態而異,並非固定的數值。現行的做法既無法精確掌握刀片狀況外,亦無法認定刀片之剩餘壽命,因此不利於數位化掌握加工過程,對於物料耗損及生產程序都只能模糊中推測。 本研究中,使用CCD攝影機即時擷取切削完畢時的刀刃數位影像,藉由選配多角度之軸向光源照射,提升高磨耗邊緣輪廓之影像對比,再使用影像處理技術定義出刀片磨耗之輪廓與範圍,最後將所得影像與全新刀片之影像做重疊比對,得出完整磨耗輪廓。未來即可搭配類神經網路等數位工具,更精確掌握刀片損耗狀況與剩餘壽命,直接提升生產品質與管理效能。
Cutter sharpness is the key control factor to the quality and precision of a CNC machined part. It is very important to timely monitor the cutter edge to manipulate the production quality. The monitoring was usually carried out by human visual or tactile examinations when spindle stopped, or by reading spindle resistance or cutter vibration frequency during operation. Most machine operators would assume a constant tool life and accumulate cutting durations to determine the residual life of a cutter. However, tool life is not constant and it highly depends on the course of the processes. This research proposes a mechanism that will take pictures of the cutter tool tips, using CCD camera, with lights from several directions, then extract images of the tip edges from the pictures to reconstruct the appearance of the worn edge. The method will facilitate the cutter monitoring process, and may closely bundled with a numerical AI procedure to emulate worn prediction and tool life. The very function will enhance the intelligence of a CNC machine tool.