Quality assessment of aluminum alloy resistance spot welding based on wavelet and statistic analysis

Abstract

The quality real-time monitoring of aluminium alloy resistance spot welding is realized by distributed multiple-sensor synchronous collection system. Displacement signal time-frequency domain wavelet analysis and further Butterworth band-pass filter statistic analysis show that the range of the electrode displacement signal of the expulsion welding spots can reach about 0.5mm, which far exceeds the normal and undersize welding spots, between 40 to 80Hz. And the electrode displacement original signal statistic range of the undersize welding spots is lower about 1.0mm than other two kinds of welding spots, therefore using the explicit characteristic information can implement resistance spot welding quality assessment manually and further also can finish quality judgment by machine for the high signal-to-noise displacement signal and the briefly easy characteristic extracting method. Moreover, the statistic analysis of 197 samples indicates that the percentage of accuracy can reach 97.6% at the acceptable weld zone. This study has built a theoretical basis for further hardware designing and controlling the quality of aluminium alloy resistance spot welding.