Font Size: a A A

Study On The Seam Tracking System Based On Vision Sensing In FSW

Posted on:2017-02-17Degree:MasterType:Thesis
Country:ChinaCandidate:D D FengFull Text:PDF
GTID:2271330488951859Subject:Mechanical engineering
Abstract/Summary:PDF Full Text Request
As a new kind of solid-phase welding technology, friction stir welding (FSW) has been widely used for high-strength aluminum alloy welding in aerospace industry. However, upset force and shoulder rotation friction force will be created during the welding process, so workpiece should be clamped, which will cause the deformation of workpiece and lead to the deviation of weld-seam. At the same time, in order to accommodate to the development of intelligent manufacturing in the future, an automatic weld-seam tracking system for FSW is developed in this paper.According to the accuracy demand of weld-seam detection in FSW, CMOS image sensor and 650nm semiconductor laser are used to compose a structured light vision sensing module in which the CMOS sensor is perpendicular to workpiece and the laser is installed inclined to CMOS sensor. Compared with other detection models, it shows that the resolution in depth of field is higher (0.0781 mm/pixel) and the structure of vision sensing module is more compact.Basic principle of calibration with standard tool is analyzed in this paper, and practical calibration experiment is conducted through which regional calibration parameters are obtained. Then calibration parameters are optimized via nonlinear minimization model which improves the accuracy of calibration.Communication module based on OPC is designed for data interchange between PC application program and Siemens CNC System which has good ability of anti-jamming without needing communication card.The main program of the system is completed by Lab VIEW 2011. As to pre-positive distance in the system, a new method of "coordinate storage-coordinate output-coordinate contrast" is used in the program. This paper has also deeply studied image processing methods in each process, and image processing module is designed with functions of image preprocessing, image binarization, skeleton and feature extraction. Meanwhile, smoothing of weld-seam tracking path is also included in the program.At last, experiments and accuracy evaluation method of weld-seam tracking are designed, including diagonal-line welding, polyline welding and curve welding. Through the experimental analysis, accuracy of 3 weld-seam tracking experiments are ±0.741mm,± 0.853mm and±1.086mm respectively. The result shows that it can meet the accuracy demand for common welding in FSW.The main research of this paper is completed in Shanghai Horizon Electronic Technology Co., LTD and receives great support and guidance.
Keywords/Search Tags:FSW, weld-seam tracking, vision sensing, OPC
PDF Full Text Request
Related items