Modeling of membrane bioreactor of wastewater treatment using support vector machine

Abstract

Membrane bioreactor (MBR) is one of the advanced and new efficient reliable technology that replace the conventional activated sludge process in wastewater treatment plant. Therefore, understanding of dynamic behaviour of membrane filtration process is crucial to ensure good estimation of the filtration process. This paper presents the support vector machines (SVM) and artificial neural network to model and predict the membrane fouling. The predicted models are validated using an experimental data from a pilot scale palm oil mill effluent MBR located at Process Control Laboratory, Universiti Teknologi Malaysia. Simulation results showed that SVM able to produce good prediction as neural network model.