The main purpose of splitter tower process of pentane unit is to separate pentane gas from light gasoline as a valuable and value-added product. Having knowledge of the product quality is necessary to help proper process control, more production and sale and ultimately more profit margins. For this purpose the process must be identified. In this study the Splitter Tower is identified.
In this paper, in addition to studying different identification methods and studying how the pentane unit works, the new model based identification and estimation methods are implemented. These methods include implementation of intelligent control technology based on soft computing techniques such as neural networks. Using these model based methods, a comparison between two different methods is carried out and the best one is chosen. To build a system model the identification using nonlinear intelligent methods such as MLP and RBF neural networks have been examined. Finally, the best one with the least mean square method is introduced.