Curved surface machining cutter shaft angle optimization method based on PSO-BP neural network

The invention relates to the technical field of complex surface machining parameter optimization and artificial intelligence, in particular to a curved surface machining cutter shaft angle optimization method based on a PSO-BP neural network, and the method comprises the steps: collecting the arithmetic mean surface roughness of a workpiece after machining according to different cutter shaft angle parameters in a preset cutter shaft angle parameter range, establishing a cutter shaft angle-roughness corresponding database; establishing a cutter shaft angle-roughness prediction model based on a BP neural network; the initial network weight and the initial network threshold are optimized and substituted into the cutter shaft angle-roughness prediction model to obtain a final cutter shaft angle-roughness prediction model; and selecting an optimal cutter shaft machining angle according to the final cutter shaft angle-roughness prediction model. The method can effectively solve the problem that most cutter shaft an