Pipe fitting welding process automatic generation method and system based on machine learning

The invention discloses a pipe fitting welding process automatic generation method and system based on machine learning, and the method comprises the following steps: collecting welding workpiece parameters of pipe fittings of different types and specifications and corresponding optimal welding process parameter data, and constructing a sample set; constructing a BP neural network; training the BPneural network by using the sample set to generate a welding process generation model; and utilizing the welding process generation model to automatically generate optimal welding process parametersof different types and specifications of pipe fittings. Aiming at the problems that the design of a pipe fitting welding process in a machining workshop is mainly based on a traditional manual mode, alarge amount of repeated operation is caused and the production quality and the cost are reduced in the work of the pipe fitting welding process, the welding process parameter adjusting process is simplified, high-quality proc