The invention relates to intelligent
welding control, in particular to a TIG (
Tungsten Inert Gas)
welding method based on model-free adaptive
iterative learning control, which comprises the following steps of: setting an initial
welding current and a
weld pool back width expected value; establishing a model-free adaptive
iterative learning control scheme about the TIG welding controlled
system; a model-free self-adaptive
iterative learning control model is used for calculation to obtain welding currents at all moments in the
welding process, and a workpiece is welded; the
weld pool back width at each moment in the
welding process is detected in real time, and the error between the
weld pool back width at each moment and the weld
pool back width expected value is calculated; if the
maximum error absolute value is larger than a preset threshold value, the welding currents at all moments in the
welding process are updated, the next workpiece is welded, and otherwise, the welding currents at all the current moments are directly used for welding the next workpiece; according to the technical scheme provided by the invention, the defect that relatively high welding quality is difficult to maintain in a complex TIG welding environment can be effectively overcome.