Initial, irreducuble water saturation, Swir is an important parameter that needs to be determined accurately when attempting to characterize hydrocarbon reservoirs. Swir is also one of the key parameters in relative permeability relationships. Furthermore, an unrepresentative value of Swir may lead to invalid residual oil saturation estimates when the latter is correlated with the former. Swi may have a dependence on several other parameters, including: absolute rock permeability, porosity, pore size distribution and capillary pressure. The above parameters are directly influenced by geological deposition and subsequent changes, such as diagenesis effects (for example clay-filled pores). It is a common practice to measure Swir utilizing representative core plugs by measuring capillary pressure with a centrifuge, at speeds equivalent to the maximum representative (reservoir) capillary pressure. However, a semi-empirical model that could estimate Swir to a good degree of accuracy would be of significant value. Over the last few years, artificial neural networks have found their application in petroleum engineering. In some cases such models have outperformed models employing conventional statistical and regression analysis. In this study, an Artificial Neural Network (ANN) model has been developed for the prediction of Swi (specifically irreducible saturation, Swir) using data from a number of onshore and offshore Australian hydrocarbon basins. The paper outlines a methodology for developing ANN models and the results obtained indicate that the ANN model developed is successful in predicating values of Swir over the range of data used for calibration. This neural network based model is believed to be unique for Australian reservoirs Introduction The accurate estimation of initial water saturation for a hydrocarbon reservoir is an essential practice in the oil and gas industry. The exact determination of Swi leads to a precise evaluation of initial hydrocarbon in place, which in turn provides valuable insight into further field development plans. Moreover, and as an important aspect in multiphase flow problems, the calculation of valid Swir values is an important requirement for obtaining accurate relative permeability relationships and for determining residual oil saturation. Swir is measured as part of primary drainage capillary pressure experiments on core plugs obtained from the reservoir under consideration. In a centrifuge test, a plug is saturated with formation or synthetic brine and positioned in the centrifuge cup filled with oil. The centrifuge is then spun at different speeds, pushing the water phase outward. The volume of the expelled water is measured and the centrifuge speed is increased until it reaches a pre-determined limit, which represents the maximum capillary pressure. Having that in mind, a careful calculation of the maximum reservoir capillary pressure and consequently the maximum capillary pressure under laboratory conditions are of importance (1). As an alternative to the centrifuge test described above, a core-flood test may be conducted. In such tests, the plug is initially saturated with brine. The oil phase is then introduced, resulting in a reduction in water saturation. The test continues until the plug reaches a stabilized condition where no more water is expelled. Care must be taken when interpreting the results from both tests described above. This is attributed to the fact that both tests employ different mechanisms and consider different influencing forces, including gravitational forces for centrifuge tests and viscous forces in core flooding tests(1).
Capillary pressure and associated relative permeability are special core analysis parameters which are vital in accurately describing fluid distributions and movement in porous media when two (or more) immiscible fluids are present. Conventionally, the determination of capillary pressure requires laboratory experiments, which are expensive and time consuming. Assuming that core material is available, typically a limited number of core plugs are considered for testing, often resulting in incomplete reservoir description. This situation then leads to motivation for developing mathematical capillary pressure models, as an alternative to adequately describing fluid behaviour in reservoirs.The well known Carman-Kozeny (C-K) equation (Carman 1937;Kozeny 1927) is commonly used to model permeability as a function of other pore structure parameters. More recently parametric groups were established for the C-K equation, which are in regular use in reservoir characterisation. Specifically, the determination of flow zone units for the description of distinct (litho)facies is now widely used in linking fluid flow to geological descriptions. In this paper, a new and innovative deployment of the C-K equation is investigated to derive capillary pressure relationships. The new mechanistic, drainage capillary pressure model uses the principle of "effective saturation", and is developed and validated with data sets from onshore and offshore Australian hydrocarbon basins, giving a good comparison with laboratory measurements and other established models. Most significantly, the new mechanistic formulation does not just fit capillary relationships but is able to predict drainage curves by knowing or assuming up to six properties: absolute permeability; effective porosity; irreducible water saturation; maximum capillary pressure, which may be related to a particular reservoir situation or laboratory standard; capillary entry pressure; and associated entry water saturation. The last two parameters are typically required for a higher permeability (or better quality) rock. Absolute permeability is incorporated indirectly. Furthermore, there is no requirement for correlation parameters or constants but the analysis determines two parameters which give a measure of heterogeneity of the sample being investigated.
Artificial neural networks theory creates, with other theories and algorithms, a new science. This science deals with the human body as an excellent source, through which it can simulate some biological basics and systems, to be used in solving many scientific, and engineering problems. Neural networks are tested successfully in so many fields as pattern recognition or intelligent classifier, prediction, and correlation development. Recently, Neural network has gained popularity in petroleum applications. In this paper we applied this technique in PVT parameters determinations. The application interests in the estimation of the bubble point pressure through a designed neural network. As this value well estimated, it then used with other variables in a second network to determine oil FVF at this value of bubble point pressure. A comparison study between the performance of neural network and other published correlations has shown an excellent response with smallest absolute relative average error, and highest correlation coefficient for the designed networks among all correlations. Introduction PVT properties always determined experimentally from calculation based on samples collected from either well bore or at the surface. Such samples may be very expensive to obtain. Hence, in case of the absence of the experimental measured PVT properties, it is necessary to use the empirically derived correlation to predict the PVT data.1 Many correlation already exist in the oil and gas industry such as: Standing correlation, Glaso correlation, Beggs and Vasquez correlation,....etc. Many investigators recognize that the neural network can serve the petroleum industry to create more accurate predication PVT correlations. So, there is a number of papers in this area. R. B. Gharbi, and A. M. El- Sharkawy2,3, in 1997, published two papers in this field. The first paper use the neural system to estimate the PVT data for middle east crude oil reservoirs28, while the next one was interest in developing a universal neural network for predicting PVT properties for any oil reservoir3.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.