Refinery crude units often process variety of feedstocks. If properties of the feed entering the crude unit are not precisely known, it is difficult to predict how changes in the tower operating conditions will affect properties of the products. Similarly, if on-line analyzers are not available, it is not possible to implement very accurate control of the product properties.
This work introduces a new class of models (hybrid models) which can be used to: (i) identify feed distillation curve, (ii) predict product properties from tower operating conditions, and (iii) predict product properties from feed properties and material and energy balances. Hybrid models consist of first principles equations (e.g. balances) and of Partial Least Squares (PLS) models for prediction of product or feed properties. A model of a crude unit has several hundred equations and is practically linear; this model predicts target properties within 1% of the rigorous tray to tray model which consists of 10k-20k nonlinear equations.
In industrial practice there is a very limited number of tray temperature measurements; the models we have developed enable us to achieve the accuracy of prediction that is the same as if there are tray temperature measurements available on every tray. In addition, inclusion of material and energy balances in the hybrid model equations enables optimization of tower operating conditions based on the hybrid model.
We present results of a) optimization of tower operating conditions based on the hybrid model and based on the equation-oriented AspenPlus model; b) identification of the crude feed distillation curve based on the limited number of tray temepratures and the hybrid model variables and compare it to predictions one would obtain if there were tray temperatures available on every tray; c) prediction of product properties from the available tray temperatures and hybrid model variables.
Accuracy of the hybrid model predictions and almost linear nature of its equations makes it suitable for applications in on-line monitoring, or as inferential predictor for model-based controllers, or for optimization of operating conditions, or for crude unit modeling in the refienry plannign and scheduling tools.
Would you like to access this content?
No problem. You just have to complete the following steps.
You have completed 0 of 2 steps.
You must be logged in to view this content. Log in now.
Purchase Technical Presentation
You must purchase this technical presentation using one of the options below.
If you already purchased this content recently, please click here to refresh the system's record of ownerships.
|Credits||0.5 Use credits|
|List Price||$25.00 Buy now|
|AIChE Members||$15.00 Buy now|
|AIChE Fuels and Petrochemicals Division Members||Free Free access|
|AIChE Undergraduate Student Members||Free Free access|
|AIChE Graduate Student Members||Free Free access|