TARGETED MEASUREMENTS FOR FORMATION EVALUATION AND RESERVOIR CHARACTERIZATION

Abstract

A downhole-reconfigurable tool string is utilized to obtain targeted measurements for formation evaluation and reservoir characterization. Tool string hardware and corresponding analysis software are dynamically adjusted to meet specific reservoir description needs identified during or before the logging run with the downhole-reconfigurable tool string. A technique for utilizing the downhole-reconfigurable tool string includes four different steps. The first step includes detecting first order features and identifying second order features. The second step includes enhancing measurements relative to the second order features by reconfiguring tool hardware to extract at least the desired information of the reservoir in terms of its static and dynamic properties, and software focusing to enhance the sensitivity to a particular reservoir attribute. The third step is performing an inversion to reconstruct the reservoir property as accurately as possible. The fourth step is using these properties and the planned production/injection data to recommend suitable monitoring schema.

Description

FIELD OF THE INVENTION

The invention is generally related to oil and gas wells, and more particularly to a downhole-reconfigurable tool string and analyzer unit which facilitate gathering targeted measurements for formation evaluation and reservoir characterization.

BACKGROUND OF THE INVENTION

Wireline logging tools are used to measure physical, chemical, and structural characteristics of formations surrounding a borehole. For example, data gathered by logging tools can be used to interpret formation stratigraphy, lithology, and mineralogy. An individual wireline logging tool measures physical properties of a formation and may be divided into different sections that are assembled at the wellsite. These sections include cartridges and sondes. The sonde is the section of the logging tool that contains the measurement sensors. The cartridge contains the associated electronics and power supplies.

In order to prepare for a logging run, the logging tool is first lowered into the borehole on a wireline cable. Measurements are then obtained as the tool is pulled back toward the surface. Multiple logging runs are made in some boreholes to improve coverage, confirm the accuracy of logged data and monitor progressive changes in the formation. However, because of the expense associated with the duration of logging operations, particularly in the case of offshore boreholes, it is desirable to minimize the amount of time required to obtain the necessary data.

In order to reduce the number of logging runs, and also because data interpretation is often based on multiple properties, logging tools are often joined together in a “tool string.” The tool string permits multiple properties to be measured during a single logging run. Flexible joints are added in long tool strings to ease passage in the borehole, and to allow different sections to be centralized or eccentralized. If the total length of the tool string is very long, it may be necessary to make two or more logging runs with shorter tool strings. However, the additional time required may add considerably to the expense of the logging operation.

The current state of the art in commercial borehole logging is modular tool strings. Modular tool strings permit a particular hardware configuration to be selected at the surface before commencement of logging operations. One aspect of hardware configuration is tool selection. In particular, a subset of tools is selected from an available suite of tools based on expected environmental and formation characteristics. An example of tool selection based on expected environmental and formation characteristics is described in Griffiths, R., Barber, T. and Faivre, O., 2000 Optimal evaluation of formation resistivities using array induction and array laterolog tools, SPWLA 41^st logging symposium, in which an array resistivity tool is selected rather than an array induction tool for a conductive environment. Similarly, with formation testers, either a probe or packer module is selected depending upon permeability. Some multi-probe tools used in formation testing also permit the distance between sink and the observation probes to be selected at the surface prior to logging. Similarly, a sonic logging tool may be preprogrammed for the frequency bandwidth for a fast or a slow formation. Nevertheless, further improvements in logging tools and techniques are desirable, because of the limited flexibility with these choices.

SUMMARY OF THE INVENTION

The present invention is predicated in-part on recognition that surface-configurable, suit-for-purpose logging tool hardware cannot always be properly configured a priori because of a lack of sufficient information about the geological formations needed to select the appropriate hardware configuration. Further, regardless of the extent to which a proper hardware adaptation has been made, it is desirable to complement the acquisition with algorithmic processing and inversion to enhance the characterization of a desired formation attribute.

In accordance with one embodiment of the invention, a method of obtaining targeted measurements from a logging tool in a borehole comprises the steps of: identifying at least one second order feature associated with the formation; calculating a logging tool sensor configuration for the identified second order feature; adjusting the logging tool to achieve the calculated sensor configuration, while the logging tool is in the borehole; and logging the identified second order feature with the logging tool.

In accordance with another embodiment of the invention, apparatus for obtaining targeted measurements from a borehole environment comprises: a logging tool operable within the borehole environment in response to data or signaling to adopt a specified sensor configuration; and an analyzer unit operable to identify at least one second order feature associated with the formation, calculate a logging tool sensor configuration for the identified second order feature, and signal to the logging tool, thereby prompting the logging tool to adjust to achieve the calculated sensor configuration, while the logging tool is in the borehole, and log the identified second order feature.

One advantage of at least one embodiment of the invention is reduction in time required for logging. The reduction in time is generally provided by reducing the number of logging runs required to achieve a desired result, which is accomplished by reconfiguring the logging tool in the borehole to log different second order features, rather than configuring a tool at the surface for each logging run, and executing a new logging run for each configuration.

Further features and advantages of the invention will become more readily apparent from the following detailed description when taken in conjunction with the accompanying Drawing.

BRIEF DESCRIPTION OF THE DRAWING

FIG. 1 illustrates a downhole-reconfigurable tool string which facilitates targeted measurements for formation evaluation and reservoir characterization.

FIG. 2 is a flow diagram illustrating formation evaluation and reservoir characterization with the downhole-reconfigurable tool string.

FIG. 3 illustrates an embodiment of the downhole-reconfigurable tool string in which tool spacing is adjusted by physically moving tool components.

FIG. 4 illustrates an embodiment of the downhole-reconfigurable tool string in which tool spacing is adjusted by enabling and disabling tool components.

DETAILED DESCRIPTION

Referring to FIGS. 1 and 2, a downhole-reconfigurable tool string (100) is utilized to measure physical, chemical, and structural characteristics of formations surrounding a borehole (101). The tool string operates under the control of an analyzer unit (102) which may be disposed at the surface. The analyzer unit is also capable of data analysis with reference to a reservoir model maintained in a memory. A wireline cable (104) connects the tool string (100) to the analyzer unit (102). The downhole-reconfigurable tool string is lowered into the borehole to measure physical properties associated with the formation, which typically includes a reservoir (106) adjacent to an impermeable layer (108), and various other layers which make up the overburden (110). Data gathered by the tool may be communicated to the analyzer unit in real time via the wireline cable. Unlike conventional modular tool strings, the downhole-reconfigurable tool string (100) and analyzer unit (102) can be both hardware and software reconfigured while in the borehole, with little or no interruption of logging operations. This innovative feature facilitates gathering of targeted measurements, which in turn are utilized for formation evaluation and reservoir characterization.

The initial step in obtaining targeted measurements is feature detection (200). In the feature detection step the tool string and analyzer unit, or some other devices, are utilized to detect first order features, as shown in step (202), and identify second order features, as shown in step (204). The first order features are major features of the borehole, such as layering adjacent to the borehole, changes in lithology or facies, etc. These first order features may be detected in either a single step or multiple steps. In the single step approach, a priori knowledge is used to select the logging tool suite for the downhole-reconfigurable tool string, and to decide on the most useful set of measurements to be made. In the multi-step approach, an initial logging run may be executed with either a simple, fixed configuration logging tool or the downhole-reconfigurable tool string in a basic configuration. Data gathered from the initial logging run is then used to select a more comprehensive logging tool suite for the downhole-reconfigurable tool string, as will be described in greater detail below. Depending on conditions and required accuracy and resolution, the downhole-reconfigurable tool string may even detect the first order features while moving toward the bottom of the borehole in preparation for a more comprehensive logging run. In any case, the first-level description of the reservoir, a.k.a., the “background,” is processed by the analyzer unit to identify zones of interests within the formation, i.e., “second-order features.” Second order features include variations within a given facies or lithology, fractures, sealing or open faults, variations in filtrate invasion, microstructural variations, i.e., the extent of intergranular and intrgranular porosities and vugs, etc. The background is detected and then filtered to remove data in order to facilitate reconfiguration and operation of the tool for purposes of enhancing the sensitivity of the focused measurements to higher-order features.

Dynamic downhole configuration of the tool string and analyzer unit are executed in an enhancement step (206). In particular, the tool string and analyzer are configured based on data obtained from the detect step. By obtaining a large-scale description of the reservoir from the detect step, and removing its effect from the data, and furthermore by identifying the zones of interest, it is possible to reconfigure the tool string and analyzer unit to obtain measurements which are maximally sensitive to the second-order features of interest. The enhance step includes the following operations: hardware configuration (208) through adjustment of control parameters, and; software-focusing (210) for deliberately resolving any formation property within a specified zone, assuming that the implemented hardware has enough sensitivity to the formation property of interest. Downhole configuration can be utilized to configure the tool string for an entire logging run, or to configure the tool string multiple times during a single logging run. For example, the tool string may be uniquely configured for each individual second order feature being logged during a logging run.

In order to automate the enhancement operations, each identified zone of interest can be regarded as a perturbation relative to the background, thereby permitting linearization of the system of equations representing the measurements. Given a discrete number of measurements, {M_i, i=1, . . . ,m}, corresponding to various source and receiver (or probe) pairs [{ R, R,}_i, i=1, . . . , m], where R_r, R_s are their respective locations, we have the following representation:

∫d r′K_i( r′)P( r′)=M_i, i=1, . . . , m,   (1)

where P( r′) is a spatially dependent physical parameter (such as permeability, resistivity, porosity, etc.) perturbed over the background property that we wish to invert for or estimate from the collected data. {M_i, i=1, . . . ,m} are understood to be dependent on time, in general. {K_i( r′), i=1, . . . , m} is a set of kernels (or transfer functions) that describes the response of the medium, i.e., the reservoir, at the receiver/probe location, R_r, given a source located at R_s. Again, without loss of generality, we approximate the system to be linear since the parameter is regarded as a perturbation over a spatially varying background. In this case, the set of kernels {K_i( r′), i=1, . . . , m} describes the response in the background medium. In a general nonlinear system, the above equation corresponds to a linearization around a particular iteration in the estimation process. This approach applies to those systems for which such a response may be constructed.

In order to perform the software focusing operation, the goal is to design a linearly weighted combination of these measurements, {M_i, i=1, . . . ,m}, such that the outcome is maximally sensitive to the value of the parameter P at a certain reservoir location r. We refer to such combinations as “targeted measurements” in that they target a certain specified parameter at a particular location in the reservoir space. For the purpose of focusing the measurements onto a specific location, r, in space, we multiply the above measurement equation by w_iK*_i( r) and sum over all measurements, where the asterisk in K*_i( r) represents complex conjugation and {w_i, i=1, . . . , m} is a set of weights that will lead to the desired focusing. We then obtain:

$\int {\stackrel{\_}{r}}^{\prime }H\left(\stackrel{\_}{r},{\stackrel{\_}{r}}^{\prime }\right)P\left({\stackrel{\_}{r}}^{\prime }\right)=D\left(\stackrel{\_}{r}\right),\mathrm{where}H\left(\stackrel{\_}{r},{\stackrel{\_}{r}}^{\prime }\right)=\sum _{i=1}^mw_iK_i\left({\stackrel{\_}{r}}^{\prime }\right)K_i^{*}\left(\stackrel{\_}{r}\right)=H^{*}\left({\stackrel{\_}{r}}^{\prime },\stackrel{\_}{r}\right)$ $\mathrm{and}D\left(\stackrel{\_}{r}\right)=\sum _{i=1}^mw_iM_iK_i^{*}\left(\stackrel{\_}{r}\right).$

We choose the weights {w_i, i=1, . . . ,m} in such a way that H( r, r′) is highly peaked at r′= r such that H( r, r′) approximates a Dirac delta functional. That is: H( r, r′)≈h*( r)h( r′)δ( r- r′) ,where h( r) is a normalization function. By doing so, we have maximized the sensitivity of the combined measurement to the value of the parameter P at the location r in the reservoir. For {M_i, i=1, . . . ,m}, which may be time dependent, it is understood that we sum the measurements over time. For optimized H( r, r′) we obtain:

$P\left(\stackrel{\_}{r}\right)\approx \frac{D\left(\stackrel{\_}{r}\right)}{{h\left(\stackrel{\_}{r}\right)}^2}=\frac{1}{{h\left(\stackrel{\_}{r}\right)}^2}\sum _{i=1}^mw_iM_iK_i^{*}\left(\stackrel{\_}{r}\right).$

Hence, in doing so, we have focused the measurements (in software) to provide a direct estimate of P( r). This was achieved by designing the weights {w_i, i=1, . . . ,m} such that:

$\sum _{i=1}^mw_iK_i^{*}\left(\stackrel{\_}{r}\right)K_i\left({\stackrel{\_}{r}}^{\prime }\right)\approx h^{*}\left(\stackrel{\_}{r}\right)h\left({\stackrel{\_}{r}}^{\prime }\right)\delta \left(\stackrel{\_}{r}-{\stackrel{\_}{r}}^{\prime }\right),\mathrm{or}$ $\sum _{i=1}^mw_i{\Omega }_i^{*}\left(\stackrel{\_}{r}\right){\Omega }_i\left({\stackrel{\_}{r}}^{\prime }\right)\approx \delta \left(\stackrel{\_}{r}-{\stackrel{\_}{r}}^{\prime }\right),\mathrm{where}$ ${\Omega }_i\left(\stackrel{\_}{r}\right)=K_i\left(\stackrel{\_}{r}\right)/h\left(\stackrel{\_}{r}\right).$

This can be implemented in a least squares sense by minimizing the following cost function:

$C_{\Omega }\left(w_i\right)={\sum _{i=1}^mw_i{\Omega }_i\left({\stackrel{\_}{r}}^{\prime }\right){\Omega }_i^{*}\left(\stackrel{\_}{r}\right)-\delta \left(\stackrel{\_}{r}-{\stackrel{\_}{r}}^{\prime }\right)}^2$

or by minimizing the alternative cost function:

$C_{\theta }\left(w_i\right)={\sum _{i=1}^mw_i{\theta }_{\mathrm{ij}}{\theta }_{\mathrm{ik}}^{*}-{\delta }_{\mathrm{jk}}}^2,$

where θ_ij=∫d rΩ_i( r)φ_j( r) and {φ_j( r), j=1, . . . , N≧√{square root over (m)}} is any orthonormal set of functions: ∫d rφ_j( r)=δ_jk, where δ_jk is the Kronecker delta. Any of the above two cost functions can be used as a measure of how well the hardware is synthesized to achieve the desired focusing on the particular parameter of interest. The smaller the value of these cost functions, the better the focusing of the measurement and maximization of its sensitivity to the particular reservoir parameter of interest. In view of the focusing scheme described above, it will be appreciated that for hardware reconfiguration it would be preferable to construct the hardware such that the kernel approximates a delta functional as closely as possible. However, only an approximation is practical because of component constraints, so deviation from the delta functional will be accomplished through software. The weight functions are chosen such that the residual kernels when constructed as stated above approximate a delta functional. Hence, this may be viewed as a software-focusing enhancement to the partial realization of the hardware implementation. Thus, both hardware and software focusing are achieved. Clearly, each measurement type will have its own optimal set of control parameters.

Referring now to FIGS. 1 through 4, hardware reconfiguration (208) may be accomplished by adjustment of one or more control parameters. For example, and without limitation, the control parameters may include: (1) number of sources/sinks and receivers/observers; (2) source/sink-receiver/observer spacing; (3) frequencies of operation or testing protocol; (4) vector components or polarizations; (5) data sampling; and (6) linear combinations thereof. By being able to reprogram these parameters downhole, it is possible to dynamically adjust the depth of investigation, the sensing volume, the azimuthal directionality, the sensitivity to anisotropy, and the lateral and longitudinal resolutions. Selection of new excitation frequencies and sampling rates of logging tools can be accomplished by the analyzer unit with reference to a database in view of the identified second order features. Other parameters may be adjusted as described below.

For adjusting source-receiver spacing, in an embodiment illustrated in FIG. 3 a tool string variant (100a) is equipped with slidably moving sensor components (300), e.g., transmitters and receivers, which are operative in response to signaling from the analyzer unit. In particular, the transmitters and receivers can be repositioned relative to one another downhole in order to achieve better placement of these sensors along the borehole for the purpose of enhancing the sensitivity of the measurements to the formation properties of interest. For formation testing, this will imply a flexible line with an internal elevator for probes, such as described in U.S. Pat. No. 5,195,588, which is incorporated by reference. Packers could also be mounted across tubing, the open interval of which may be expanded or shrunk, e.g., with a screw driven inside of a shaft within the tubing and a flexible hose for inflating and deflating the packers.

In an alternative embodiment illustrated in FIG. 4, a tool string variant (100b) is equipped with redundant arrays (400a-400f) of fixed-position sensor components which are operative in response to signaling from the analyzer unit. In particular, individual transmitters and receivers in each array can be selectively activated and deactivated downhole in order to achieve better separation and placement of the active sensor components along the borehole for the purpose of enhancing the sensitivity of the measurements to the formation properties of interest. For example, a sensor pair configured initially as transmitter (402) and receiver (404) could be reconfigured as transmitter (402) and receiver (406). Further, combinations of arrays may be selectively activated to achieve a desired number of sources/sinks and receivers/observers. The number and layout of arrays, and the number and layout of components in individual arrays, are implementation details which may depend on intended use, and the illustrated example is not intended to be limiting in those respects.

Referring again to FIGS. 1 and 2, the next step in obtaining targeted measurements is reconstruction (212). Having determined a qualitative estimation of a particular reservoir property from the enhancement step, a nonlinear inversion (214) is executed to reconstruct the reservoir properties to achieve a greater degree of accuracy. In particular, the analyzer unit is operable to compare the measured data obtained by the downhole-reconfigurable tool string with data produced by simulation with a reservoir model. The analyzer unit then operates to reduce or minimize the difference between the measured data and simulation data. The match between the measured and simulated data can be accomplished by adjusting the reservoir model parameters associated with the simulation in order to arrive at an approximate match. Possible inversion methods include, but are not limited to, deterministic (least squares) and probabilistic (Bayesian).

A final, optional, step is monitoring (216). The results of the detect, enhance and reconstruct steps provide a good description of the reservoir for both geometry (structure) and physical/chemical properties. With this knowledge, a permanent monitoring sensor array, e.g., resistivity, acoustic, pressure, temperature, gravity, etc., and its location could be calculated. The permanent monitoring sensor array may be mounted as a part of the completion. Further, the permanent sensor array may be made partly reconfigurable as shown in step (218) in FIG. 2. For example, although the sensor locations may be fixed if the array is mounted rigidly within cement, it may still be possible to change the excitation frequencies and sampling rates of the measurement, which is valuable if one is interested in maximizing information content of the data acquired during fluid movement. Still further, certain arrays may be deployed in a semi-permanent mode. An example would be a resistivity array with flexible spacing run by a wireline within a cased or open borehole.

While the invention is described through the above exemplary embodiments, it will be understood by those of ordinary skill in the art that modification to and variation of the illustrated embodiments may be made without departing from the inventive concepts herein disclosed. Moreover, while the preferred embodiments are described in connection with various illustrative structures, one skilled in the art will recognize that the system may be embodied using a variety of specific structures. Accordingly, the invention should not be viewed as limited except by the scope and spirit of the appended claims.

Claims

1. A method of obtaining targeted measurements from a logging tool in a borehole environment comprising the steps of:

identifying at least one second order feature associated with the formation environment;

calculating a sensor configuration for the logging tool which will enhance logging of the identified second order feature;

adjusting the logging tool to achieve the calculated sensor configuration, while the logging tool is in the borehole; and

logging the identified second order feature with the logging tool.

2. The method of claim 1 wherein the logging tool includes sensor hardware and wherein calculating the tool sensor configuration includes calculating a weighted combination of different sensor hardware measurements to achieve a predetermined level of sensitivity to the identified second order feature, and employing a cost function to quantify the level of sensitivity to the identified second order feature.

3. The method of claim 1 wherein adjusting the logging tool includes adjusting at least one control parameter selected from the group including: number of sources/sinks; number of receivers/observers; source/sink spacing; receiver/observer spacing; frequencies of operation; testing protocol; vector components; polarization; data sampling; and linear combinations thereof.

4. The method of claim 1 including the further step of performing a nonlinear inversion to reconstruct properties to achieve a greater degree of accuracy.

5. The method of claim 4 including the further step of performing a nonlinear inversion by comparing measured data obtained by the logging tool with data produced by simulation with a model, and adjusting model parameters in order to arrive at an approximate match.

6. The method of claim 1 including the further step of detecting first order features associated with the formation.

7. The method of claim 1 including the further step of calculating a logging tool sensor configuration for a different second order feature.

8. The method of claim 7 including the further step of readjusting the logging tool to achieve the sensor configuration calculated for the different feature while the logging tool is in the borehole.

9. The method of claim 1 wherein adjusting the logging tool includes the step of physically moving at least one sensor component.

10. The method of claim 1 wherein adjusting the logging tool includes the step of selectively activating at least one sensor component from an array of redundant components disposed at different locations on the logging tool.

11. Apparatus for obtaining targeted measurements from a borehole environment comprising:

a logging tool operable within the borehole environment in response to signaling to adopt a specified sensor configuration; and

an analyzer unit operable to identify at least one second order feature associated with the formation, calculate a logging tool sensor configuration to enhance logging of the identified second order feature, and signal to the logging tool, thereby prompting the logging tool to adjust to achieve the calculated sensor configuration, while the logging tool is in the borehole, and log the identified second order feature.

12. The apparatus of claim 11 wherein the logging tool includes sensor hardware and wherein the analyzer unit is further operable to calculate a weighted combination of different sensor hardware measurements to achieve a predetermined level of sensitivity to the identified second order feature, and employ a cost function to quantify the level of sensitivity to the identified second order feature.

13. The apparatus of claim 11 wherein the logging tool is operable to adjust at least one control parameter selected from the group including: number of sources/sinks; number of receivers/observers; source/sink spacing; receiver/observer spacing; frequencies of operation; testing protocol; vector components; polarization; data sampling; and linear combinations thereof.

14. The apparatus of claim 11 wherein the analyzer unit is operable to perform a nonlinear inversion to reconstruct properties to achieve a greater degree of accuracy.

15. The apparatus of claim 14 wherein the analyzer unit performs a nonlinear inversion by comparing measured data obtained by the logging tool with data produced by simulation with a model, and adjusting model parameters in order to arrive at an approximate match.

16. The apparatus of claim 11 wherein the logging tool is operable to detect first order features associated with the formation.

17. The apparatus of claim 11 wherein the analyzer unit is further operable to calculate a logging tool sensor configuration for a different second order feature.

18. The apparatus of claim 17 wherein the analyzer unit is operable to prompt the logging tool to readjust to achieve the sensor configuration calculated for the different feature while the logging tool is in the borehole.

19. The apparatus of claim 11 wherein the logging tool is operable to physically move at least one sensor component.

20. The apparatus of claim 11 wherein the logging tool is operable to selectively activate at least one sensor component from an array of redundant components disposed at different locations on the logging tool.