System identification of ethane-ethylene distillation process using linear and Nonlinear ARX approaches

This paper present a comparison study toward process modeling and identification using linear ARX and Nonlinear ARX (NARX) approaches in an attempt to capture the dynamics of a continuous distillation column derived from simulated data of ethane and ethylene separation process obtained in DaIsy database. Each of the resultant modeling were compared and tested to understand which method perform better for the given sets of data using established model verification method. Result shows NARX modeling approach was found to be superior to the linear ARX approach in defining the studied process.