Title :
Evaluation of friction welded joint performance by recognition of burr shape (in the case of A5056 aluminum alloy)
Author :
Ogawa, Koichi ; Kawai, Gota ; Sakurai, Keizo ; Yamaguchi, Hiroshi ; Takashima, Yoshinao
Author_Institution :
Osaka Prefecture Univ., Sakai, Japan
Abstract :
A strong correlation is believed to exist between the burr shape and friction welded joint performance. The possibility of evaluating welded joint performance from burr shape, which is a non-destructive inspection method, will lead to its application in evaluating joint performance. Thus, in this study burr shape data is fed into an image processing equipment and a precise measurement of the various burr features is performed. The following conclusions were derived from the joint performance experiment: for A5056 aluminum alloy friction welded joints, a clear line could not be drawn between perfect and imperfect welding solely from the tensile test of the welded joint. Thus, the discriminant analysis of whether the weld is perfect or imperfect using variables such as height of burr groove, width of burr, burr height and sectional area of burr only gave a 90% of right answers
Keywords :
aluminium alloys; automatic optical inspection; computer vision; feature extraction; image processing; welding; A5056 aluminum alloy; burr shape recognition; computer vision; discriminant analysis; feature extraction; friction welded joint; image processing; non-destructive inspection; performance evaluation; Aluminum alloys; Chemicals; Friction; Image processing; Inspection; Performance evaluation; Shape measurement; Shape memory alloys; Testing; Welding;
Conference_Titel :
SICE '95. Proceedings of the 34th SICE Annual Conference. International Session Papers
Conference_Location :
Hokkaido
Print_ISBN :
0-7803-2781-0
DOI :
10.1109/SICE.1995.526649