Abstract: A welding parameter self-regulating algorithm based on long-step calibration and partial Newton interpolation were put forward.By this algorithm,not only the unified parameters regulation could be realized,but also the database of the digital welder could be transformed from static mode into dynamic mode,in which the self-learning and self-regulating function were adopted.It might provide an effective strategy for the intelligent evolution of the digital welder.The results of the tests indicated that the welding parameters could be continuously self-regulated in a wide range.By the P-GMAW parameters generated by this algorithm,a stable welding process and well-formed seams could be acquired.By employing the parameter storing method with various priorities,the optimization of the parameters could be achieved.