Abstract:In view of the problems of high energy consumption, low efficiency and surface quality hard to control in complex surface milling, as well as the problem of determining the relationship between processing parameters and target, proposed the dual neural network optimization method considering complex surface features. First, described the surface characteristics with the curvature representing the complexity of complex surface machining, and established the mathematic mode of milling parameters optimization for complex surface with machining complexity, spindle speed, feed, feed velocity and path spacing as design variables and processing time, energy consumption and surface roughness as objective functions; Secondly, using black-box method with BP neural network established the nonlinear relations of milling parameters to optimizing objects and combined neural network solved by ALM method to optimize the milling parameters. The method solved the parameter optimization of complex surface machining and has an important theoretic guiding role in improving the machining efficiency and quality of complex surface.