Non-equilibrium phase behavior of hydrocarbons has a great influence on the development of gas condensate and oil and gas condensate deposits, as well as oil deposits after evolution of dissolved gas. The key parameter of mathematical models that take into account non-equilibrium phase behavior is the characteristic relaxation time. The paper presents examples of assessing its value according to the results of laboratory experiments and well tests, which also allow verifying the mathematical model and obtaining input data for the production forecast taking into account non-equilibrium processes. A model of non-equilibrium phase behavior in a compositional formulation, based on relaxation of the chemical potentials difference, was verified on the data of a laboratory experiment in a calorimeter cell. The numerical algorithm for calculating the dynamics of phase fractions and compositions in a non-equilibrium process was modified to take into account the isochoric condition, which made it possible to directly compare the simulation results to the real experiment. A good match of the experimental pressure-temperature hysteresis curves was obtained without additional adjustment of the characteristic relaxation time as compared with the estimate obtained from processing the measured pressure relaxation curve. Thus, the adequacy of the mathematical model and the possibility of evaluating its parameters directly from the experimental measurements were confirmed. An example of assessing the characteristic relaxation time from well test data is presented for the pressure buildup (PBU) curve recorded under the conditions of gas dissolution in oil. A sector blackoil-type model was used for the well drainage area. For matching the PBU data, various modeling methods for non-instant gas dissolution were considered, including a model based on relaxation of solution gas-oil ratio (sGOR). The results of simulations with the sector model showed that the traditional options for limiting the sGOR increasing rate do not allow adequate matching of the PBU curve under conditions of gas dissolution. For the model that takes into account the sGOR relaxation, an in-situ estimate of the characteristic relaxation time was obtained which provided a fine match of the PBU data. As expected from theoretical concepts, the characteristic relaxation time for the in-situ process has a much larger order of magnitude than for the laboratory experiments on centimeter-sized cells. The results of the study show the adequacy of relaxation models for non-equilibrium phase behavior in compositional and black-oil formulations for describing real processes and demonstrate the possibility to estimate the characteristic relaxation time from laboratory experiments and well tests, taking into account its dependence on the process scale.

The tendency of recent years in the sphere of modeling oil and gas fields and the processes of oil and gas fields' development is an increase of complication of models, revision and abandonment of a number of simplifications, which were necessary earlier due to the limited computing and time resources available to engineers. This tendency makes more widespread such approach as accounting for the performance and optimization of surface gathering systems becomes a critical task in the case of gas, gas condensate fields, and also - light oil fields. Given the apparent growth in the interest of oil and gas companies in such assets, the popularity of integrated modeling in the industry as a whole is extremely high. The problem with this technology is that the integrated simulation models are complex in the creation, they require more knowledge from the engineers involved in the project in the related field of activity (the reservoir engineer should have an idea of the principles of operation and methods of managing the surface network, production engineer should understand the physical basis of the work not only of the reservoir, but also of its simulation model, etc.). Both the network model and the reservoir model can be performed in different variants in terms of the complexity and accuracy of the transfer of the boundary conditions and taking into account the necessary physical effects. This article offers a comparative analysis and possible applications of different types of models, and also describes a number of additional tools that help to eliminate some typical problems of this class of tasks.

A physically-consistent method is suggested for calculation of coexistent hydrocarbon phases' compositions in flow simulations with non-equilibrium phase transitions. The method preserves standard flow terms' formulation and can be incorporated in existing commercial simulators without principal modifications to flow simulation algorithms. Through the analysis of laboratory experiments data and computational studies, we show that the necessity for non-equilibrium phase behavior model depends on simulation scale. We introduce an upscaling algorithm to estimate parameters of a non-equilibrium coarse grid model. An algorithm is proposed to calculate non-equilibrium phase compositions with desired level of non-equilibrium behavior and model different speeds of relaxation of fluids' compositions to their equilibrium states. Relaxation dynamics is determined by pressure changing rate as well as initial phase compositions and pressure. These effects are essential for consistent history matching of a flow simulation model of an oil or gas-condensate reservoir with minimal number of matching parameters. Examples of using the method to simulate non-equilibrium phase behavior of real oil and gas-condensate mixtures are presented.

The paper argues in favor of the need to adjust pVT properties of the hydrocarbon system and apply non-equilibrium models of phase transitions for purposes of upscaling and history-matching flow simulation models (FSMs) to actual production data. Since most of FSMs are based on the Black Oil model, it features prominently in the paper. The paper considers certain information from laboratory experiments and production data that points to non-equilibrium phase transitions in the in-situ hydrocarbon mix. Provided are also the results of numerical experiments demonstrating the need for upscaling the fluids' pVT properties, and evidence of non-equilibrium effects from the enlarged size of cells in the FSM computational grid. In conclusion, the paper considers capabilities of various commercial flow simulators and new approaches to simulation of non-equilibrium phase behavior effects in the Black Oil model.

Application of air injection as an enhanced oil recovery method involves high risks, both economic and technological. One of the main risks of this technology is the lack of oil self-ignition during air injection and difficulties in creating a combustion front in the reservoir. In a number of fields the reservoir thermal and pressure conditions and the oil properties cause spontaneous ignition (mainly light oil fields with a high reservoir temperature), while in other fields special ignition programs are required. In the present paper three possible ways of air injection process development are shown: the method of "pushing" oil with air without combustion front appearance, in-situ combustion (ISC), and thermogas treatment, or, using the American terminology, high pressure air injection (HPAI). The field’s properties that determine which way in-situ combustion process will develop are identified. One way to determine in which of these three ways the process will develop is to study the oxidation characteristics using the pressurised differential scanning calorimetry (PDSC). For the improved study of the combustion process DSC experiments of oil samples oxidation reactivity, oil-saturated core and other actively reactive substances are performed. High precision of modern thermo-analytical PDSC equipment enables investigating the initial stages of oxidation reactions of hydrocarbons and describing the initial stages of the reaction in terms of the Arrhenius theory as well as the theory of chain reactions. This approach allows us to identify the most reactive oil components (oil paraffins), and describe the process of their oxidation in terms of the chain mechanism of birth and death of free radicals. It is shown how oil can change its reaction activity at the reservoir rock samples; the activity among others, is caused by the adsorption of asphaltenes and oxidised components on the core surface during the heating process. In addition, the unsteady type of oil (tarry light oil) has been studied, oxidation of which is accompanied by transition of resins to secondary asphaltenes, which, as it has been first shown, in the presence of core significantly increases the amount of fuel. In the case when the pressure and temperature conditions of the reservoir and oil reactivity prevent spontaneous ignition, some cases for combustion initiation have been considered. It is shown that due to PDSC experimental results a comparison of hydrocarbon and other fluids in order to find a reagent with the lowest self-ignition temperature can be achieved. It is shown that vegetable oil additions to oil can significantly reduce the temperature of the heat output, and can serve as oil ignition initiators for air injection. Technical aspects of combustion ignition are also discussed in the paper. The technology integrate heating agent and reactive hydrocarbon liquid injection, which reduces the lack of ignition risks in the oilfields with low temperature.

Specific of gas field development suggests a constant level of gas production in order to provide stable loading of gas field processing and transport facilities. At this point, when production of major development objects commences to fall, start of minor objects production is supposed to occur. Having united surface facility and separate reservoir simulation of segregated layers with joint well stock a question of accuracy in measurement of hydrocarbon production is arisen. In addition, at the entry of production of minor objects with distinctive reservoir pressure a problem of appropriate distribution of wellhead and downstream pressures can be revealed. In this paper mechanism of building of a multiple layer oil and gas condensate field integrated simulation model (ISM) is described. Construction of integrated model will allow resolving the problems outlined above. Besides, running of field development simulation cases with different order and tempo of production will be significantly simplified. Separate compositional models made for each development object have been selected for integration in the ISM. However, consolidation of these models is complicated by the issue of demanding a lot of resources for storage and computation time. Therefore, a transition from compositional to black oil models has been performed in order to obtain production data. Detailed gas composition required for integration with a gas processing system can be gained by a delamping procedure. As a result of building integrated simulation model a flexible tool of effective management of field development has been obtained that makes it possible to run simulations for both major and minor objects simultaneously and taking into account the field infrastructure (gathering and compressor facility); to assess the possibility of joint production of separate layers and consider the impact of gathering system on hydrocarbon production as well. When comparing the simulation time in different types of models, without significant loss of simulation quality, the average acceleration is about twenty times. This allows obtaining results much faster and further improving it.

Abstract The paper is devoted to the prediction of the condensate content in a gas-condensate mixture during depletion, and the estimation of the condensate recovery factor. It is shown that the traditional approach to tackle this problem may result in both overestimated and considerably underestimated condensate production forecast. Individual and combined influence of the formation temperature and dispersed hydrocarbon (HC) liquid on the variation of the C 5+ content in the reservoir gas, and the condensate recovery factor for the depletion of the condensate gas cap are analyzed with examples of the Orenburg and Dulisma fields. It is proved that the traditional approach to predicting the presence of C 5+ group in the formation gas by ignoring the presence of the dispersed hydrocarbon liquid (DHCL) in the reservoir can lead to overestimated condensate production rates. Particular influence of the anomalous low formation temperature on the retrograde condensation rate and the associated condensate recovery factor are analized. Advanced methods for modeling multicomponent phase behavior are used to analyze the impact of the HC fluids properties on the expected C 5+ content in the produced mixture, and on the condensate recovery factor for the depletion of near critical gas condensate reservoirs. It is shown that when dealing with such HC systems, the traditional approach to predicting the C 5+ hydrocarbon content in the produced gas, based on differential condensation (constant volume depletion, CVD) studies, may result in significantly underestimated condensate production, and consequently its recovery factor. Introduction Pore space of the natural gas reservoirs is partly filled with other fluids, including DHCL. Their amount and distribution within the reservoir are controlled by the reservoir formation peculiarities. Data supplied by A.G. Durmishian 1 is one of the first oilfield evidence supporting the presence of DHCL in gas zones. K.V. Pokrovsky and his teammates 2 were the first researchers in the USSR who analyzed physical and chemical peculiarities of gas/condensate systems that contained liquid hydrocarbon phase prior to the field production start up. Those studies showed that the presence of DHCL in reservoirs VII-VIII of the Karadag gas/condensate field (KGF) resulted in more rapid C 5+ condensation. M.T. Abasov, A.G. Kovalev, R.M. Kondrat, V.A. Nikolaev, F.G. Orudjaliev, V.I. Petrenko, G.V. Rassokhin, R.M. Ter-Sarkisov, M.S. Razamat, G.S. Stepanova and others 3–10 also studied the phase behaviour and hydrocarbon recoveries for the gas/condensate reservoirs that contained DHCL. It was established that the component recovery at early development stages of the reservoir containing DHCL is below that of the DHCL-free reservoirs. The current component recovery of a mature field depends on whether or not the previously immobile liquid hydrocarbon phase is involved in the reservoir fluid flow. The condensate recovery factor (CRF, condensate recovery factor) and the potential C 5+ yield behaviour of depletion-drive reservoirs are based on the reservoir gas CVD (constant volume depletion) data. These tests are usually performed when the field is being prepared for commercial production. Lessons learned in operating the Orenburg field showed that in case the reservoir-gas dew point equals the initial reservoir pressure, the predictions should presume a possible DHCL influence on the condensate production dynamics. Reservoir temperature is another critical factor that controls the intensity of the retrograde phenomena in the natural gas/condensate systems. The example of the Dulisma field, which exhibits anomalously low reservoir temperature (ALRT) is cited to analyze both the DHCL and reservoir-temperature impacts on the CRF relationships and the behaviour of the C 5+ yield of the gas produced.