Study on Mo process optimization of double glow plasma surface alloying based on BP neural network

Abstract

On the basis of orthogonal experiment in double glow plasma suface alloying permeability Mo process, using the genetic algorithm of back propagation neural network (BPNN) GA, pole spacing and heat preservation temperature, holding time was studied, the source voltage and working pressure of process parameters on the double glow ion permeability molybdenum permeability layer thickness, the influence of the optimization of the double glow ion permeability Mo process test parameters. The predicted results are in good agreement with the actual test results, and the absolute coefficient (R2) is 0.964. The recommended optimal prediction method can get representative results for both the optimal and various penetration thickness predictions. This paper provides a new method for the selection of the optimum process scheme of double brilliance ionic infiltration Mo process.