Development of electronic nose and near infrared spectroscopy analysis techniques to monitor the critical time in SSF process of feed protein.

In order to assure the consistency of the final product quality, a fast and effective process monitoring is a growing need in solid state fermentation (SSF) industry. This work investigated the potential of non-invasive techniques combined with the chemometrics method, to monitor time-related changes that occur during SSF process of feed protein. Four fermentation trials conducted were monitored by an electronic nose device and a near infrared spectroscopy (NIRS) spectrometer. Firstly, principal component analysis (PCA) and independent component analysis (ICA) were respectively applied to the feature extraction and information fusion. Then, the BP_AdaBoost algorithm was used to develop the fused model for monitoring of the critical time in SSF process of feed protein. Experimental results showed that the identified results of the fusion model are much better than those of the single technique model both in the training and validation sets, and the complexity of the fusion model was also less than that of the single technique model. The overall results demonstrate that it has a high potential in online monitoring of the critical moment in SSF process by use of integrating electronic nose and NIRS techniques, and data fusion from multi-technique could significantly improve the monitoring performance of SSF process.