MODERN FUEL BLENDING-1 BLENDING TECHNOLOGY KEY TO MAKING NEW GASOLINES

Yalcin Serpemen, Fritz W. Wenzel
Veba Oel Technologie GmbH
Gelsenkirchen, Germany

Alfred Hubel
Veba Oel AG
Gelsenkirchen, Germany

Advanced gasoline blending technology employed at one of Europe's most complex refineries can help meet the major and economic demands of making reformulated gasoline.

The U.S. refining industry is facing tough requirements, both in time and cost, to comply with the Clean Air Act of 1990. Much has already been written on the various aspects of oxygenates, new processing units, the disposition of butanes and aromatics, and the logistical problems associated with the new fuel.

In short, U.S. refiners face major cost increases in making transportation fuels to meet air quality requirements.

A system that can help meet this challenge at a relatively low investment has been in operation for gasoline blending since September 1987 in Germany at Ruhr Oel GmbH's 215,400 b/cd Gelsenkirchen-Horst refinery. Ruhr Oel is jointly owned by Veba Oel AG and Petroleos de Venezuela.

As Fig. 1 shows, the refinery produces, in addition to petrochemicals, a complete line of transportation fuels.

Typical production volumes are: unleaded regular, 11,400 b/d; unleaded premium (Eurosuper), 9,500 b/d; leaded premium, 9,000 b/d; unleaded "premium plus," 3,700 b/d; diesel, 20,500 b/d; and light heating fuel oil, 30,500 b/d. Jet fuel is also produced.

Blending systems are employed for both gasoline and diesel production. The diesel blending system will be covered in a subsequent article in Oil & Gas Journal.

BLEND OBJECTIVES

One way to enhance refining benefits is to improve blending operations. Blending of transportation fuels with a great number of single components differing in their properties is a classical optimization problem, because of the quality specifications to be met and the goal to keep the costs for finished products at a minimum.

The most modern, reliable, and economic way to solve this problem is the application of an advanced on-line blend optimization technology, consisting of different integrated hardware and software components. This permits LP-based optimization, multivariable feedback quality control, and supervisory blend control.

Such a system configuration meets the most important blend objectives:

• Minimization of quality giveaways and reblends

• Production of the most valuable products at lowest possible costs by minimizing the use of expensive high-value components, such as reformate or purchased components (oxygenates)

• Accurate quality control with use of on-line analyzers

• Increased flexibility of the blending operation

• Reduction of inventory of components and finished products.

The optimal use of high-value blend components will become increasingly important in gasoline blending for closing the octane gap caused by lead phaseout and limitations on aromatics and volatility, which call for the use of expensive oxygenates.

In the cases of diesel and light heating fuel oil, the future reduction in aromatics and sulfur content influences the chemical composition of the pool components. The most likely effects will be on cold filter plugging point (CFPP) and cloud point additives.

BLEND SYSTEM

The main features of the blending system are highlighted by using the Ruhr Oel Gelsenkirchen installation for gasoline blending as an example (Fig. 1).

Because of the complex structure of this refinery with integrated petrochemical production, the gasoline pool consists of a great number of components with very different properties. They are blended into four different grades of gasoline. Up to 16 single components, as listed in Table 1, including oxygenates and tetraethyl lead (TEL), are involved in each blend.

The system configuration, as schematically shown in Fig. 2, consists of:

• Component tanks (static, once-through, and bypass modes possible) and transfer lines with flowmeters and control valves

• A blend header with a special integral in-line mixer providing a homogeneous blend of involved components

• A fast sample loop with on-line analyzers

• A process computer as a central supervising unit, collecting all data (including product tank heel volume and quality), calculating the optimal recipe, and downloading it directly to the ratio controller

• A ratio controller (in-line blender) controlling and transmitting the volumetric throughputs of the components during the blend.

This configuration of the blending unit allows for blend control in a closed-loop mode.

The blending optimization and supervisory system applied by the process computer is using on-line and real time multivariable LP-based, profit-driven optimization techniques. The system communicates by interfacing with the ratio controller, the online analyzers, and an automatic tank gauging (ATG) system.

It provides actual information on components and finished product tanks during the blend.

The blend operation and control is performed by means of the "operators interface," providing the shift personnel with all necessary functions (blend setup, start, stop, suspend, etc.) and information (displays and status and trend reports), including alarms during the execution of the blend. This is incorporated in a control room.

The functions and elements of the blend optimization and control system are depicted in Fig. 3.

During the blending operation, the on-line analyzer measurements are used, after the completion of a validity check, to update the blending model periodically. A quality integration of the blended product (in the product tank) is performed using on-line analyzer readings and taking into account heel volume and qualities.

Within the given constraints of, for example, equipment, component availability, and final product qualities, the LP-optimizer ensures the most economic usage of the given components. It is a price-driven objective function.

The result of the calculation is then directly downloaded to the component ratio controller.

In case of infeasible operating conditions, the operator is guided toward a feasible solution.

The function of the blend quality control algorithm is demonstrated in Fig. 4, which shows the progress of research and motor octane numbers during the batch-time for a regular gasoline blend. The accumulated quality of the blended product in the product tank, including the volume and quality of the tank heel, is calculated every 6 min using the on-line RON and MON measurements, shown as green curves in Fig. 4. These curves illustrate the dynamic response of the system.

The quality set points calculated by the system for achieving the required value in the tank are shown as red curves.

Because the on-line analyzers in question need different lengths of time to generate data points, synchronization of the analyzer readings has to be performed. Therefore, the instrument with the longest response time dictates both the cycle time for the LP-model update and the setting of new ratio values at the blend header,

As a result, the calculated red quality set point curve has a more graduated shape, as compared to a more continuous shape for the accumulated quality curve.

From the inspection of the plots, it can readily be seen that the accumulated quality follows the set points.

During the first half of the blend, the control action is performed by small steps. After reaching the required "on spec point" (in this case 60% of the batch volume), the system takes bigger step changes in set points in order to bring the accumulated quality to the target within the next blend cycle. The blend quality control function can be modified by changing tuning factors.

On-line analyzers for measuring density, RON, and MON, Reid vapor pressure (Rvp), and the distillation curve of the product at the blend header are used within this application, allowing for the feedback control of eight blend qualities in total. These qualities are listed in Table 2.

TEL content is not measured by means of an analyzer, but controlled by volume proportional addition to the blend.

To allow for maximum TEL utilization within a blend, a weighing system with high accuracy is employed.

The on-line process analyzers embedded in a properly designed fast sample loop are essential for achieving a high blend optimization performance. On-line analyzers, such as the knock engines for octane measurement (Figs. 5 and 6), represent a significant portion of the investment of an overall blending modernization project. It is obvious that this investment can only be justified if two main requirements are fulfilled during the blending operation.

They are:

• The on stream factor, or the availability of the analyzer must be high, providing blend control capability by means of blend quality corrections.

• Reliability must be high at the same time, allowing for a blend quality control as close as possible toward the target, remarkably reducing quality giveaways.

A fast sample loop design, accounting for proper sample conditioning for the integrated on-line analyzers, along with calibration facilities and correct handling of the reference fuel, can contribute significantly to achieving the just mentioned requirements for the analyzers.

The on stream factors achieved, in percentage of the individual batch time of single blends (volume-based average values), for MON and Rvp on-line analyzers are 95 and 97%, respectively.

This provides a good basis for a successful blend quality control.

Fig. 7 shows the parity plot for the key quality MON, where the accumulated quality (system prediction) is compared with the laboratory analysis. A very good agreement is observed, indicating precise operation of the MON analyzer.

In order to judge the significance of the results shown in this plot, one has to take into account that the calculation of the accumulated tank quality is based on a large number of MON measurements (analyzer frequency 3.5 min), compared with a single sample analyzed in the laboratory.

ECONOMIC RESULTS

The expected economic success using such a blend optimization system, as implemented by Ruhr Oel GmbH, strongly depends on the extent of the realization of the objectives mentioned earlier.

The possible reduction of quality giveaways usually justifies the application of the advanced on-line blend optimization system.

The related cost savings resulting from the reduction of octane and Rvp give-away and the use of the maximum allowed amount of lead components in gasoline (where permissible) can easily be calculated.

This has been the basis to justify the gasoline blending project for Ruhr Oel's Gelsenkirchen refinery.

However, additional credit potentials resulting from the elimination of reblends, the optimal use of high-cost, high-value blend components, and a possible reduction in laboratory work (certification) are to be considered. The quantification of the individual benefits, however, is rather difficult.

On the other hand, the quality give-away and the extent of required reblend operations and laboratory effort are strongly correlated. Usually, quality give-away results from limitations on the time available for performing reblends for correcting either off-spec qualities or quality giveaways.

BLENDING IMPROVEMENTS

In order to demonstrate the achieved improvements in reducing the quality give-away, plots showing two key qualities- MON and Rvp-are depicted for two different gasoline grades.

These plots compare the former operation in 1986 with the most recent case using the on-line blend optimization system.

Because only a small amount of unleaded regular gasoline was produced in 1986, the comparison has been made between regular leaded gasoline in 1986 and regular unleaded with blend optimization.

MON and Rvp figures, as a result of the first blend, are given in Fig. 8 for regular gasoline (1986 operation). The required specs are also indicated as dotted lines. For both qualities, large deviations from the specs are evident, where a great number of batches are obviously off-spec.

The resulting quality figures after at least one reblend (characterized as last blend) are shown in Fig. 9. By evaluating this plot, it clearly can be demonstrated that the off-spec qualities are corrected to a large extent. Nevertheless, the resulting quality giveaways for MON and Rvp are still significant.

Similar plots for premium leaded gasoline are given in Fig. 10 (first blend) and Fig. 11 (last reblend). As can be seen from the MON values, the data scattering is not as pronounced as compared to the regular gasoline case. The resulting MON values after the last blend are very close to the spec, resulting in only a marginal octane giveaway.

For Rvp, the picture is more or less the same as for regular gasoline, with considerable scatter after the first blend and remaining large giveaways after the last blend.

This clearly indicates that it was not possible to obtain satisfactory results for more than one quality at the same time.

Appreciably better performance of the blend operations with the on-line blend optimization system is shown in Figs. 12 and 13.

The system is obviously able to control all qualities simultaneously, resulting in drastically reduced quality give-aways.

Furthermore, the optimization goals are obtained without any reblend. Thus, a more adequate approach to assess this powerful tool would be a comparison with the results from the first blend of the former operation, where one reblend was necessary for 50% of the cases, and an additional one for 17% of the blends.

Another important aspect of the application of an on-line blend optimization system is the impact of the optimum utilization of the individual blend components. This puts an economic target function on the average blend composition.

The on-line LP-optimizer can help reduce to a minimum the use of high-value, high-cost components, and simultaneously maximize the use of low-cost components.

COMPONENT SAVINGS

This effect is demonstrated in Fig. 14, where the change in average blend composition by application of the on-line blend optimization system is plotted for all blend components, as a difference from the former operation. The year 1986 is the reference case (base line).

The biggest gain, in terms of high-value component savings, is observed for the reformate, which is the most valuable high-cost base blend component in the pool. The reformate content in the gasoline could be significantly reduced.

This shift contributes appreciably to the overall economic improvement. Because of the existing structure of the refinery, with integrated petrochemicals production, the saving of reformate in the gasoline pool can be quantitatively utilized for the production of BTX aromatics.

By-products from the aromatics extraction-C9+ aromatics and Aromex light raffinate (Arlraf)-are used as blend components in the gasoline pool. In line with the above observation, the contents of these low-value components in gasoline, especially Arlraf, are increasingly beneficial.

Because of the realized maximization of Rvp, significantly more butane is now used in the gasoline pool. The system encouraged a higher utilization of butane, as one would expect, by the reduction of Rvp give-away. This is caused by the economic optimization tool because butane, as a low-cost component, is preferably used, as compared to low Rvp, high-cost components.

The changes in oxygenates are more or less balanced and only affected by different component prices. Further, more low-value components, such as hydrocracker light naphtha and pyrolysis heavy naphtha, are used as blend components in increasing amounts, whereas the iso-pentane content is not affected.

The gasoline blending system as operated in the Gelsenkirchen refinery since 1987, and by two licensors, is a strategic tool to improve refineries' operating margins at a moderate investment (typically, US$ 3-4 million). The payout is less than 1.5 years, and projects require less than 13 months.

The second article of this series will report on the application of this system for diesel/light fuel oil blending.

BIBLIOGRAPHY

1. "Motor vehicle emission regulations and fuel specifications-1990 up-date," Concawe, report no. 2/90.

2. Piel, W.J., Thomas, R.X., "Oxygenates for Reformulated Gasoline," Hydrocarbon Processing, July 1990, p. 68.

3. Crow, P., "U.S. industry refighting battles on Clean Air Act," OGJ, July 23, 1990, pp. 15-17.

4. Tallett, M.R., Dunbar, D.N., "Impact of Green Gasoline," Petroleum Economist, April 1990, p. 126.

5. Unzelman, G.H., "Reformulated gasolines will challenge product-quality maintenance," OGJ, Apr. 9, 1990, pp. 43-48.

6. Crow, P., Williams, B., "U.S. refiners facing squeeze under new federal, state air quality rules," OGJ, Jan. 23, 1989, pp. 15-18.

7. Unzelman, G.H., "Higher diesel quality would constrict refining," OGJ, June 29, 1987, pp. 55-59.

8. Unzelman, G.H., "U.S. gasoline pool octane increase may be limited, " OGJ, Apr. 4, 1988, pp. 35-41.

Copyright 1991 Oil & Gas Journal. All Rights Reserved.