Predicting Nanofiltration Performance during Treatment of Welding Electrode Manufacturing Wastewater, Using Artificial Neural Networks
This paper presents artificial neural network (ANN) predictions for a nanofiltration membrane used to treat wastewater of welding electrode manufacturing in a cross flow set up. The main parameters were time, feed flow rate, and transmembrane pressure (TMP). The experimental data were correlated and analyzed using ANN. ANN’s prediction of the permeate flux, turbidity, total dissolved solids (TDS), hardness for various TMPs, and flow rates are discussed. The effects of the training algorithm, neural network architectures, and transfer function on the ANN performance, as reflected by the percentage average absolute deviation, are discussed. A network with one input layer, 50-100 hidden layers, and one output layer is found to be adequate for mapping input–output relationships and providing a good interpolative tool. A good agreement has been obtained between the ANN predictions and the experimental data with a deviation below 2% for all cases considered.