TY - CONF A1 - Krämer, D. A1 - Violet, N. A1 - Paul, Andrea A1 - Maiwald, Michael A1 - King, R. T1 - Robustification of partial least squares predictions based on near infrared spectroscopy trough nonlinear state estimation N2 - The Cultivation of “Saccharomyces cerevisiae” for enzyme production was monitored using Near-infrared spectroscopy. An inline NIR optrode was therefore immersed in a 15 L vessel. The calibration was done using a Partial Least Squares (PLS) model with reference measurements of glucose, ammonium, phosphate, ethanol, and optical density. A nonlinear biological process model based on an extended Kalman Filter (EKF) was used to describe the fermentation behavior. It was found that EKF corrects inaccurate PLS predictions. T2 - 11. Kolloquium Arbeitskreis Prozessanalytik CY - Wien, Austria DA - 30.11.2015 KW - PLS-R KW - Near infrared spectroscopy KW - Prozess-Spektroskopie KW - Process analytical technology KW - Fermentation PY - 2015 SP - 77 EP - 77 PB - Bundesanstalt für Materialforschung und -prüfung (BAM) CY - Berlin AN - OPUS4-38844 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER -