Construction and Analysis in Combining the Taguchi Method and the Back Propagation Neural Network in the PEEK Injection Molding Process

Abstract

In order to promote the accuracy and reduce the dimension deviation of injection molding, the Taguchi quality method and the Back Propagation Neural Network (BPNN) were adopted in the injection molding process. The aim of the research is to enhance the optimization of the injection molding process using the most precursory material, Polyether Ether Ketone (PEEK), in the current plastic industry. First, the Taguchi quality method is applied to establish the design of experiment. An analysis of variance is done to arrive at the significant factors that influence the injection molding quality the most. Also, the significant factors are used to construct the BPNN prediction system. Next, the BPNN is employed to fine-tune the optimum conditions obtained from the Taguchi method for achieving optimum quality. The experimental results are within the 95% confidence interval, confirming that the factor's effects are reproducible and that the experimental results are reliable.