A NEW METHOD FOR MONITORING AND TUNING PLASTIC INJECTION MOLDING MACHINES

H.Y. Lau, X. Li, and R. Du

References

  1. [1] W. He, Y.F. Zhang, K.S. Lee, & T.I. Liu, Development of a fuzzy-neuro system for parameter resetting of injection molding, Trans. ASME, Journal of Manufacturing Science and Engineering, 123 (1), 2001, 110–118. doi:10.1115/1.1286732
  2. [2] E.B. Garvey, On-line quality control of injection molding using neural networks, master’s thesis, Royal Melbourne Institute of Technology, Australia, 1997.
  3. [3] B.H.M. Sadeghi, A BP-neural network predictor model for plastic injection molding process, Journal of Materials Processing Technology, 103, 2000, 411–416. doi:10.1016/S0924-0136(00)00498-2
  4. [4] R. Ivester, K. Danai, & D. Kazmer, Automatic tuning of injection molding by the virtual search method, Trans. ASME, Journal of Manufacturing Science and Engineering, 120 (2), 1998, 323–329. doi:10.1115/1.2830130
  5. [5] V. Vapnik, Statistical learning theory (New York: Wiley, 1998).
  6. [6] B. Schölkopf et al., Input space vs. feature space in kernelbased methods, IEEE Trans. on Neural Networks, 10 (5), 1999, 1000–1017. doi:10.1109/72.788641
  7. [7] J.A.K. Suykens & J. Vandewalle, Least squares support vector machine classifiers, Neural Process Letter, 9 (3), 1999, 293–300. doi:10.1023/A:1018628609742
  8. [8] J.A.K. Suykens, J. De Brabanter, L. Lukas, & J. Vandewalle, Weighted least squares support vector machines: robustness and sparse approximation, Neurocomputing, 48, 2002, 85–105. doi:10.1016/S0925-2312(01)00644-0
  9. [9] S.B. Schölkopf, A tutorial on support vector regression: NeuroCOLT, Technical report NC-TR-98-030, Royal Holloway College, University of London, 1998.
  10. [10] M. Orr, Regularisation in the selection of RBF centres, Neural Computation, 7 (3), 1995, 606–623. doi:10.1162/neco.1995.7.3.606
  11. [11] S. Chen, C.F.N. Cowan, & P.M. Grant, Orthogonal least squares learning algorithm for radial basis function networks, IEEE Trans. on Neural Networks, 2 (2), 1991, 302–309. doi:10.1109/72.80341
  12. [12] Y.-H. Cheng & C.-S. Lin, A learning algorithm for radial basis function networks with the capability of adding and pruning neurons, Proc. IEEE Signal Processing, 1994, 797–801.
  13. [13] T. Poggio & F. Girosi, Regularization algorithms for learning that are equivalent to multilayer networks, Science, 247, 1990, 987–982. doi:10.1126/science.247.4945.978
  14. [14] S. Haykin, Neural networks: A comprehensive foundation (New York: Macmillan, 1994).
  15. [15] D. Lowe, Adaptive radial basis function nonlinearities and the problem of generalization, First Int. Conf. on Artificial Neural Networks, London, 1989, 171–175.
  16. [16] M. Kubat, Decision trees can initialize radial-basis function networks, IEEE Trans. on Neural Networks, 9 (5), 1998, 813– 824. doi:10.1109/72.712154
  17. [17] H.Y. Lau, X.L. Li, K. Yeung, & R. Du, Predicting nozzle pressures in plastic injection molding processes using a hybrid radial basis function neural network, 2004 Automation & Assembly Summit Conference, Fort Worth, TX, May 2004. Appendix A Results of the Two Design of Experiments A.1. The DOE for Case 1 Figure A.1. The Pareto chart of |Effect/2| in Case 1. Table A.1 Design Factors and their Levels Factor A B C D Name Nozzle Injection Injection Holding temperature pressure speed pressure High 205◦ C 98 bar 80% 40 bar level (+) Low 195◦ C 80 bar 55% 5 bar level (−) 134 Table A.2 The DOE Results (I: Mean of High Level; II: Mean of Low Level) Run 1 + + + + + + + 35.5675 2 + + + − − − − 36.1950 3 + − − + + − − 27.4925 4 + − − − − + + 26.5225 5 − + − + − + − 37.0425 6 − + − − + − + 37.8625 7 − − + + − − + 30.8825 8 − − + − + + − 28.1400 I 31.4444 36.6669 32.6963 32.7463 32.2656 31.8181 32.7088 II 33.4819 28.2594 32.2300 32.1800 32.6606 33.1081 32.2175 Effect −2.0375 8.4075 0.4663 0.5662 −0.3950 −1.2900 0.4913 Effect/2 −1.0188 4.2038 0.2331 0.2831 −0.1975 −0.6450 0.2456 A.2. The DOE for Case 2 Figure A.2. Pareto chart of |Effect/2| in Case 2. Table A.4 The DOE Results (I: Mean of High Level; II: Mean of Low Level) Variables A B A×B C A×C B×C D Results (yi) Trail 1 + + + + + + + 35.4 2 + + + − − − − 35.5975 3 + − − + + − − 33.4125 4 + − − − − + + 33.26 5 − + − + − + − 35.0075 6 − + − − + − + 34.7775 7 − − + + − − + 33.33 8 − − + − + + − 32.64 I 34.4175 35.1956 34.2419 34.2875 34.0575 34.0769 34.1919 II 33.9387 33.1606 34.1144 34.0688 34.2987 34.2794 34.1644 Effect 0.4787 2.0350 0.1275 0.2187 −0.2412 −0.2025 0.0275 Effect/2 0.2394 1.0175 0.0637 0.1094 −0.1206 −0.1013 0.0137 Table A.3 Design Factors and their Levels Variable A B C D Name Nozzle Injection Injection Holding temperature pressure speed pressure High 202◦ C 90 bar 75% 35 level (+) Low 198◦ C 82 bar 60% 15 level (−) 135

Important Links:

Go Back