Refinery-Wide Optimization

REFINERY-WIDE OPTIMIZATION WITH RIGOROUS MODELS... 

Refinery- Wide Optimization with Rigorous Models 259... 

The refinery-wide optimization (RWO) initiative began in July 1998. Three phases were defined for the project ... 

Table 1. Scope of Refinery-Wide Optimization Applications...

Refinery-Wide Optimization -with Rigorous Models... 

Mudt, D. R. Pedersen, C. C. Jett, M. D. Karur, S. McIntyre, B. Robinson, P. R. Refinery-wide Optimization with Rigorous Models, In Practical Advances in Petroleum Processing, Hsu, C. S. Robinson, P. R. (Eds.), Springer ... 

Some recent applications have benefited from advances in computing and computational techniques. Steady-state simulation is being used off-line for process analysis, design, and retrofit process simulators can model flow sheets with up to about a million equations by employing nested procedures. Other applications have resulted in great economic benefits these include on-line real-time optimization models for data reconciliation and parameter estimation followed by optimal adjustment of operating conditions. Models of up to 500,000 variables have been used on a refinery-wide basis. 

As discussed in Chapter 22, model-predictive control (MPC) and real-time optimization (RTO) can improve the profitability of a major conversion unit by 5 to 10%. Refinery-wide, benefits from MPC and RTO can exceed US 0.40 per barrel, or US30 million per year for a high-conversion 200,000 b/d refinery. 

Since the early 1980s olefin plants in the United States were designed to have substantial flexibiHty to consume a wide range of feedstocks. Most of the flexibiHty to use various feedstocks is found in plants with associated refineries, where integrated olefins plants can optimize feedstocks using either gas Hquids or heavier refinery streams. Companies whose primary business is the production of ethylene derivatives, such as thermoplastics, tend to use ethane and propane feedstocks which minimize by-product streams and maximize ethylene production for their derivative plants. 

Linear programming can be used to solve very large problems, with thousands of variables and constraints. The method is widely used in operations, particularly in optimization of oil refineries and petrochemical plants. It is used a lot less in design, as design problems almost inevitably contain many nonlinear equations. 

There are still many needs required for improvements to refineries. One is a wide array of sensors, which will help to provide the foundation for real-time optimization of the integrated refinery and help to manage every point in the refinery. To help meet these needs, a key interface must be developed between chemical engineering and applied mathematics. 

Production planning is an important activity in modem refineries. The modem refinery is a combination of many complex units such catalytic reforming, fluid catalytic cracking (FCC), hydroprocessing, etc. While it is possible to tune each unit to an optimal yield, the optimum yield of a particular unit may not reflect a true optimum because of the demands and prices for the wide range of products that the refiner produces. Therefore, it is important to consider each unit in the context of the whole refinery. The activity of choosing feedstock to refinery (and... 

---