Inferential sensor – Based adaptive principal componentsanalysis for mechanical properties prediction and evaluation

Type : Publication
Auteur(s) :  Bouhouche Salah, Laib dit Leksir Yazid, Hazem Tarek, Bast Jurgen
Année :  2013
Domaine : Automatique
Revue : measurement
Résumé en PDF :  (résumé en pdf)
Fulltext en PDF :  (.pdf)
Mots clés :  Soft sensor, Mechanical testing, Adaptive principal component analysis, Uncertainties evaluation

Résumé : 

This paper is concerned with a method for on-line quality prediction and evaluation ofmechanical properties in metal testing. This method uses an Adaptive Principal ComponentAnalysis (APCA) as a multi predictor of different sub-models defining the mechanical prop-erties such as constraints limits and elongation. The PCA technique, characterized by itsmultivariate component, is strongly recommended to model a multi-input-multi-outputsystem. The complex system is generally known as a non-linear and unsteady state pro-cess. The PCA method is a linear projection. To adapt it and to improve the prediction accu-racy, a variant of this method is considered based on iteratively using a specific algorithm.This kind of approaches is applied for constructing an inferential model, which allows areliable and accurate predictor. Simulation results, based on the measured and computeddata using the above-cited method, show that the proposed approach is easily implement-able and give an accurate prediction.