(185a) Advanced Process Modelling for Refinery-Petrochemical Integration

Oil refineries focus primarily on producing transport fuels. Feedstocks that flow to petrochemical plants from refineries tend to be consequential. There is now an accelerating trend to reduced dependency on transport fuels and this coincides with an increased demand for petrochemical derivatives. As a result, there is a growing effort to integrate refineries and petrochemical plants, so that refinery products are more aligned with usable petrochemical plant feedstocks and overall economics and environmental performance improve.

There are integration opportunities available today with minimal change to current refinery operating conditions. Refinery streams such as fuel gas, LPG, FCC light ends and naphtha can be fed to petrochemical plants and petrochemical plant streams such as hydrogen and C4s can be fed back to refineries. Energy, in terms of steam and electricity, can also be easily moved between the two, providing significant financial benefit.

The traditional route to generating lighter components from crude oil in a refinery is by feeding VGO or Vacuum/Atmospheric Resid to an FCC unit to generate olefins and aromatics. Extending this technology, crude oil can be cracked directly or the crude can be distilled first and the products cracked via steam or catalytic methods, or even both.

Refinery/petrochemical integration tends to increase the overall capital investment but the improved generation of higher-value products and the reduction in previously ‘stranded’ products makes integration financially very attractive.

When assessing new design or revamp, one significant challenge is how to adequately evaluate the costs and environmental benefits of refinery/petrochemical integration. The standard approach is to use Linear Programming models with simplified plant feed/product/cost models. However, this approach is fundamentally flawed because although it accounts for flexibility in stream routing, it does not take account of reactor catalyst performance in the units and the wide range of operating conditions across the refinery that are possible. A suitable overall process model is needed where stream routing, catalyst performance and operating conditions can all be simultaneously manipulated to achieve maximum profit. The associated model complexity is high yet it is essential that this be considered, otherwise significant potential benefit may be left unrealized. It could also mean that a profitable project is stopped due to an inappropriately chosen set of operating parameters.

This paper shows a new approach, where advanced process modelling techniques are used to tackle such complex problems and deliver insights to design parameters and stream allocation. The approach combines equation-based models with advanced model-pruning and equation solving techniques. These are used to determine optimum stream routing, plant configurations and operating conditions. A case study shows the benefits of the new approach. As a result, new or revamp refinery/petrochemical integrated designs are generated which fully exploit the operating freedoms available and deliver maximum profit.