VMGSim 10.0 and Phase Equilibrium Review
Phase equilibrium and physical property estimation are the foundation of any process simulation. VMG has a history of innovation and excellence in thermophysical properties. VMG consistently includes improvements and updates of our thermodynamic predictions in every release of VMGSim. This includes state of the art estimation methods and routine review of the most up to date data available.
The release of VMGSim 10.0 incorporates our latest extensive review of data to validate and improve where appropriate the estimations of several systems of importance to the oil and gas industry. Our review concluded that the majority of our estimations were already of great quality, however we found some niche areas with room for improvement.
The recommended property packages in VMGSim for oil & gas applications are Advanced Peng-Robinson (APR) and APR for Natural Gas 2 (APRNG-2). The latter was specifically designed to provide consistent modeling for hydrocarbon-rich and water-rich phases using a flexible and consistent mixing rule. The mixing rule formulation in APRNG-2 is completely generic and can handle non-polar or polar mixtures. The interaction parameters are easily accessible through the Thermodynamic or Model Regression environments. This facilitates a rapid response from our Technical Support team in cases where unique interaction parameters can be regressed from equilibrium data and quickly deployed in VMGSim.
At VMG we believe that there are no shortcuts when it comes to providing the best “out of the box” predictions in our property packages. We continuously gather, review and regress as much published data as possible. Once the new regressed parameters are available, we validate the predictions and fine tune them as needed. This process requires the review of thousands of data points and hundreds of simulations. Fortunately, we can use VMGSim to keep track of this work and go back to it when this process has to be repeated.
Every release of VMGSim includes major milestones in the development of our thermodynamic engine. These are some of the most notable developments in the last 10 years:
- VMGSim 2.2 (2004) - Introduction of APRNG for accurate modeling of glycols, methanol, acid gas water contents, hydrates and other important systems
- VMGSim 2.7 (2005) - Introduction of Claus and Amines packages
- VMGSim 4.0 (2008) - Added 20,000 components to database, introduction of Gasification package
- VMGSim 7.0 (2012) - Major improvements to oil characterization, added RefProp property package
- VMGSim 8.0 (2013) - Introduction of PIONA based characterization, comprehensive review of hydrate data from literature
- VMGSim 9.0 (2014) - Introduction of APRNG-2 property package including a comprehensive review of data from literature
The list above marks only some of the major milestones in the development of VMGSim but there are a considerable number of improvements smaller in scope that are routinely done to improve predictions.
VMGSim 10.0 - Thermodynamics Data Review
Research Report 131 looks at triethylene glycol with methane and BTEX at conditions similar to those seen in a glycol dehydration process. It includes a variety of pure glycol streams as well as ones with water. The biggest improvement in this system was the amount of BTEX leaving in the vapour phase. BTEX are volatile organic compounds and considered harmful to health and the environment, therefore accurate prediction of their phase equilibrium is important.
Figure 1 and Figure 2 below show the improvement from VMGSim 9.5 to VMGSim 10.0. The error in toluene mole fraction in the vapour phase was reduced from 37.2% to 16.1%, while the error in ethylbenzene mole fraction in the vapour phase was reduced from 18.7% to 4.9%. Similar quality results can be seen for the liquid phase, with little change necessary from version 9.5 to 10.0. In VMGSim 9.5 the errors for methane, toluene and ethylbenzene are, respectively, 3.2%, 33.6% and 1.0%, while in VMGSim 10.0 the errors for methane, toluene and ethylbenzene are, respectively, 3.1%, 34.9% and 1.2%. The figures that follow show the changes to the vapour phase results graphically.
Figure 1. RR131 Table 2 Vapour Phase composition in VMGSim 9.5 using APRNG2
Figure 2. RR131 Table 2 Vapour Phase composition in VMGSim 10.0 using APRNG2
Research Report 149
Research Report 149 covers a wide range of acid gas and hydrocarbon components in equilibrium with ethylene glycol solutions. The investigation looked into ethylene glycol, water solutions in contact with various mixtures of methane, propane, n-heptane, methylcyclohexane, toluene and hydrogen sulfide. This combination of light hydrocarbons with C7 hydrocarbons led to a study that produced vapour-liquid-liquid equilibrium. Table 23 looks at acid gas solubility with ethylene glycol and water solutions. This data can be seen in Figure 3 and Figure 4 below.
Figure 3. RR149 Table 23 Liquid Phase composition in VMGSim 9.5 using APRNG2
Figure 4. RR149 Table 23 Liquid Phase composition in VMGSim 10.0 using APRNG2
Research Report 92
Research Report 92 looks at a variety of gas condensate hydrate inhibition results. This inhibition is done by both ethylene glycol and methanol. As was mentioned in the “Major Milestones” section, in VMGSim 8.0, a significant revision was made to the hydrate calculations, allowing us to match both inhibited and non-inhibited hydrate formation conditions with great accuracy. This can be seen in the figures that follow. Figure 5 and Figure 6 look at the hydrate inhibition performed be ethylene glycol in Table 5 of RR 92. Most of the data points collected are for vapour-liquid-liquid-hydrate equilibrium. The data is collected for non-inhibited temperatures, as well as 25 and 50 weight percent ethylene glycol inhibition. The average deviation in hydrate formation temperature in VMGSim 9.5 is 2.81F (1.55 C), while it is 2.76F (1.53C) in VMGSim 10.0.
Figure 5. RR92 Table 5 Hydrate formation temperature in VMGSim 9.5 using APRNG2
Figure 6. RR92 Table 5 Hydrate formation temperature in VMGSim 9.5 using APRNG2
Table 6 shows the results for methanol inhibition and is shown in Figure 7 and Figure 8. These values are for 25 and 50 weight percent methanol inhibition. In VMGSim 9.5 there is an average deviation in hydrate formation temperature of 4.04F (2.25C), while in VMGSim 10.0 the average deviation is reduced to 3.98F (2.21C).
Figure 7. RR92 Table 6 Hydrate formation temperature in VMGSim 9.5 using APRNG2
Figure 8. RR92 Table 6 Hydrate formation temperature in VMGSim 10.0 using APRNG2
A more complete list of the reports studied and the components examined in the reports follow.
Dehydration with Glycols
Acid Gas Treating
Water and Hydrocarbon Solubiity
VMGSim Thermodynamics - The Future
At VMG we recognize that there is always room for improvement when it comes to property estimation. In the previous section we included a summary of the validation work we did for the release of VMGSim 10.0. This work was mainly targeted to typical vapor-liquid and vapor-liquid-liquid systems. In future releases we will be doing similar development work in other areas such as solids (CO2, H2O, Benzene) and asphaltene precipitation. These developments are driven by the feedback we get from our users. We can also make available any of the regression cases or further details in these and any other validation work we do inside of VMGThermo.
Carl Landra, P.Eng., Simulation Software Developer & Linnea Russell, P.Eng., Technical Support Engineer
Please contact your local VMG office for more information or if you have any comments or suggestions.