Validation of a simple fouling model for submerged membrane bioreactor

International audienceMost of the published membrane bioreactor (MBR) models have been proposed for process description and cognition, resulting in a large number of parameters to estimate from experimental data. These models are usually too complex for process control, and there is a need for simple, dedicated, dynamic models. In this study, attention is focused on the fouling phenomenon, which hampers the efficient operation of MBRs, and a simple model is proposed and validated using a large data base collected from a pilot plant. This model includes several manipulated variables, e.g., the effluent flow rate and the air crossflow, as well as a measurable perturbation such as the ambient temperature. To ease the identification procedure, a separation of the time scales of the process in slow and fast dynamics is exploited. The results show that the model can be used to predict the trans-membrane pressure behavior in a medium-term, about 10 days ahead