Development of Expert System for Biological Wastewater Treatment Using Membrane Separation

Membrane fauling process in membrane bioreactors was modeled and an expert system for the maintenance of membrane bioreactors was developed. The developed model considered production and decomposition of membrane fouling substances (extracellular polymeric substances), which were divided into high molecular, middle-molecular and-low molecular parts. Accumulated high-molecular weight EPS on membrane surface were found to be decomposed to low-molecular weight EPS resulting in the decrease of trans-membrane pressure. Mechanism of membrane cleaning by air-bubbling in MBR reactors was discussed using shear-stress sensors. Existence of bubbles and particles in reactors were found to cause strong shear stress on membrane surface. Time-series data of shear stress were analyzed and a model was developed to relate cleaning efficiency of membrane surface and shear stress fluctuation. A mathematical model was developed to predict trans-membrane pressure considering the production, accumulation, decomposition and detachment of EPS and model parameter were successfully identified using Kalman Filter. Prediction of membrane fouling using the developed model and identified parameters was successful as the simulated results were in good agreement with the measured results.