Abstract
Abstract The development of modelbased process software sensors for monitoring of biomass concentration and product concentration in fed-batch and continuous yeast fermentations is presented, followed by a validation of the sensors using data from industrial fermentations. Alternatively, using multiway projection to latent structures (MPLS) algorithm, a model for prediction of one-step ahead and end point product. concentrations is developed and demonstrated on industrial process data. The one-step ahead MPLS-predictor is compared to the modelbased product. concentration software sensor. The comprison indicates a better performance by the MPLS-predictor.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have