Modeling at Molecular Level in Petrochemical Planning Optimization
This contribution proposes a method to introduce the concept of molecular management into production planning optimization models to achieve molecular-level modeling of petrochemical processes.The conventional production planning optimization models used in China’s some petrochemical enterprises are generally too macroscopic to accurately reflect the impacts of feedstock composition changes on product yields and properties.Therefore,it is necessary to introduce molecular-level modeling in optimization models.Herein,an approach to molecular modeling in overall refinery optimization models was proposed:First,secondary reaction units are constructed as multiple logical devices.Thereafter,multi-delta-base structures are used to describe the correlation between product and feedstock compositions at molecular level.Finally,the scheme-relevant property transfer technology was employed to synthesize molecular feeds into actual products,thus dynamically reflect the molecule compositions in each product feed.This method is implemented on RIPO,the self-developed planning optimization platform.Using data from a certain plant,an overall production planning optimization model involving molecular models of atmospheric distillation units,continuous catalytic reforming units and aromatics complex was established.Results displayed this method can achieve sophisticated molecular-level modeling and optimization.The optimization results revealed the impacts of different crude oil compositions and processing schemes on target product yields and properties,providing support for improving refining economics.
NETL
NSTL
万方数据
