Study On Thermal-magnetic Inversion Imaging Of Nugget In Aluminumalloy RSW

Resistance spot welding (RSW) is one of the important parts in the welding technologies, and it is a material joining method which is high efficient and easy to be automotive. RSW has been widely used in the fields of automobile, areospace and so on. RSW process is a complicated phenomenon, which involves mechanical, thermal, electrical, magnetic and metallurgical factors. Because the process is ephemeral and welding joint region is close, it is difficult to monitor the weld quality effectively. Although the current quality detection techniques have made certain effect, they still have some shortcomings and are only used on certain conditions. Hence, the disadvantage of RSW quality monitoring system limits its application on the fields where higher welding quality is required. Hence, in order to develop new monitoring method, this paper introduces the inverse problem theory, and utilizes the magnetic field and temperature field during RSW to calculate the current distribution in nugget cross-section. The inversion image can be used to evaluate welding quality intuitively, which provides new criterion to judge welding quality.Firstly, thermal-magnetic coupling inversion model is established, and the ill-posedness of the model is analyzed. Secondly, this paper establishes the forward model, based on inversion model’s need, and its reliability is verified by experiments. In order to solve the ill-posedness, this paper introduces the three-dimensional inverse heat conduction model, simulating the temperature field in the nugget cross-section by conjugate gradient method. Finally, this paper combines inverse heat conduction model with the inversion model, using iterative Tikhonov regularization method to solve the current distribution. Thereby, we can analyze the characteristics of inversion image, the effect of different sensor arrangements and the relations between current distribution and temperature field through theories and welding experiments.The thermal-magnetic inversion results show that the inversion image reflects the characteristics of nugget current distribution to a certain extent, which can be used to evaluate the weld quality. The theories results show that the current distribution expands to the edge of nugget as the time increases. While in the welding experiments, sensor arrangement has large effect on inversion, especially in the region of large error, the inversion result is poor. During four arrangements, the result of line arrangements is best, which fits close to the theories result. Applying inverse problem theory and regularization method to RSW is still in the beginning stage, and hence need further research.