|Theme||Visible||Selectable||Appearance||Zoom Range (now: 0)|
Valova, Elena (OneSubsea a Schlumberger company) | Eliseeva, Ekaterina (OneSubsea a Schlumberger company) | Roberts, Ian (OneSubsea a Schlumberger company) | Meyer, Joerg (OneSubsea a Schlumberger company)
It is a sobering statistic that 68 of 76 proposed major offshore developments were cancelled in 2015. There has been gradual improvement since then with system cost reductions and the oil price recovery, but nonetheless prospects that could be very profitable with the right development concept have been rendered uneconomic by overblown cost estimates. Conversely prospects with poor economics may go ahead due to overestimated production expectations.
Production forecasts are generated by reservoir models. These models are often first constructed early in the development process, and refined multiple times as the process progresses and more information becomes available, e.g., from appraisal wells and production history. Meantime, production system design and sizing are often performed by different groups in the E&P company or by subcontractors in virtual isolation from each other. Updates from the reservoir modelling are not necessarily shared between these groups resulting in disconnects that can impact project viability and economics.
Better collaboration between these groups and use of integrated reservoir and production system models address and sometimes resolve these issues. However, even in cases where integrated models are used, rather than harnessing the full power of the reservoir model, it is often dumbed down. This is done primarily in an attempt to save time, but it results in poorer representation of the actual reservoir response, e.g. losing the effect of well interference and failing to capture the pressure distribution and pressure changes with time in the reservoir.
In this paper some real examples are provided of disconnects and lack of modelling fidelity with the potential for severe consequences on project economic performance.
The four examples that will be discussed in this paper are as follows:
Over-simplification of the reservoir representation for a gas field led to a proposed high- capacity subsea compression system being uneconomical. Reevaluating with a full reservoir model showed the compression system was over-sized and would underperform. Correcting these issues transformed the project economics and enabled the project to progress towards the FID.
A simplified representation of the reservoir for a tieback resulted in insufficient natural production (no artificial lift) to be economic and insufficient production increase to justify subsea pumping. However, when reevaluated with more detail, the project became economical using a pumping concept.
Passing a simplified production forecast from the reservoir team to the flow assurance team for a long tieback development resulted in overlooking the most onerous period of production for thermal management. If this oversight had made its way into the final design of flowline insulation, the tieback may well have plugged with wax deposits within a few years of operation. Use of complete forecasts within an integrated simulation framework prevents this kind of issue.
Use of an inappropriate simplifying assumption while evaluating a brownfield redevelopment resulted in an overestimated production potential of the proposed new wells. When the error was corrected, an additional iteration of the development concept to downsize it was required which delayed first oil. Integrated reservoir and production system modelling avoid these kinds of planning and investment mistakes happening in the first place.
This paper challenges the notion that detailed, integrated simulation is not required until later stages (FEED, execution) of a field development project, if at all. The presented examples are based on actual project experience. They provide quantitative evidence that integrated simulations have avoided costly development concept iterations and even more costly mistakes. Sometimes this allowed assets with marginal economics to be reconsidered for the operators’ project execution pipeline. Sometimes it resulted in the need to revisit economic analysis and development concepts due to initial overestimation of well performance.
A great deal of interest in composite reservoir flow solutions has been shownby the increasing number of publications dealing with this problem. So far,only a limited number of solutions have appeared. The solutions available donot seem to be of completely general utility. The purpose of this paper is todemonstrate a class of approximate solutions to composite reservoir flowproblems which may be generated readily. These solutions are applicable to welltest analysis and, although approximate, have the advantage of involving simplefunctions only. One interesting result is that secretary approximating formswhich ha""e been presented previously derive readily from this class ofsolutions. Another interesting result is that outer boundary conditions ofeither a closed boundary or a constant pressure boundary may be approximated aslimits of composite reservoir solutions.
THE TERM 'COMPOSITE' was originally used in very early studies of heatconduction in solids composed of two different materials; seeCarslaw(l) for example. Despite the well-known analogy between heatconduction and laminar fluid flow through porous media, intense interest influid flow through composite solids has occurred recently. Hurst(2)and Mortada(3) analyzed the interference between oil fields in acommon aquifer of two different permeabilities in 1960. Loucks andGuerrero(4) presented an analysis of drawdown and buildup in radialcomposite systems in 1961. Rowan and Clegg(5) presented approximatesolutions in 1962. Van Poollen and associates have presented several papersconcerning radial and linear discontinuities within porous media and theresultant effects upon well test analysis(6,7). Carter(8)presented an analysis of the depletion of a closed composite radial reservoir.and discussed reservoir limit tests for this class of reservoirs. Adams, etal.(9) employed Hurst's approximate result for radial compositereservoirs to interpret pressure buildup tests in a fractured dolomitereservoir. Recently, Wattenbarger and Ramey(10) employed a radialcomposite reservoir model to simulate the skin effect of a finite storagecapacity.
The references given above concerning oil and gas reservoir applications of thecomposite reservoir problem are provided to show the increasing interest inthis class of flow problem. No attempt has been made to provide a completelisting. The bibliographies of the papers cited list other pertinent studies.The main point is that composite reservoir flow problems are significant, andthat the connection between various applications of the solutions is notclearly evident.
One interesting fact is that papers concerning composite reservoir problemshave often raised controversies or debates, or left other evident problemsunsettled. Part of the reason must lie in the complex nature of these problems.Fortunately, the approximate solutions which will be discussed here containonly elementary functions.
The experimental data of Donnelly and Katz(1) were used for themethane - carbon dioxide system and the data of Price andKobayashi(10) were used for the methane-ethane-propane system. Theresults of the calculations are shown graphically in Figure 6, where again itis seen that the convergence pressure concept has failed to correlate theK-factors for methane in the presence of the non-hydrocarbon component.
Abstract Simple methods are presented for productivity computations for multilateral, branched and other generalized and extended well concepts, mostly based on computations of pseudoradial skin factors for systems of connected wells, intervals or fractures with a common bottomhole pressure (or potential). Also included are applications to perforation size situations. For symmetric cases, either with respect to formation boundaries or with respect to a rotational surface around the well-system axis, simple closed-form expressions are presented. Similar, although more complex, expressions are also given for linear symmetric cases with 3 or 4 wells or branches. For other cases a numerical approach is outlined. The accuracy of the direct results depends on the vertical component and horizontal distance between well elements, and on methods used to compute skin values of individual well elements, but numerical schemes to improve the results have also been outlined. Introduction Although several methods have been introduced in the literature to determine productivity indices or skin values for multilateral and multifractured wells (see for instance Refs. 1–4), the cases covered are mostly restricted to highly symmetric well configurations. The present paper, on the other hand, considers a more general approach that can be used to determine productivity indices or skin values for quite arbitrary well configurations in homogeneous reservoirs of constant thickness. The approach used assumes a certain distance between the midpoints of the well elements, e.g., branches, drainholes, and fractures, of the well or well system in question, and then assumes the pressure development to depend on the distance only. Even relatively close to a horizontal well, it is for instance possible to estimate the pressure development with negligible error by using a fully penetrating vertical line well. Moreover, since the line-source solution can be approximated with a simple logarithmic expression at sufficiently late times, it is possible to combine the effect of for instance several wells or branches by elementary methods. This is especially simple if the rate of each well or branch is known, e.g., if they can all be assumed identical from symmetry considerations. Otherwise the rates must be determined by solving a system of linear equations as part of the computation of the productivity index or skin value. For cases where simple line-source solutions cannot be assumed to accurately represent the pressure development at a given observation point, it might be necessary to use more accurate analytical solutions to compute adjustment factors. These adjustments can be applied to the distance between well elements (e.g., midpoint of branches) as has been outlined in Appendix B. Skin Factors and Productivity Indices For the developments of this paper it suffices to consider pseudosteady state productivity indices of the form (1) in practical SI units, where y =0.57722156… denotes Euler's constant, CA is the Dietz form factor for the shape of the drainage area and well location, and S is the skin factor of the well. For a circular drainage area with the well at the center, Eq. 1 takes the familiar form (2) where re is the outer radius of an equivalent circular drainage area. P. 739
This article is a synopsis of paper SPE 49222, "Analyzing Well Production Data Using Combined Type-Curve- and Decline-Curve-Analysis Concepts," by R.G. Agarwal, SPE, D.C. Gardner, SPE, S.W. Kleinsteiber, SPE, and D.D. Fussell, SPE, Amoco E&P Co., originally presented at the 1998 SPE Annual Technical Conference and Exhibition, New Orleans, 27-30 September.