OPTIMIZATION OF PLASTIC INJECTION MOLDING PROCESS VIA TAGUCHIS PARAMETER DESIGN METHOD, BPNN, AND DFP

摘要：

This research integrates Taguchis parameter design method, back-propagation neural networks (BPNN), and Davidon-Fletcher-Powell (DFP) method to optimize the process parameter settings of plastic injection molding. Taguchis parameter design method is used to arrange an orthogonal array experiment and to reduce the number of set-test cycles. Injection Time, Velocity Pressure Switch Position, Packing Pressure, and Injection Velocity are selected as process control parameters and product weight is selected as the single product quality. The experimental data of Taguchis parameter design method are used for effectively training and testing BPNN. Then, BPNN is combined with DFP to search out the final optimal process parameter settings. Finally, a confirmation experiment is performed to confirm the effectiveness of the final optimal process parameter settings. The experimental results show that the proposed effective process parameter optimization approach can avoid shortcomings inherent in the application of trial-and-error processes or the conventional Taguchi parameter design method. Furthermore, the proposed approach can effectively assist engineers in determining optimal initial process parameter settings and achieving competitive advantages on product quality and costs.

关键词： Plastic injection molding Back-propagation neural networks Taguchis parameter design Davidon-Fletcher-Powell method

作者: W.C.CHEN M.W.WANG G.L.FU C.T.CHEN

作者单位: Graduate School of Industrial Engineering and System Management, Chung Hua University No.707, Sec.2, Department of Mechanical Engineering, National Kaohsiung University of Aapplied Sciences415 Chien Ku Graduate Institute of Technology Management, Chung Hua University, No.707, Sec.2, Wu Fu Rd., Hsinchu Department of Computer Science and Information Engineering, Ta Hwa Institute of Technology, 1 Tahwa

会议类型: 国际会议

会议名称: 2008 International Conference on Machine Learning and Cybernetics(2008机器学习与控制论国际会议)

会议地点: 昆明

会议语种:英文

页码: 3315-3321

在线出版日期: 2008-07-12（万方平台首次上网日期，不代表论文的发表时间）