— Adsorption experiments were carried out in fixed bed column. Neural network (NN) was used to describe the fixed bed adsorption of POME pigment by resin. The general breakthrough models such as Thomas and Yoon–Nelson models resulted in poor fitness with experimental data (R2
< 0.8). A wavelet neural network model (WNN) was developed to model the breakthrough curves in fixed bed column for multicomponent system and WNN model successfully described the adsorption process (R2
= 1). At the initial stages, BDST model showed good agreement with the experimental data but diverged at higher Cb/C0
ratio (> 0.11).
— Neural network, fixed bed column, adsorption, modeling.
M. M. Nourouzi and Luqman Chuah Abdullah are with the Department of Chemical and Environmental Engineering, Faculty of Engineering, Universiti PutraMalaysia, 43300 UPM Serdang, Selangor, Malaysia (email: email@example.com).
S. Keshani is with Fuel Cell Institute, Universiti Kebangsaan Malaysia, 43600 UKM Bangi, Selangor, Malaysia.
Cite: M. M. Nourouzi, S. Keshani, and L. Chuah Abdullah, " Neural Network Application in Fixed Bed Column Adsorption," International Journal of Innovation, Management and Technology vol. 5, no. 2, pp. 130-133, 2014.