Industrial process control and characterization by globalizing methods
Keywords:
Process control, Complex material characterization, Specific instrumentation, Data miningAbstract
We describe in this summary basic elements and applications of a concept of industrial process control and characterization of complex materials. This idea of "globalizing control" is presented as an alternative or complement to traditional methods that are defined by the acquisition of multiple individual parameters and the juxtaposition of partial results. Moreover it is usual to search for a fundamental understanding of the physicochemical conditions, even through numerical modeling. Given the complexity of some industrial materials, in our method, we decided to give up this ultimate goal, to get effective control by means of indirect ways.
Downloads
References
[2] S. Poggi, C. Ribeiro – Réseaux de neurones, machines à support et cartes topologiques – ALCTRA séminaire École supérieure d'électronique de l'Ouest 40pp.
[3] M. Norgaard, O. Ravn, N.K. Poulsen, L.K. Hansen – Neural Networks for Modelling and Control of Dynamic Systems. Springer – Verlag London. 2005.
[4] M. Tenenhaus - La regression PLS. Ed. Technip Paris. 2000
[5] L. Eriksson, E. Johansson, N. Kettaneh – Wold, J. Trigg, C. Wikstrom, S. Wold – Multi and Megavariate data Analysis. Umetrix Academy. Umea – Sweden. 2006.
[6] I. Berechet, G. Alcuri - Miniaturized refractometric sensor for processing control-Proceedings Société Française d’Optique 10pp. 2002.