Referencia completa:
N. M. Fuentes-García, J. M. González-Martínez, G. Maciá-Fernández and J. Camacho, "PARAMO: Enhanced Data Pre-processing in Batch Multivariate Statistical Process Control", Oral Presentation in the Scandinavian Symposium on Chemometrics (SSC16), 2019.
Abstract:
Since the pioneering works by Nomikos and MacGregor [1], the Batch Multivariate Statistical Process Control (BMSPC) methodology has been extensively revised and a sheer number of alternative monitoring approaches have been suggested. The different approaches vary in the batch data alignment, the pre-processing approach, the data arrangement and/or the type of model used, from two-way to three-way and from linear to non-linear [2]. One of the most accepted pre-processing schemes, referred to as Trajectory Centering and Scaling (TCS), is based on the normalization to zero mean and unit variance around the average trajectory [1]. However, the main drawback of TCS is the inherent increase of the level of uncertainty in the estimation of model parameters [2]. In this work, two main open questions are addressed: i) can the estimation of pre-processing parameters be improved, thereby reducing the parameter instability in the bilinear modeling of batch data? and ii) does the parameter stability have a statistical significant effect on fault detection?
We illustrate how to improve parameter estimation whilst maintaining the good properties of
TCS. We propose a new pre-processing approach, PARAMO (PARAmeters from More Observations), which uses more observations than TCS to estimate the pre-processing parameters. We assess PARAMO and TCS by using the data generated from the Saccharomyces Cerevisiae cultivation process [3-4]. PARAMO outperforms the established methodology for pre-processing batch data in BMSPC. Using this proposal, both the parameter stability and the monitoring performance are improved. The results of this research work affect a large amount of the monitoring approaches proposed to date, and we advocate that the pre-processing procedure proposed here should be generally applied in BMSPC.