A Mixed SVD-Neural Network Approach to Optimal Control of Ceramic Mould Manufacturing in Lost Wax Cast Processes

Abstract:

We account for the problem of optimal control of ceramic mould manufacturing in lost wax cast processes with the aid of a mixed linear algebraic-statistical approach based on the employment of Singular Value Decomposition (SVD) and Neural Networks (NN).We consider the peculiar aspect of minimizing ceramic inclusions occurrence in equiaxed superalloy turbine components which are manufactured resorting to gravitational pouring. The optimization consists in finding optimal extrema of scalar and/or vectorial functions of the type Rk->Rm i.e. Key Process Variable domain (KPV) vs. Target Variable domain (TV) over a large set of experimental data affected by acquisition noise leading to a typical sparse multiblock array.The goal of the work consists in the assessment of possible significant statistical multivariate correlations amongst the KPV and TV when the dimension of domain space, k, has an order of magnitude of tens, in the presence of quasi-rank deficient input matrix.

• Periodical: Advances in Science and Technology (Volume 87)
• Pages: 105-112
• DOI: https://doi.org/10.4028/www.scientific.net/AST.87.105
• Online since: October 2014
• Authors: Ciro Caramiello, Stefania Iannuzzi, Alessandro Acernese, Doriana D'Addona
• Keywords: Ceramic Inclusion, Gas Turbine Manufacturing, Lost Wax Process, Neural Network (NN), Process Optimization, Subspace Algorithms, Superalloy, SVD

References

[1] G.H. Golub - C.F. Van Loan, Matrix Computations, The Johns Hopkins University Press, Baltimore and London, (1993).

[2] G. W. Stewart and J. G. Sun, Matrix Perturbation Theory, Academic Press, Boston, MA, (1990).

[3] Y. P. Hong, C.T. Pan, Rank-Revealing QR Factorizations and the Singular Value Decomposition, Mathematics of Computation, vol. 58, No. 197 (Jan. 1992), pp.213-232.

DOI: 10.2307/2153029

[4] P. V. Overshee and B. D. Moor, Subspace algorithms for the stochastic identification problem, Automatica, vol. 29, pp.649-660, (1993).

DOI: 10.1016/0005-1098(93)90061-w

[5] M. Viberg, Subspace-based methods for the identification of linear time-invariant systems, Automatica, vol. 31, no. 12, pp.1835-1853, (1995).

DOI: 10.1016/0005-1098(95)00107-5

[6] Douglas C. Montgomery – Introduction to statistical Quality Control, John Wiley and Sons, Inc. (2005).

[7] Thomas Pyzdek, The Six sigma Handbook, Mc Graw Hill, (2003).

[8] Di Foggia, M., D'Addona, D.M., Identification of critical key parameters and their impact to zero-defect manufacturing in the investment casting process, Procedia CIRP, Vol. 12, 2013: 264-269.

DOI: 10.1016/j.procir.2013.09.046

[9] Fahlman, S.E. and Lebiere, C., 1990, An Empirical Study of Learning Speed in Back Propagation Networks, Tech. Rep., CMU-CS-88-162.

[10] D'Addona, D., Teti, R., Caprino, G., 2012, Ultimate Strength Prediction of Artificially Damaged Composite Laminates Based on Neural Networks, Journal of Intelligent & Fuzzy Systems, ISSN: 10641246, 23/5: 217–223.

DOI: 10.3233/ifs-2012-0511

[11] Masters, T., 1993, Practical Neural Network Recipes in C++, Academic Press, San Diego.