The Effect of Input sequence to nonlinear artificial neural network (NN) performance in modeling CSTR

• Main Authors: F. D., Somanti, S., Yusup, H., Zabiri
• Format: Conference or Workshop Item
• Published: 2009
• Subjects: TP Chemical technology
• Online Access: http://eprints.utp.edu.my/10752/1/fariha1.pdf; http://eprints.utp.edu.my/10752/2/the_effect_of_input.pdf; http://eprints.utp.edu.my/10752/
• Institution: Universiti Teknologi Petronas

This paper considers the aspects of the system identification of nonlinear black box empirical models for chemical process dynamics. The core of this research is to study the effect of existing input sequence to nonlinear Artificial Neural Network applied in nonlinear dynamic system. To illustrate the practical utilization of the various types of input sequences used, NARXSP dynamic Neural Network model is applied to approximate the dynamics of a first-principles model of first order kinetic reaction in a simple Continued Stirred Tank Reactor.