Multilayer perceptron modeling for UASB Reactor treating tannery effluent

In this paper, the use of multilayer perceptron neural network for modeling is investigated for a bench scale system of upflow anaerobic sludge blanket (UASB) reactor. The system study is the anaerobic digestion of synthetic wastewater derived from the starch processing industries. The experiment is carried out in a bench scale Upflow Anaerobic Sludge Blanket reactor. It is proven that multilayer perceptron modeling has great adaptability to the variations of system configuration and operation condition. Multilayer perceptron neural network is trained with the experimental values obtained. The output parameters predicted for respective inputs have been found to be very much closer to the corresponding experimental ones and the model was validated by replicative testing.