Abstract: Based on an information acquisition system, the dynamic characteristics of the pressure signal for resistance spot welding of aluminum alloy. The pressure signals of splashing spot and normal spot are compared and analyzed in time and frequency domains. It was found that there is a sudden change in the pressure signal of splashing spot with obvious characteristic frequency, while the normal spot has no such phenomenon; The db5 wavelet is used to decompose the pressure signal into seven levels to obtain the waveforms in time and frequency domains and the energy ratio of detail signals in different frequency bands. The results showed that according to different forms and characteristics of the pressure signal, the standard deviation of the pressure signal during the power phase, the peak-peak value in the target level, the highest value of the energy ratio and the number of level in the target layer of the wavelet decomposition detail signal can be extracted as the characteristic index to identify splashing spot.