Modeling and experimental prediction oil refinery wastewater treatment using activated sludge process

In this study, activated sludge process in refinery was investigated for the treatment of refinery wastewater. In order to investigate the system a laboratory scale rig was built. The effect of several parameters on COD removal efficiency, temperature, residence time, effect of Leca (Percentage fill the reactor by Leca), and UV radiation on percentage of COD removal were examined experimentally. Maximum COD removal efficiency, was obtained 94% after final testing. An artificial neural network (ANN) was applied to evaluate the effect of operational parameters on the COD removal efficiency in the next step of study. A two-layered ANN was provided the best results, using Levenberg–Marquardt back propagation learning algorithm (trainLM), tansig and purelin as transfer functions in hidden and output layers. Furthermore, application of three neurons in the hidden layer caused to gratify network training while over fitting was hindered. ANN model, provided good estimation with correlation coefficient and the mean square error (MSE) of 0.997 and 0.5 × 10-3 respectively.

Keywords:

Wastewater treatment , COD , Active Sludge , Artificial Neural Network