Studies On The Asymmetric Reaction Catalyzed By Small Organic Molecules

Real time control of weld pool characteristics during pulsed GTAW process is very important for weld quality, and many factors such as the fluctuation of welding current, displacement of the workpiece, and distortion of the workpiece, will disturb the welding quality. And the control should be based on the precise measurement of the weld pool size. In this thesis, many researches have been carried out on the visual sensing technology of GTAW process, and the 2D and 3D image processing algorithms. Based on these algorithms, software named Obtaining and Image Processing of Weld Pool Vision Characteristics System (OIPWPVCS) has been developed. The software is widely adaptive to different images apart from the ones described in the thesis because many algorithms are integrated to deal with different situations.Firstly, a visual sensing system was described for GTAW experiments to capture the pictures of weld pool image. The proper visual sensor parameters were determined according to the strength distribution of the arc light and spectrum distribution. And a compound filtering system was used to avoid the disturbance of arc and reflection on the surface of the workpiece.The"Obtaining and Image Processing of Weld Pool Vision Characteristics System"is developed based on Visual C++/ MFC according to the linear model of software engineering. The software includes six modules, namely, fiels management, image checking, preprocessing, edge detecting, curve fitting and 3D information aquicision. Shape from Shading (SFS) method was applied to obtain the 3D information of the weld pool.The software is widely adaptive, fast and precise. The experiments on the weld pool of Aluminium, Low carbon steel and Stainless steel demonstrated the effectiveness of the algorithms. And the software is applicable for images besides the ones in the thesis.