以電腦視覺畫面鑑定車削刀片等級之技術研究

Lathe Cutter Assessments via Computer Vision

陳健立(Chein-Li Chen)

指導教授 ： 鄭宗明

朝陽科技大學/理工學院/工業工程與管理系/碩士(2012年)

摘要

在數位技術發達的今日，製造業亦逐步提高資訊化與自動化的程度，令使用機器取代人力也就越顯重要。CNC工具機之切削刀具的磨耗，對於加工精密度和工件品質乃扮演著重要的角色，而刀片磨耗之檢測卻仍使用目視、觸感、主軸阻抗、或刀尖震動來認知。前二者屬於技術經驗，需要依賴直覺卻易受干擾而不穩定。後二者屬於事後認知，當發現切削不順暢時，不良的切削已然發生，再作任何處理都已是亡羊補牢。使用刀具壽命的小時數去累計與追蹤，常因不想超越極限而須提早停用，造成刀片的浪費。況且，刀片的壽命常因加工的型態而異，並非固定的數值。現行的做法除了無法精確掌握刀片狀況外，亦無法量化地認定刀片等級，因此不利於數位化地掌控加工過程，對於物料耗損及生產程序都只能從模糊中推測。

本研究中，使用CCD攝影機擷取切削完畢時的即時刀刃數位照片，藉由影像處理技術將刀片磨耗之輪廓與範圍定義出，再使用染色程序，將所得影像與全新刀片之影像作重疊比對，得出磨耗、燒焦或斷裂等缺陷。接著將刀片缺陷以量化方式表述，並建立量化的刀刃等級機制，使未來可直接由影像鑑定刀片之等級，能更精準掌控刀片損耗狀況與剩餘壽命，直接提升生產品質與管理效能。

關鍵字

刀具管理 ； 刀片等級鑑定 ； 電腦視覺

並列摘要

Automation has been an inevitable trend in many aspects in the manufacturing industry. The rapid evolution of the information technology also helps speeding up the trend. The purpose is to replace human efforts with automatic mechanisms. In special, the tasks of machining have been carried out by CNC machine tools for decades, for their precise control and repeatability. However, the assessments of cutters were still carried out separately by experienced human workers. Some automation attempts on cutter manipulations were done by cutter tool-life tracking or spindle-torque detections. All of the above were carried out based on vague information, and the results are unpredictable to some extents.

In this thesis, real-time cutter images are captured right after every operation. The tool-tip images are then electronically measured and analyzed. An image evaluation method was then developed to derive a cutter competence level. At the same time, the cutter and its image picture are also evaluated by a tool specialist and acquire a cutter rank. The research method then successively adjusts the relationship weights so that the competence level can be highly related to the cutter rank. A hundred of lathe cutter images were evaluated in this research, and the result shows that the method is capable of ranking cutter images as a tool specialist.

並列關鍵字

Tool Management ； Computer Vision ； Cutter Assessment

參考文獻

Anna, Z., Classifying the wear of turning tools with neural networks, Journal of Materials Processing Technology, vol.109, pp.300-304, 2001.

Balan, G.C., and Epureanu A., The monitoring of the turning tool wear process using an artificial neural network, Intelligent Production Machines and Systems, vol.47, pp.20-25, 2006.

Haber, R. E., Alique, A., Intelligent process supervision for predicting tool wear in machining processes, Mechatronics, Vol.13, pp.825-849, 2003.

Huang, H., Li,A., amd Lin X., Application of PSO-based Wavelet Neural Network in Tool Wear Monitoring, Automation and Logistics, pp.2813-2817,2007.

Huang, K., Wu, Z., and Wang, Q., Image enhancement based on the statistics of visual representation, Image and Vision Computing,vol.109,pp. 51-57, 2005.