Combined Reservoir Simulation And Seismic Technology, A New Approach For Modeling CHOPS

Author(s):  
Hossein Aghabarati ◽  
Carmen Camelia Dumitrescu ◽  
Larry Lines ◽  
Antonin Settari
2017 ◽  
Author(s):  
B. Kayode ◽  
S. Surdiman ◽  
Z. Ghareeb ◽  
H. Salem

2012 ◽  
Vol 30 (18) ◽  
pp. 1920-1930 ◽  
Author(s):  
M. H. Sefat ◽  
K. Salahshoor ◽  
M. Jamialahmadi ◽  
B. Moradi

SPE Journal ◽  
2017 ◽  
Vol 22 (06) ◽  
pp. 1999-2011 ◽  
Author(s):  
Guohua Gao ◽  
Hao Jiang ◽  
Paul van Hagen ◽  
Jeroen C. Vink ◽  
Terence Wells

Summary Solving the Gauss-Newton trust-region subproblem (TRS) with traditional solvers involves solving a symmetric linear system with dimensions the same as the number of uncertain parameters, and it is extremely computational expensive for history-matching problems with a large number of uncertain parameters. A new trust-region (TR) solver is developed to save both memory usage and computational cost, and its performance is compared with the well-known direct TR solver using factorization and iterative TR solver using the conjugate-gradient approach. With application of the matrix inverse lemma, the original TRS is transformed to a new problem that involves solving a linear system with the number of observed data. For history-matching problems in which the number of uncertain parameters is much larger than the number of observed data, both memory usage and central-processing-unit (CPU) time can be significantly reduced compared with solving the original problem directly. An auto-adaptive power-law transformation technique is developed to transform the original strong nonlinear function to a new function that behaves more like a linear function. Finally, the Newton-Raphson method with some modifications is applied to solve the TRS. The proposed approach is applied to find best-match solutions in Bayesian-style assisted-history-matching (AHM) problems. It is first validated on a set of synthetic test problems with different numbers of uncertain parameters and different numbers of observed data. In terms of efficiency, the new approach is shown to significantly reduce both the computational cost and memory usage compared with the direct TR solver of the GALAHAD optimization library (see http://www.galahad.rl.ac.uk/doc.html). In terms of robustness, the new approach is able to reduce the risk of failure to find the correct solution significantly, compared with the iterative TR solver of the GALAHAD optimization library. Our numerical results indicate that the new solver can solve large-scale TRSs with reasonably small amounts of CPU time (in seconds) and memory (in MB). Compared with the CPU time and memory used for completing one reservoir simulation run for the same problem (in hours and in GB), the cost for finding the best-match parameter values using our new TR solver is negligible. The proposed approach has been implemented in our in-house reservoir simulation and history-matching system, and has been validated on a real-reservoir-simulation model. This illustrates the main result of this paper: the development of a robust Gauss-Newton TR approach, which is applicable for large-scale history-matching problems with negligible extra cost in CPU and memory.


2016 ◽  
Author(s):  
Emil Mamedov ◽  
Ernest Zakirov ◽  
Vladislav Arekhov ◽  
Atlas Ahmetzyanov

1998 ◽  
Vol 1 (01) ◽  
pp. 18-23 ◽  
Author(s):  
Yu Ding ◽  
Gerard Renard ◽  
Luce Weill

Summary In reservoir simulation, linear approximations generally are used for well modeling. However, these types of approximations can be inaccurate for fluid-flow calculation in the vicinity of wells, leading to incorrect well-performance predictions. To overcome such problems, a new well representation1 has been proposed that uses a "logarithmic" type of approximation for vertical wells. In this paper, we show how the new well model can be implemented easily in existing simulators through the conventional productivity index (PI). We discuss the relationship between wellbore pressure, wellblock pressure, and flow rate in more detail, especially for the definition of wellblock pressure. We present an extension of the new approach to off-center wells and to flexible grids. Through this extension, the equivalence of various gridding techniques for the well model is emphasized. The key element is the accurate calculation of flow components in the vicinity of wells. Introduction The well model plays an important role in reservoir simulation because the precision of calculation in well-production rate or bottomhole pressure is directly related to this well model. The main difficulty of well modeling is the problem of singularity because of the difference in scale between the small wellbore diameter (less than 0.3 m) and the large wellblock grid dimensions used in the simulation (from tens to hundreds of meters), and to the radial nature of the flow around the well (i.e., nonlinear but logarithmic variation of the pressure away from the well). Thus, the wellblock pressure calculated by standard finite-difference methods is not the wellbore pressure. Peaceman2,3 first demonstrated that wellblock pressure calculated by finite difference in a uniform grid corresponds to the pressure at an equivalent wellblock radius, r0, related to gridblock dimensions. Assuming a radial flow around the well, he demons-trated that this radius could be used to relate the wellblock pressure to the wellbore pressure. However, there are problems with this approach in many practical reservoir simulation studies:For routinely used nonuniform Cartesian grids,4 there is no easy means to determine an r0 value.In three-dimensional (3D) cases with non-fully-penetrating wells, the basic radial flow assumption does not apply,5 whereas vertical flow effects must be included.6Off-center wells are not correctly treated.7,8Treatment of the well model is much more complicated with non Cartesian or flexible grids.9–11 The aim of this paper is to show that the new well representation1 proposed in a previous paper can handle these problems accurately. Wellblock Pressure Calculation A previous paper1 presented a new approach particularly well-suited to nonuniform grids for the modeling of vertical wells in numerical simulation. The principle of this new approach, which is based on a finite-volume method, is to calculate new interblock distances that improve the modeling of flow in the vicinity of wells. Because the new approach was originally presented for two-dimensional (2D)-XY problems, it was shown that for such problems the wellbore pressure could be calculated without both the intermediate computation of the wellblock pressure and introduction of an equivalent wellblock radius. However, for at least two reasons, it is convenient to keep this standard method commonly used in numerical models, which consists of relating the wellbore pressure and wellblock pressure through the use of a numerical PI and equivalent wellblock radius. One reason is practical. To implement the new approach more easily into standard numerical models, it is better to keep their internal structure unchanged. The other reason is dictated by the necessity of having a wellblock pressure in particular 3D simulation studies. When a well partially penetrates the reservoir or when there is communication between different layers, there is a vertical flow component in the vicinity of the well that necessitates that the wellblock pressure be calculated. How should the new approach be implemented in standard reservoir simulators- In these simulators, a numerical PI is used in the well model to relate the wellbore pressure, pw, to the wellblock pressure, p0. Usually, this PI is written as where r0 is the equivalent wellblock radius at which the pressure is equal to p0. Within the new well representation,1 to obtain a pressure p0 corresponding to a radius r0, it is sufficient to use equivalent wellblock transmissibilities relating p0 to the pressures of adjacent blocks through equivalent interblock distances, Leq, i (Fig. 1: where ?x0, ?y0 are the wellblock dimensions. For instance, in the x+ direction, Leq,1 is written where ?1+2 arctg (?y0 /?x0) is the angle formed by the right wellblock interface seen from the well. Because wellblock transmissibilities in standard models are conventionally expressed by the new approach can be implemented easily in standard models multiplying the conventional wellblock transmissibilities by constant factors. For instance, in the x+ direction, this factor is By use of equivalent transmissibilities, the calculated wellblock pressure, p0, should correspond to the equivalent wellblock radius, r0, which is involved in transmissibility calculations (Eq. 3). Then, the wellblock pressure can be related to the wellbore pressure with the conventional PI (Eq. 1).


2016 ◽  
Author(s):  
Emil Mamedov ◽  
Ernest Zakirov ◽  
Vladislav Arekhov ◽  
Atlas Ahmetzyanov

Sign in / Sign up

Export Citation Format

Share Document