A systematic optimization approach in the MISO Plastic Injection molding process

In this research, Taguchi method, back-propagation neural networks (BPNN), and genetic algorithms (GA) are applied to the problem of process parameter settings for multiple-input single-output (MISO) plastic injection molding. Taguchi method is adopted to arrange the number of experimental runs. Injection time, velocity pressure switch position, packing pressure, and injection velocity are engaged as process control parameters, and product weight as the target quality. Experimental data from Taguchi method are used to train and test BPNN. Engineering optimization concepts are employed to establish the fitness function for using in GA. Then, BPNN and GA are applied for searching the final optimal parameter settings. Two confirmation experiments are performed to verify the effectiveness of the proposed approach. Experimental results reveal that the proposed approach not only can avoid shortcomings inherent in the commonly used Taguchi method but also can result in significant quality and cost advantages.

[1]  Chao-Ton Su,et al.  Optimization of parameter design: An intelligent approach using neural network and simulated annealing , 2000, Int. J. Syst. Sci..

[2]  Hung-Chang Liao,et al.  A data envelopment analysis method for optimizing multi-response problem with censored data in the Taguchi method , 2004, Comput. Ind. Eng..

[3]  Kang Tai,et al.  An evolutionary approach for cooling system optimization in plastic injection moulding , 2004 .

[4]  B.H.M. Sadeghi,et al.  A BP-neural network predictor model for plastic injection molding process , 2000 .

[5]  Wen-Chin Chen,et al.  A neural-network approach for an automatic LED inspection system , 2007, Expert Syst. Appl..

[6]  Wen-Chin Chen,et al.  Process parameter optimization for MIMO plastic injection molding via soft computing , 2009, Expert Syst. Appl..

[7]  Jasbir S. Arora,et al.  Introduction to Optimum Design , 1988 .

[8]  Tsann-Tay Tang,et al.  Parameter optimization in melt spinning by neural networks and genetic algorithms , 2006 .

[9]  Bong-Jin Yum,et al.  Robust design of multilayer feedforward neural networks: an experimental approach , 2004, Eng. Appl. Artif. Intell..

[10]  Kok Kiong Tan,et al.  Neural-network-based predictive learning control of ram velocity in injection molding , 2004, IEEE Transactions on Systems, Man, and Cybernetics, Part C (Applications and Reviews).

[11]  Prasad K. Yarlagadda,et al.  Development of an integrated neural network system for prediction of process parameters in metal injection moulding , 2002 .

[12]  Furong Gao,et al.  Injection molding product weight: Online prediction and control based on a nonlinear principal component regression model , 2006 .

[13]  W. I. Patterson,et al.  Control of part weight in injection molding of amorphous thermoplastics , 1999 .

[14]  Henry C. W. Lau,et al.  Neural networks for the dimensional control of molded parts based on a reverse process model , 2001 .

[15]  Hasan Kurtaran,et al.  Efficient warpage optimization of thin shell plastic parts using response surface methodology and genetic algorithm , 2006 .

[16]  Holger R. Maier,et al.  Understanding the behaviour and optimising the performance of back-propagation neural networks: an empirical study , 1998 .