通过测得的熔化极气体保护焊 (GMAW )过程中的电参数 ,利用Kohonen神经网络 ,直接依据不同焊接工艺条件下焊接电压和焊接电流的概率密度分布 (PDD)曲线以及短路过渡时间和燃弧时间等的时间频数分布 (CFD)曲线 ,可以自动识别焊接过程中的各种干扰信号 ,从而实现焊接质量监测。实验验证了该方法的可行性。 Based on the measured electrical parameters of GMAW and the Kohonen neural network, the probability density distribution (PDD) curve of welding voltage and welding current under different welding conditions, as well as the short-circuit transition time and arcing Time and other time frequency distribution (CFD) curve, you can automatically identify the welding process of a variety of interference signals, in order to achieve welding quality monitoring. The experiment verifies the feasibility of this method.