Browsing by Author "Marchetti, Jacinto"

Filter results by typing the first few letters
Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    A fault detection and diagnosis technique for multivariate processes using a PLS-decomposition of the measurement space
    (2013) Vega, Jorge Ruben; Godoy, José Luis; Marchetti, Jacinto
    A newstatisticalmonitoring technique based on partial least squares (PLS) is proposed for fault detection and di- 24 agnosis inmultivariate processes that exhibit collinearmeasurements. A typical PLS regression (PLSR)modeling 25 strategy is first extended by adding the projections of the model outputs to the latent space. Then, a PLS- 26 decomposition of the measurements into four terms that belongs to four different subspaces is derived. In 27 Q2 order to online monitor the PLS-projections in each subspace, new specific statistics with non-overlapped do- 28 mains are combined into a single index able to detect process anomalies. To reach a complete diagnosis, a further 29 decomposition of each statistic was defined as a sum of variable contributions. By adequately processing all this 30 information, the technique is able to: i) detect an anomaly through a single combined index, ii) diagnose the 31 anomaly class from the observed pattern of the four component statistics with respect to their respective confi- 32 dence intervals, and iii) identify the disturbed variables based on the analysis of themain variable contributions 33 to each of the four subspaces. The effectiveness observed in the simulated examples suggests the potential appli- 34 cation of this technique to real production systems.
  • Thumbnail Image
    Item
    New contributions to non linear process monitoring through kernel partial least squares
    (2013) Vega, Jorge Ruben; Godoy, José Luis; Marchetti, Jacinto; Zumoffen, David
    The kernel partial least squares (KPLS) method was originally focused on soft-sensor calibration for predicting online quality attributes. In this work, an analysis of the KPLS-based modeling technique and its application to nonlinear process monitoring are presented. To this effect, the measurement decomposition, the development of new specific statistics acting on non-overlapped domains, and the contribution analysis are addressed for purposes of fault detection, diagnosis, and prediction risk assessment. Some practical insights for synthesizing the models are also given, which are related to an appropriate order selection and the adoption of the kernel function parameter. A proper combination of scaled statistics allows the definition of an efficient detection index for monitoring a nonlinear process. The effectiveness of the proposed methods is confirmed by using simulation examples. Keywords: KPLS Modeling, Fault Detection, Fault Diagnosis, Prediction Risk Assessment, Nonlinear Processes.