Hydrocarbon Processing Copying and distributing are prohibited without permission of the publisher
Email a friend
  • Please enter a maximum of 5 recipients. Use ; to separate more than one email address.

Enhance refinery profitability with modeling innovations

07.01.2013  |  Beck, R.,  Aspen Technology Inc., Burlington, MassachusettsAjikutira, D.,  Aspen Technology Inc., Burlington, MassachusettsHerrmann, L.,  Aspen Technology Inc., Burlington, MassachusettsYe, V. ,  Aspen Technology Inc., Burlington, Massachusetts

Several improvements are on the horizon, which will add powerful optimization capabilities into the process modeler’s simulation work space in areas like heat integration, column optimization and economics.

Keywords: [process modeling] [software] [training] [refining] [heat exchangers] [engineering & design] [modeling ]

Refining operations continue to be a crucial but challenging element in the petroleum value chain; critical to the ability to bring hydrocarbons to market, but, in many instances, still a challenge in terms of attaining targeted levels of profitability. Fortunately, there have been rapid innovations and advances in several of the key enabling modeling technologies that contribute to the ability to react to technical and business changes. There are more innovations on the horizon. Refinery assets and operations are in the midst of an evolution, both in terms of the worldwide distribution and age of refining capacity, and the demand for flexibility in crude selection and product portfolios.

Software innovations discussed here fall into four general groups. One area of innovation is the workflow and usability of sophisticated refinery modeling tools. Increasingly, high-fidelity models are being used for refinery operation, planning, maintenance, cost feasibility and troubleshooting. The importance of these developments cannot be understated as a new generation of engineers enters the workforce.

A second area involves breakthroughs in the integration of models that enable optimization and improvement on a refinery-wide basis, such as heat integration, energy savings and capacity improvement. A third area involves advances in underlying science and methods, including improved molecular modeling and better statistical models to characterize and represent properties of crude oil, and transparent reactor models that provide rapid updating of planning models. A final and increasingly important advance encompasses new usability paradigms that provide mobile access to all plant models and data anytime, anywhere.

The business result of the above advances is a significantly improved ability of the organization to respond to crude selection and product contract opportunities, reduction in energy consumption, and improved sustainability and control. These are the requirements of the next generation of refineries.

Faster training

Process modeling systems that can represent, characterize, model and optimize refineries involve considerable complexity of functionality, tools, options and reporting capabilities. As the functionality of these models increased over the past 15 years, the usability of the modeling systems did not improve at a parallel pace, and, in many cases, it became problematic.

Over the past year, in-depth analyses of how a modeler builds and employs a refinery simulation model has led to more easily accessed models, more easily understood interfaces and improved workflows.1 Fig. 1 shows a stylized program ribbon as implemented for properties analysis and for simulation modeling.


  Fig. 1. Ribbon bars for physical properties and simulation modes characterize the new generation of process modeling tools. 

High-fidelity exchanger models

Refineries have an ongoing focus on reducing their energy footprint by optimizing energy use through operating, equipment, maintenance and process-configuration strategies. The use of more rigorous models for heat exchangers, and the embedding of those models within the refining simulation model, have proven to be an important step for making refineries more energy efficient. By modeling heat exchangers more accurately, process design optimization can be less conservative, making the process more profitable. Early identification of structural challenges, such as vibration or fluid momentum (Rho V2) problems, means a shorter cycle from conception to feed.

Rigorous models also calculate pressure drops that help design related equipment, such as pumps. These rigorous modeling results are used to plan the upgrading or reconfiguring of heat exchangers as a method of utilizing energy more efficiently and to ensure that the selected exchanger configurations are optimal for the conditions under which they will operate.

A full geometric predesign of heat exchangers and incorporation of the exchanger in the simulation is now possible. This has resulted in engineers, equipment designers and operators having early and greater fidelity in examining the impact of design and operations decisions on heat use in the plant.

Another important aspect of heat exchanger design that relates to energy efficiency, as well as to efficient operation, is equipment fouling. Rigorous heat exchanger models can be used to calculate fouling resistance. This provides operating benefits because, as fouling increases, equipment throughput sharply decreases, creating serious efficiency issues, as shown in Fig. 2. 


  Fig. 2.  Example of impact of fouling on refinery
  furnace systems.

This heat exchanger modeling for fouling was used by a major chemicals producer and refiner when modeling a heat exchanger train. The refinery was experiencing significant increases in operating costs because of heat exchanger fouling. The lack of a rigorous model meant that the refiner was unable to accurately calculate the fouling of single exchangers, as well as the entire train, and it had no means to estimate output temperature increases after the train was cleaned. With a rigorous model, the company was able to accurately model the fouling and determine how often the heat exchangers needed to be cleaned for optimal operation. The profit resulting from those improvements alone is estimated at between $1.5 million (MM)/yr and $2 MM/yr.2

Molecular modeling and assay

In most geographies, the selection of crude oils available as refinery inputs has widened, introducing more operating choices. Crude oils are complex mixtures, and, depending on where the crude is sourced, they can vary greatly in composition. Due to the large number of different hydrocarbon molecules that can be present in petroleum, it is infeasible to fully define the composition of the mixture. However, each crude oil type has unique molecular and chemical characteristics, so an assay is used to evaluate the properties of the petroleum and obtain data to characterize crude oil feeds. By characterizing these assays, refiners can see whether a specific crude oil feedstock is compatible with a particular petroleum refinery or if it will cause quality, yield, economic or environmental issues.

Evaluations are costly and tedious, and they result in a limited set of property measurements for the crude. Thus, statistical extrapolation and interpolation, as well as estimation methods, are used to predict missing properties for refinery planning and process simulation.

These statistical methods have been used extensively in the industry, but the limited assay data makes precise fitting difficult, which can result in incorrect characterizations that will impact the accuracy of the model. Modelers must take advantage of special factors or handles provided by traditional assay characterization tools to ensure that the results are correct and lead to realistic modeling outcomes. Recent research has focused on improving such tools to more fully incorporate the engineering knowledge of the problem, with the goal of better results.

Traditional analytical approaches suffer from extrapolation limitations, depending on the assay data available. With the selection of crudes on the market becoming heavier, the need for a new and fundamentally better methodology for crude assays is more important than ever.

An exciting new innovation, molecule-based characterization, offers the strongest scientific basis for the prediction of crude oil properties, as it bases its calculations on the chemical compositions of the hydrocarbon constituent molecules and on accurate molecular thermodynamic models for hydrocarbon mixtures.3

This approach to crude characterization has, as a basis, the principle that all hydrocarbon molecules can be constructed from a set of different structural segments, which can be described as a specific structural combination of carbon, hydrogen, sulfur, nitrogen and oxygen atoms. By modeling the complex hydrocarbons in the crude oil as a series of repeating molecular segments, the assay characterization has significantly improved accuracy, especially for heavier and increasingly varied crudes, such as high-sulfur oil. Refiners and planners are better able to estimate the properties of the crude oil feedstock, which results in more accurate reaction modeling.

Furthermore, using the same, improved assay characterization method for both the simulation model of the refinery and reactors, and the planning model for refinery operations, leads to overall improvements in the ability to make economically optimized decisions and to successfully predict refinery conditions and performance.

High-fidelity reactor models

The improvements seen in modeling software in recent years have been especially remarkable in the reactor design area. These rigorous reactor models have accounted for significantly more accurate determinations of equipment operations and easier identification of possible optimizations.

In addition, whereas before it was necessary to model these reactors separately and modify connecting streams by hand, it is now possible to model them in one integrated flowsheet.4 This advancement has become more significant as more reactors are added to refineries, since it enables modeling of the interactions between reactors to better understand and optimize the process.

The increase in the number of reactors is due primarily to the availability of heavier crudes. To keep up with this trend, modeling software has focused on expanding the properties available to model heavy crudes and, therefore, improve their reaction modeling, as mentioned in the previous section.

To make the use of heavier crudes more economically feasible, refineries have been adding cracking units to break the heavier crudes into simpler hydrocarbons to obtain the desired product blends. These sophisticated and vital cracking units must be rigorously modeled to serve this purpose, which increases the need for rigorous modeling software that can perform within the context of a wider simulation model.

The trend toward adding reactors is seen, for example, in Royal Dutch Shell adding hydrocrackers to three refineries in Holland, China and Poland. In all three cases, the hydrocrackers were added to take advantage of heavier crudes, which are more economical than lighter crudes.5 Modeling these hydrocrackers is extremely important for making this process feasible and will become more so, as more of these units are added to refineries.

Planning model

Another important development in refinery modeling involves advances in the ability of integrated software to update widely used refinery planning models more accurately, more frequently and with less specialized expertise.6 Support for this activity within the newest releases of refinery simulation models leads to better accuracy in the planning model for feedstock selection. The automation and demystifying of these interfaces is increasingly important as the addition of reactors increases the complexity of the process.

With the ability to model how reactors influence the end product, modeling software can interface with planning software to assist in the selection of the best feedstock for the desired products. Since the reactors involved in the process need to be rigorously modeled to give an accurate result, it is crucial to have planning software that can interface with the modeling software to give the most accurate results possible. The result is optimized reactor operating conditions and product output. In the past, this activity always required a heavy dose of expert consulting input, the resources for which were not always available; however, that is no longer the case.

Mobile interfaces to refinery models

With the advent of mobile technology and the increasing number of users of portable devices, it seems logical for refining and engineering companies to take advantage of these new capabilities to maximize productivity. By introducing applications that allow users to access process charts and data securely, without having to be onsite and without having to be experts in the underlying tools and models, managers and engineers can instantly receive updates and access models to keep track of plants wherever they are located (Fig. 3).


  Fig. 3.  Innovative access to refinery models on mobile
  devices is growing.

The ability to manage various processes from any location can help companies substantially reduce costs, while also increasing flexibility and worker efficiency. With the popularity of tablets and smartphones on the mobile devices market, these applications can take advantage of the convenience of monitoring ongoing activities at the plant from anywhere.

Additionally, these devices have the added benefit of being always on, always with the user and usually connected to a global network. By providing a mobile interface for engineering applications, users will rarely be in a position where they will not be able to access data securely and instantly. This key innovation ensures that companies keep up with, and take full advantage of, modern shifts in technology.

More innovations to come

The innovations described here have largely been introduced over the past two years, greatly accelerating the pace of innovation and modeling power for refinery operations, improvements and design. This pace of innovation is unlikely to slow down. A number of exciting improvements are on the horizon, which will add powerful optimization capabilities into the process modeler’s simulation work space in areas such as heat integration, column optimization and economics.

Additionally, the refinery manager can expect important advances in the power and value of mobile interfaces to give the manager access to key refinery performance information, anywhere and anytime, and to give the engineer access to the technical details required to make improvements to the performance. HP


1 Press release, Aspen Technology, December 10, 2012.
2 Berlin, G. Y. and P. Robert, “Quantifying and monitoring fouling of refinery heat exchangers,” INEOS, AspenTech User Conference, 2008.
3 Press release, Aspen Technology, “New assay management functionality in Aspen PIMS software optimizes crude purchasing decisions and increases profitability,” August 27, 2012.
4 Shethna, H., “Operations support for refinery planning and simulation,” Saudi Aramco, 2013 Optimize Global Conference, May 7, 2013.
5 Shell Global Solutions, “Key lessons from successful hydrocracker projects,” Hydrocarbon Processing supplement, September 2011.
6 Briggs, B. and K. Lau, “Webinar with BP: Improving refinery margins with hydroprocessing model applications,” BP Refining and Logistics, January 10, 2012.

The authors
  Ron Beck is the engineering product marketing director for Aspen Technology, covering the aspenONE engineering suite. He has worked for AspenTech for five years and is the marketing manager for aspenONE Engineering. Mr. Beck spent 10 years in a research and development organization commercializing fluidized bed technologies, enhanced oil recovery methods and environmental technology. He has 20 years of experience in the development, adoption and marketing of software solutions for engineering and plant management. Mr. Beck has been involved with the development of integrated solutions for several global chemical enterprises. At AspenTech, he has also been involved with AspenTech’s economic evaluation products. Mr. Beck is a graduate of Princeton University in New Jersey. 
  Dinu Ajikutira is Aspen Technology’s senior product manager for the Aspen HYSYS family of process optimization products for the energy industry. He has worked with Aspen Technology for nearly seven years, in both product management and research and development. Prior to working at AspenTech, he developed modeling technology for the process industries and also worked on the plant floor. Mr. Ajikutira is a chemical engineer with an MS degree from the University of British Columbia in Canada. 


Have your say
  • All comments are subject to editorial review.
    All fields are compulsory.


I would appreciate having a chance to talk with the authors of the article

Related articles


Sign-up for the Free Daily HP Enewsletter!

Boxscore Database

A searchable database of project activity in the global hydrocarbon processing industry


Is 2016 the peak for US gasoline demand?




View previous results

Popular Searches

Please read our Term and Conditions and Privacy Policy before using the site. All material subject to strictly enforced copyright laws.
© 2016 Hydrocarbon Processing. © 2016 Gulf Publishing Company.