Prediction of NH4+ concentration during the temperature triggered glutamate fermentation using at-line near-infrared spectroscopy

Abstract

Near-infrared (NIR) spectroscopy, a rapid and nondestructive analytical technique, was applied to predict the NH4+ concentration of the culture broth from temperature triggered glutamate fermentation. NIR data of 164 samples of supernatant were analyzed by partial least squares (PLS) regression with several spectra preprocessing methods. The coefficient of determination (R2), model root mean square error of calibration (RMSEC), and root-mean-square error of prediction (RMSEP) of the test calibration for NH4+ concentration were 0.9839, 4.34 and 4.62 mmol/l, respectively. These results suggested that the model had an accurate predictive capacity for NH4+ concentration. The proposed model provides a potential fast way for control and optimization of glutamate fermentation process. © Springer-Verlag Berlin Heidelberg 2014.