EP2568110B1 - Exploitation method of an oilfield on the basis of a selection technic of to be drilled borehole positions - Google Patents

Description

The present invention relates to the technical field of the petroleum industry, and more particularly to the operation of underground reservoirs, such as petroleum reservoirs or gas storage sites.

In particular, the invention makes it possible to effectively plan the development of a reservoir by selecting the positions where to drill new wells, for which the production potential will be maximum.

The optimization and exploitation of oil deposits is based on as precise a description as possible of the structure, the petrophysical properties, the properties of the fluids, etc., of the deposit studied. To do this, the specialists use a tool that allows these aspects to be accounted for in an approximate way: the reservoir model. Such a model is a model of the subsoil, representative of both its structure and its behavior. Generally, this type of model is represented on a computer, and one speaks then of numerical model. A reservoir model comprises a mesh or grid, generally three-dimensional, associated with one or more petrophysical property maps (porosity, permeability, saturation, etc.). The association consists of assigning values of these petrophysical properties to each grid cell.

To be considered reliable, the reservoir model should verify as much as possible all the data collected in the field: the logging data measured along the wells, the measurements made on rock samples taken from the wells, the data deduced seismic acquisition campaigns, production data such as oil flow, water flow, pressure flow ... These data are insufficient to precisely characterize the values of the petrophysical properties to be attributed to the meshes of the model. This is why we usually resort to a stochastic formalism. Petrophysical properties are considered as realizations of random functions. We then generate a possible image of the reservoir, ie a model, from geostatistical simulation techniques. The resolution of the flow equations for this model provides answers in production. These responses are then compared to the production data measured in the wells. To increase the predictivity of the reservoir model, the discrepancy between the simulated responses and the data acquired in the field must be minimized. This step goes through a calibration or optimization process. The latter is generally very expensive in computation time, because it is iterative and requires a flow simulation by iteration. However, a single flow simulation often involves a few hours of computation time.

When a model respecting the data measured in the field is finally obtained, it is used to predict the movements of fluid in the reservoir and to plan the future development of the field. For example, for mature fields, it is necessary to be able to select the zones where to drill new wells, either to produce the oil by depletion, or to inject a fluid which maintains the pressure at a sufficient level in the reservoir. To appreciate the performance of a well at a point, one can rely on the reservoir model, position the well in the desired position and perform a flow simulation. The performance of a well is estimated from the amount of hydrocarbon it produces. The ultimate goal is to maximize the production or profitability of the field, it should be possible to test all possible positions and thus select the best of them. Such an approach is inappropriate in practice because it consumes too much computing time. An alternative is to start an optimization process to place a well as best as possible to optimize production. However, this approach is difficult to implement because it requires a few thousand iterations.

The concept of production indicator map, also referred to in the quality map literature, was introduced to provide a practical answer to the problem of placing new wells in a reservoir. It is a two-dimensional map, comprising a set of meshes, where each mesh is associated with a real value that shows how a new well placed in the mesh in question impacts the production or the net present value (NPV) relative to to one compared to the base case. The basic case corresponds to the initial operating scheme, ie here a scheme for which no new wells are added. ( Da Cruz, PS, Horne, RN, Deutsch, C., The Quality map: A tool for reservoir quantification and decision making, SPE ATCE, SPE 56578, Houston, TX, USA, 1999 ). A production indicator defines an impact on the production of the fluid (hydrocarbon) linked to the addition of a well in the mesh in question.

To build this map, one can make a flow simulation for each mesh where it is possible to place a well. If the reservoir comprises NX and NY meshes along the X and Y axes, the total number of meshes to be examined is NX × NY to which we subtract the numbers of non-active meshes and meshes in which we already have a well for the case of based. This approach requires significant computation time as soon as NX × NY is important. In addition, the possible meshes being considered one after the other, the interference between the new wells are not taken into account.

To reduce calculation times, an interpolation approach has been envisaged ( Cottini-Loureiro, A., Araujo, M., SPE ATCE, SPE ATCE, Dallas, TX, US, 9-12 October 2005 ). We then make a simulation for some meshes of the map, the values in the other meshes are estimated by interpolation. However, this approach does not account for interference between the new wells.

The map of production indicators quantifies for each mesh the impact on a production indicator due to the addition of a well in this mesh. It only takes into account a single well. To add several wells and to take into account the interferences between these wells, it was suggested to follow a sequential approach. The wells are added one after the other. Each time a well is added, the quality map is updated in the region encompassing the selected position. A flow simulation is made for each of the meshes of the region in question ( Cheng, Y., McVay, DA, Lee, WJ, A Practical Approach to Optimizing Inflammable Fuel Storage, Journal of Natural Gas Science and Engineering, 1, 165-176, 2005 ). This solution requires many simulations and therefore requires a large calculation time.

None of the processes developed therefore proposes a solution that both gives accurate results in a reduced calculation time and takes into account the interference with the added wells.

Thus, the subject of the invention relates to an alternative method for operating a petroleum deposit from a reservoir model. This alternative method is based on the construction of the production indicator map, comprising a set of meshes, for which certain production indicators are determined by interpolation, the interpolation method chosen being dependent on the distance between the mesh considered and the well nearest to said mesh considered. This method also makes it possible to update said production indicator map when wells are added sequentially in the reservoir model, without the need for new simulations. Therefore, thanks to this method, interferences between wells are taken into account, and this, in a limited calculation time.

The process according to the invention

• on construit ladite carte au moyen des étapes suivantes :
  1. a) on sélectionne des mailles parmi l'ensemble de mailles de ladite carte ;
  2. b) on détermine des indicateurs de production aux mailles sélectionnées ;
  3. c) on interpole lesdits indicateurs de production déterminés à l'étape b) sur l'ensemble des mailles de ladite carte, au moyen d'un modèle d'interpolation prenant en compte une distance entre la maille à interpoler et le puits le plus proche de ladite maille à interpoler ; et
• on définit la position dudit second puits par la maille où ledit indicateur de production est maximal.

The invention relates to a computer-implemented method for operating an underground reservoir, in particular a petroleum reservoir, through which at least a first well from which a fluid is produced is produced, in which a position of at least one second well to be drilled using a production indicator card comprising a set of meshes, each mesh being associated with a production indicator defining an impact on the production of the fluid of an addition of a well in this mesh. The method comprises the following steps:

• said card is constructed by the following steps:
  1. a) selecting meshes from the set of meshes of said card;
  2. b) production indicators with the selected meshes are determined;
  3. c) interpolating said production indicators determined in step b) on all the cells of said card, by means of an interpolation model taking into account a distance between the mesh to interpolate and the nearest well said mesh to be interpolated; and
• defining the position of said second well by the mesh where said production indicator is maximum.

In one embodiment, the production indicator measures a variation of parameters impacting the production of the fluid when adding a well in the mesh.

Preferably, the production indicator is an increment of fluid volume produced by placing a well in the mesh or a change in the expected net value.

According to an advantageous embodiment, the selection of the stitches is carried out by sampling.

1. i. on détermine des attributs du réservoir ;
2. ii. on construit une carte d'identification des régions par une classification des attributs ; et
3. iii. on sélectionne lesdites mailles en fonction de ladite carte d'identification de régions.

Advantageously, the meshes are selected by performing the following steps:

1. i. tank attributes are determined;
2. ii. an identification map of the regions is constructed by a classification of the attributes; and
3. iii. said meshes are selected according to said region identification map.

In one embodiment, the attributes of the reservoir used are chosen from the following attributes: the distance between each mesh and the well closest to said mesh; dynamic data, such as pressure and connected fluid volume, seismic data such as velocities and densities.

According to a preferred embodiment, the classification method is the K-means algorithm.

Preferably, steps c) and defining the position of the second well are repeated for determining a position of at least one other well, taking into account the impact of adding one or more well on the distance between a mesh and the well nearest said mesh.

Advantageously, the interpolation model used in step e) is a polynomial interpolation model, preferably of order 2, or a kriging interpolation model, or a combination of a polynomial interpolation model and of a kriging interpolation model.

The invention also relates to a computer program product downloadable from a communication network and / or recorded on a computer readable medium and / or executable by a processor, wherein it comprises program code instructions for implementing the method as defined above, when said program is run on a computer.

In addition, the invention relates to a method as defined above, in which exploratory drilling is carried out at said determined positions.

Other features and advantages of the method according to the invention will appear on reading the following description of nonlimiting examples of embodiments, with reference to the appended figures and described below.

Brief presentation of the figures

• The figure 1 illustrates a map of reference production indicators.
• The figure 2 illustrates a region identification map made from an attribute classification.
• The figure 3 illustrates several maps. The Figures 3.1 to 3.4) represent the update of the minimum distance map. The Figures 3.5 to 3.8) in the middle represent the updating of the production indicator map. The Figures 3.9 to 3.12 ) represent the position of the added well (isolated dark position). The figures 3.1) , 3.5) , 3.9) represent existing wells initially, ie 6 producers. The figures 3.2) , 3.6) , 3.10) represent the previous case to which an injector has been added. The figures 3.3) , 3.7) , 3.11) represent the previous case to which a second injector has been added. The figures 3.4) , 3.8) , 3.12) represent the previous case to which a third injector has been added.

Detailed description of the process

The process according to the invention makes it possible to efficiently exploit a petroleum deposit. The method successively selects areas where it is interesting to put a new well, producer or injector, to improve the productivity of the reservoir. It relies on the construction of a map of production indicators ( Figure 1 ) taking into account the interferences between the wells.

1. 1) on construit la carte d'indicateurs de production
   1. a) on sélectionne des mailles parmi l'ensemble de mailles de ladite carte,
   2. b) on détermine des indicateurs de production aux mailles sélectionnées,
   3. c) on réalise les étapes suivantes pour déterminer les indicateurs de production à l'ensemble des mailles de la carte :
      1. i. on définit un modèle d'interpolation prenant en compte la distance entre la maille à interpoler et le puits le plus proche et on estime les paramètres de ce modèle d'interpolation à partir desdits indicateurs de production déterminés à l'étape b),
      2. ii. on interpole lesdits indicateurs de production sur l'ensemble des mailles de ladite carte, au moyen du modèle d'interpolation et des paramètres spécifiés à l'étape c) i., et
2. 2) on définit la position dudit nouveau puits par la maille où ledit indicateur de production est maximal.

The method according to the invention comprises the steps listed below:

1. 1) build the production indicator map
   1. a) selecting meshes from the set of meshes of said card,
   2. (b) production indicators with selected meshes are determined,
   3. c) the following steps are carried out to determine the production indicators at all the cells of the map:
      1. i. defining an interpolation model taking into account the distance between the mesh to interpolate and the nearest well, and estimating the parameters of this interpolation model from said production indicators determined in step b),
      2. ii. said production indicators are interpolated on all the meshes of said map, by means of the interpolation model and the parameters specified in step c) i., and
2. 2) defining the position of said new well by the mesh where said production indicator is maximum.

Step 1) Construction of the production indicator map

This production indicator card comprises a set of meshes, each mesh being associated with a production indicator (IP). A production indicator (IP) quantifies an impact on the production of the fluid due to the addition of a well in this mesh. The production indicator (PI) measures a variation of the parameters affecting the production of the fluid when adding a well in the mesh. This production indicator (PI) may include a change in total production of all wells, a change in the expected net present value, a variation in pressure or flow. In one embodiment, the production indicator (IP) is the increment of volume of oil produced by placing a well, for example an injector well, in this mesh.

a) Mesh selection (Figure 2)

Advantageously, we select meshes of the map to estimate from a sampling technique, which can be fully computerized, or computerized, then completed manually, or performed entirely manually. For example, said sampling technique is a Latin hypercube, based on a "Maximin" criterion, which makes it possible to divide the space into equiprobable subspaces sampled in a uniform manner.

According to a preferred embodiment, a region identification map, previously prepared from attributes, is used. The process according to the invention makes it possible to efficiently exploit a petroleum deposit, for which a set of properties (petrophysical or seismic) such as permeability, porosity, saturations, etc. is known. These attributes of the reservoir, which can be measured, simulated or calculated, are called geological data, geometric data, seismic data and dynamic data such as: the pressure at the time before the addition of new wells, connected fluid volume, minimum distance from existing wells, average permeability, porosity, velocities, density ...

Attributes characterizing the reservoir being known, a classification method is applied, for analysis and separation into classes. We deduce a two-dimensional map, called an identification map of the regions, distinguishing regions for which the attributes belong to the same class. Meshes belonging to the same region are therefore characterized by similar or similar attributes. It is advantageous to use attributes because they require only negligible computation time.

In a preferred embodiment, the classification is done according to the K-means algorithm, which makes it possible to group the attributes in K classes that do not overlap. We choose a number of classes (or coefficient K), generally less than 10, in order to obtain a relatively stable result. This algorithm has the advantages of conceptual simplicity, speed of execution and low memory size requirements.

The figure 2 represents an example of map obtained by application of this method. Since the number of classes is five, there are five different attribute regions. The position of existing wells is indicated by white squares. Meshes selected by sampling are represented by a black circle and those manually added by a black dot.

The region identification map being established, it can guide the mesh selection process. It is therefore advantageous to superimpose the selected meshes on the identification card of the regions. If a class identified during the creation of the identification map of the regions is considered a priori interesting by a specialist, but includes few meshes selected, additional meshes are manually selected by said specialist. A relevant choice of the selected meshes, in particular from the identification map of the regions, makes it possible to build a map of production indicators that is more precise and more reliable.

b) Determination of production indicators with selected meshes

For each mesh selected in step a), the production indicator (IP) is determined either by measurement, by calculation or by simulation.

Preferably, a simulation of fluid flow contained in the reservoir to the producing wells is carried out for each selected mesh, on the assumption that a well is added in said selected mesh. Therefore, if in step b) of selection, N meshes are retained, one runs N flow simulations with for each of them a single well added to the considered position. These simulations give the exact value of the production indicator (IP1, IP2 ... IPN) for the selected meshes. Thanks to the invention, measurements, calculations or flow simulations are carried out only for the selected meshes. To execute a flow simulation, it is known to those skilled in the art to use software called flow simulators such as Pumaflow® (IFP Energies nouvelles, France).

c) Determination of production indicators on all the meshes of the map i. Definition of the interpolation model and its parameters

In order not to have to determine the production indicators on all the cells of the card from a time-consuming process such as a flow simulation, and consequently to reduce the calculation time, the indicator of production is estimated by interpolation on the set of unsampled meshes of the map. In order to take into account the interference with the wells, the interpolation model is constructed from a group of regressors comprising an attribute that depends on the distance between the mesh to be interpolated and the well nearest to said interpolated mesh. . This well may be an existing well or an already added well.

avec IP l'indicateur de production à la maille considérée, et IP1, IP2, IPN, les indicateurs de production connus (aux mailles sélectionnées et aux puits).In one embodiment, a polynomial interpolation model or a kriging interpolation model can be used. For these models, the meshes are characterized by the values of the regressors associated with them, for example their spatial coordinates x and y and the distance between the mesh to be interpolated and the well nearest to said mesh to be interpolated. This last regressor is introduced to take into account the interference between the added wells. The indicator of production in a mesh can thus be expressed by the following formula: $$\mathit{IP}\left(x,\mathrm{there}\right)=f\left(x,\mathrm{there},D\min ,\mathrm{}\mathit{<figure-callout\; id="1"\; label="IP"\; filenames="imgf0001.png"\; state="\{\{state\}\}">IP</figure-callout>}1\mathit{<figure-callout\; id="2"\; label="IP"\; filenames="imgf0001.png,imgf0002.png"\; state="\{\{state\}\}">IP</figure-callout>}2...,\mathit{IPN}\right)$$

with IP the production indicator at the mesh considered, and IP1, IP2, IPN, the known production indicators (with the selected meshes and the wells).

In addition, the interpolation model depends on model construction parameters, which must be adjusted to the studied reservoir. To achieve this adjustment, the values of the production indicators obtained in step b) are used. Indeed, this is made possible because at the selected meshes, only these construction parameters are unknown.

ii. Interpolation (Figure 3.5) to 3.8))

The parameters of the interpolation model having been estimated in step c) i., The production indicators with unselected meshes of the card are interpolated. The estimation of production indicators by interpolation makes it possible to dispense with an ancillary simulator and to reduce calculation times.

The Figures 3.5 to 3.8) show examples of production indicator maps for an example, each case corresponding to a different initial configuration of wells constructed for a growing number of wells. The production indicator chosen is the increment of volume of oil produced. The black areas correspond to the areas where the production indicator is minimal and the dark gray areas correspond to the areas where the production indicator is maximum.

In addition, this card has the advantage of being able to be updated to integrate the influence of successively added wells without having to restart new flow simulations. The figure 1 represents an example of a map of production indicators. The value in one position corresponds to the production indicator, it is here, in relative value (%), the increment of volume of oil produced by placing an injector well at this position. White squares indicate existing wells.

Step 2) Positioning a new well (Figure 3.9) to 3.12)

The maximum value of the production indicator thus constructed corresponds to the mesh where it will be most advantageous to position a well. A well is then added to the production scheme and integrated into the group of existing wells. The well can then be drilled later.

The Figures 3.9) to 3.12) represent an example of successive positioning of wells. The added well is represented by the mesh colored in black.

Addition of additional wells

In order to define the optimal location of at least one other new well, the minimum distances to the nearest well are updated beforehand for each cell of the card. This update takes into account the well that has just been added. Indeed, a well having been added to the group of existing wells, it is necessary to recalculate for each of the meshes the distance separating it from the existing or simulated nearest well. This gives a discounted minimum distance map, as shown for an example at Figures 3.1 to 3.4) . In fact, the coordinates of the cells of the card are modified. It is recalled that these coordinates are x , y and the distance from the mesh considered to the nearest existing or simulated well. Current production indicators are therefore out of date.

Step c) ii is then repeated, which results in the updating of the production indicator card. It should be noted that the values of the production indicators determined in step b) before the addition of the first well are retained for the selected meshes, except for the meshes selected for which the distance to the nearest well has varied. Taking in counting the distance to the nearest well in the interpolation process naturally leads to a decrease in the production indicators of these meshes.

The map of production indicators having been updated, step 2) is repeated for defining the position of a new well.

This procedure is repeated as long as you want to add a well.

Thus, thanks to the invention, the positioning of the new well is a parameter that is taken into account in the determination of the production indicators. Therefore, interferences between the wells are taken into account. In addition, step b) of determining the production indicators with the selected meshes and step c) i. the definition of the interpolation model and its parameters are not repeated, which brings a saving in process time. This gain is significant, especially when the number of meshes sampled is important and when step b) uses a flow simulator to determine the production indicators meshes sampled.

Application example

To illustrate the process, we use a test case developed as part of the European project "Production forecasting with UNcertainty Quantification" from a real oil reservoir. The field contains oil and gas. It is produced from 6 producing wells located near the line of contact between oil and gas. The basic production scheme covers the period from 01/01/1967 to 15/01/1975. The wells are then closed for three years before being put into production at the imposed rate for the last four years. At the end of eight years, there is the question of adding water injection wells to support the pressure in the tank. It is assumed that from 15/01/75 to 15/01/80, production is driven by the six producing wells and the injection wells. The problem is to identify the most strategic positions for the implantation of injection wells.

It is then a matter of constructing, using the method according to the invention, a map of production indicators (IP), and to deduce the position of the wells to be added while updating it as and when.

The reservoir model is discretized on a grid of 19 × 28 × 5 meshes, of which 1761 are active. This configuration leads us to build a map of production indicators on a grid of 19 × 28 meshes, of which 396 can accommodate a new well. The base case corresponds to the cumulated oil volume produced by the six producing wells at 15/01/80 in the absence of any injection wells. The production indicator assigned to a mesh of the production indicator card corresponds to the quantity of oil produced in addition when an injection well is placed in the mesh in question.

A flow simulation for the PUNQ case requires a very short computation time. In these very particular conditions, it is quite possible to perform a flow simulation for all possible meshes, which gives access to the exact map of production indicators (PI) ( Figure 1 ).

Several attributes were determined for the test case, including the pressure and oil volume connected on 15/01/75 as well as the average connected permeability. The K-means algorithm is then applied to identify regions. Five classes are considered for the example studied. The identification card of the resulting regions is carried over to the Figure 2 . At this stage, it is difficult to estimate the regions' interest in terms of performance or profitability. Indices are however provided by the analysis of the attributes. For example, an area where pressure is strong is probably favorable to the establishment of a new well. It is also preferable to put a well in a mesh where the connected volume of oil is important, where the permeabilities are high, in a mesh sufficiently distant from existing wells, etc. In fact, classes 1 and 4 are likely to have interesting potential for drilling new wells, unlike class 5.

We then select grid cells by sampling from a Latin hypercube based on a "Maximin" criterion. By identifying the selected meshes on the region identification map ( Figure 2 ), it is observed that two clusters of class 1, which is mainly represented and whose potential is a priori important for the specialist, are not sampled. The specialist then intervenes manually: the additional positions thus selected are indicated by black disks. 5 cells are selected in class 1 and 1 in class 5. Then a flow simulation is carried out with Pumaflow ® flow simulation means (IFP Energies nouvelles, France) with an injector well placed on each of the selected meshes , one after the other. We deduce the production indicator (IP1, IP2, ..., IPN) associated with these meshes whose coordinates are the spatial coordinates (X, Y) and the value of the distance (Dmin) which separates them from the existing well. the closest.

The minimum distance (Dmin) from the nearest wells (existing or simulated) is presented on the Figure 3.1) . The production indicators (IP) in the unselected meshes are then deduced from a kriging interpolation, the parameters of which have been determined beforehand from said flow simulations with the selected meshes. The Figure 3.5) shows the resulting map of production indicators. It is very close to the map of reference production indicators ( Figure 1 ), although it was constructed from 26 flow simulations instead of 396. We now define the position of the first well to be added by the mesh where the production indicator (quantity of oil produced) is maximal ( Figure 3.9) ), this being integrated into the group of existing wells. To place the next well, update the minimum distance map and then the production indicator map. This procedure is repeated as long as it is desired to add wells. The Figures 3.1) to 3.4) show the evolution of the minimum distance map with the successive addition of wells. The Figures 3.5 to 3.8) show the resulting evolution for the production indicator map. The Figures 3.9) to 3.12) show the selected position for the new well from the updated production indicator maps.

Claims (11)

1. A computer-implemented method of developing an underground reservoir, notably a petroleum reservoir, crossed by at least a first well from which a fluid is produced, wherein a position of at least a second well to be drilled is determined by means of a production indicators map comprising a set of cells, each cell being associated with a production indicator (IP) defining an impact on the fluid production of a well addition in this cell, characterized in that it comprises the following stages:

   - constructing said map by means of the following stages:

   a) selecting cells from the set of cells of said map

   b) determining production indicators (IP) in the cells selected

   c) interpolating said production indicators (IP) determined in stage b) on the set of cells of said map, by means of an interpolation model taking account of a distance between the cell to be interpolated and the well closest to said cell to be interpolated, and

   - defining the position of said second well by the cell where said production indicator is maximal.
2. A method as claimed in claim 1, wherein production indicator (IP) measures a parameter variation impacting the fluid production upon addition of a well in the cell.
3. A method as claimed in any one of the previous claims, wherein production indicator (IP) is a fluid volume increment produced by placing a well in the cell or a variation of the net value expected.
4. A method as claimed in any one of the previous claims, wherein selection of the cells is achieved by sampling.
5. A method as claimed in any one of the previous claims, wherein the cells are selected by carrying out the following stages:

   i. determining reservoir attributes

   ii. constructing a region identification map through attribute classification, and

   iii. selecting said cells as a function of said region identification map.
6. A method as claimed in claim 5, wherein the reservoir attributes used are selected from among the following attributes: the distance between each cell and the well closest to said cell ; dynamic data such as the fluid pressure and the connected fluid volume ; seismic data such as velocities and densities.
7. A method as claimed in any one of claims 5 and 6, wherein the classification method is the K-means algorithm.
8. A method as claimed in any one of the previous claims, wherein stage c) and the stage of defining the position of the second well are repeated for determination of a position of at least another well, by taking account of the impact linked with the addition of one or more wells on the distance between a cell and the well closest to said cell.
9. A method as claimed in any one of the previous claims, wherein the interpolation model used in stage c) is a polynomial interpolation model, preferably of second order, or a kriging interpolation model, or a combination of a polynomial interpolation model and of a kriging interpolation model.
10. A computer program product downloadable from a communication network and/or recorded on a computer readable medium and/or processor executable, comprising program code instructions for implementing the method as claimed in any one of the previous claims, when said program is executed on a computer.
11. A method as claimed in any one of claims 1 to 9, wherein exploratory drilling is carried out in said determined positions.

Images