| The fillet welding is one of familiar welding seam styles in welding structure, 80 percent of welding seams belong to fillet welding in shipbuilding. A high speed rotating scanning welding torch acts as sensor and arc welding robot acts as controlled object, information processing and path correcting for fillet welding is resolved mainly in this paper. A set of software and hardware path correcting system is built, single chip microcomputer is the core of controller in the system, intelligence of arc welding robot could be improved because of the integration between the system and robot. Low cost could be realized for welding automation.Sensor technique, signal prccessing and autocontrol technique for fillet welding seam tracking are summarized briefly in the paper, and then collection and processing of current signal transduced by rotating arc welding sensor , seam identifying and position correction all could be come ture by 80C196KC MCS. The method of spectrum is applied to identify welding seam deviation, fuzzy control method is applied to seam tracking for height and right-and-left orientation in the paper.PWM speed-regulating is realized by digital and simulative mode for speed control of rotating welding torch, and speed also could be measured. The method of three circles average filtrating and median filtrating are used in combination for pre-processing of current signal, in order to eliminate yawp disturbing.Analyze the influence of molten metal in the molten pool to current signal for fillet welding, and put forward method to eliminate the influence. A method of compensation array is used in the paper, the result after experiment states that welding seam could be tracked exactly, and the shape of molten pool is alse smooth after compensation, the influence of molten metal could be resolved primarily in welding process, so the method is approved to be affective. |
PDF Full Text Request