Abstract: The voltage between the electrodes in smallscale resistance spot welding (SSRSW) of titanium alloy was collected by data acquisition system. The experiments revealed that the dynamic voltage signal included a lot of welding quality information. In order to demonstrate this finding and monitor the welding quality,the back-propagation artificial neural network (ANN) was employed to forecast the nugget diameter. The maximum predicted error of ANN was about 6.5%. Adjusting and monitoring the voltage waveform could be used to forecast the formation of weld nugget and monitor the quality of welded joints.