Integration of Surface Topography Recognition with Laser Inspection and Automated Fractal Finishing Toolpath Generation

• Main Authors: ChenYu Wang, 王振宇
• Other Authors: Chen Chao-chang
• Format: Others
• Language: zh-TW
• Published: 2001
• Online Access: http://ndltd.ncl.edu.tw/handle/84602043651143068039

碩士 === 淡江大學 === 機械工程學系 === 89 === This research is to install a laser inspection system on a CNC machining center and then to integrate a recognition algorithm to identify rough areas on precision mold surfaces and fractal toolpath generation of such rough areas for automated precision mold finishing processes. The rough areas mean the areas requiring surface finishing, and encompasses roughness, waviness and shape error. Usually the mold inspection is performed off-line in mold production. This research is to reduce set-up time and alignment procedure with in-process laser inspection of surface topography of molds. Image processing technique has been developed and used to classify roughness data points into binary data and identify the size, location and orientation of rough areas. The geometric properties of each identified area, such as rectangle, circle, rhombus, and ellipse can be classified and exported for generating the fractal toolpaths with Simple Fractal CAM 2-D software and then converted to NC post processor for CNC grinding and further polishing. The fractal geometry has the properties of self-similarity and scale-independent. Surface finishing processes have been shown to be more lay-free distribution with fractal toolpaths than with conventional zigzag toolpaths under the constraint of the same machining time. Experiments have been performed on a prototype Automated Surface Finishing System (ASFS) based on a CNC machining center equipped with a grinder holder on spindle housing. After EDM, two steps of grinding and two steps of polishing were proceeded. Experimental results have shown the reduction of Ra= 4.7mm( Rmax= 20mm) to Ra=1.14mm (Rmax=5.8mm ). The feasibility of surface finishing error recognition and fractal toolpath generation for automated surface finishing process is then verified to be potential for industrial mold finishing processes.