Published: CABEQ 26 (3) (2012) 241–248
Paper type: Original Scientific Paper
J. Vivier and A. Mehablia
To improve traditional neural networks, the present research used the wavelet network, a special feedforward neural network with a single hidden layer supported by the wavelet theory. Prediction performance and efficiency of the proposed network were examined with a published experimental dataset of cross-flow membrane filtration. The dataset was divided into two parts: 70 samples for training data and 330 samples for testing data. Various combinations of transmembrane pressure, filtration time, ionic strength and zeta potential were used as inputs of the wavelet network to predict the permeate flux. The initial network led to a wavelet network model after training procedures with fast convergence within 30 epochs. Further, the wavelet network model accurately depicted the positive effects of either transmembrane pressure or zeta potential on permeate flux. Moreover, comparisons indicated the wavelet network model produced better predictability than the back-forward backpropagation neural network and the multiple regression models. Thus the wavelet network approach could be employed successfully in modeling dynamic permeate flux in cross-flow membrane filtration.
This work is licensed under a Creative Commons Attribution 4.0 International License
Artificial Neural Network, membrane, simulation