LAUSR

Abstract In this work, a systematic research was undertaken to predict the nugget size of the welding joints in real time through monitoring the dynamic power signature for TC2 titanium alloy. The welding current in the welding process was measured by a Rogowski coil and the welding voltage was obtained by means of clipping two leads onto the upper and lower electrodes during the welding process. The power signal and the corresponding dynamic resistance were compared with each other. As the constant high frequency alternating welding current was employed in the whole welding process, the power waveform indicated the similar characteristics of the dynamic resistance, besides; the power signal could better reveal an explicit physical meaning of the heat input applied to the welding zone. The disparities and differences of the power waveforms with different welding qualities were also investigated. Four features were extracted from the power waveform so as to indicate the variations of welding power profile in the welding process. A process quality prediction model based on Kriging method was set up to correlate the four extracted characteristics with nugget diameter of the welding joints. Verification test results proved that the method proposed in this paper effectively evaluated the welding qualities within the ranges of process parameters being studied, which could monitor the welding performances in the resistance spot welding process as feasibly and effectively as possible.