Comparative methodology of non-linear models for predicting rheological properties of rubber mixtures in industrial lines
- Urraca Valle, R.
- Sodupe Ortega, E.
- Antoñanzas Torres, J.
- Alonso García, E. 1
- Sanz García, A. 1
- Martínez de Pisón Ascacíbar, F. J.
-
1
Universidad de La Rioja
info
Argitaletxea: Asociación Española de Ingeniería de Proyectos (AEIPRO)
ISBN: 978-84-616-6454-2
Argitalpen urtea: 2013
Orrialdeak: 1346-1357
Biltzarra: CIDIP. Congreso Internacional de Ingeniería de Proyectos (17. 2013. Logroño)
Mota: Biltzar ekarpena
Laburpena
Data mining and statistics are applied to predict certain of the properties of rubber-extruded mixtures. These properties are associated to their cure curves using datafrom the mixing phase at the beginning of the process. The main goal is toautomatically obtain the model that provides operators accurate set points to controlmixing process. The operators would be able to anticipate possible failures in thequality of vulcanized rubber mixture.There are several strategies to develop optimum models. This work proposes thefollowing methodology to optimize the information extraction from the available data.First, an initial analysis of database attributes is performed seeking for significantinformation to future model derivation. Second, a wide comparison of different non-parametric methods is carried out to determine which one is the most appropriate.Instead of directly contrasting prediction errors, an automatic statistical system ofcomparison is included by using several non-parametric techniques. Third, somealternative strategies are tested taking advantage of the specific attributes of thedatabase.