法国专利FR3064755A1 AUTOMATED WELL DIAGRAPH CORRELATION USING DESCRIPTORS

专利PDF首页>>法国专利

专利附录

专利说明

权利要求

类似技术

同族专利

引用文献

法律状态

优先权

专利摘要:
Well logs can be automatically correlated using dynamic alignment with descriptors. For example, a computing device may receive a user input indicating a manual correlation between a first node in a first well log and a second node in a second well log. The computing device may use manual correlation to automatically determine a combination of descriptors that can be used to correlate the first node to the second node. The computing device may then perform dynamic alignment on other well logs using the combination of descriptors determined using the first well log and the second well log. This can provide more accurate correlations between well logs.
公开号:FR3064755A1
申请号:FR1850981
申请日:2018-02-06
公开日:2018-10-05
发明作者:Ahinoam POLLACK；Yang Peng；Kainan WANG；Ming-Kang SHIH；Krisha Hansel TRACY；Jesse Mathias Lomask
申请人:Landmark Graphics Corp；
IPC主号:

专利说明:

(54) AUTOMATED WELL LOGGING CORRELATION USING DESCRIPTORS.
FR 3 064 755 - A1 (5d) Well logs can be automatically correlated using dynamic alignment with descriptors. For example, a computing device may receive user input indicating a manual correlation between a first node in a first well log and a second node in a second well log. The computing device can use manual correlation to automatically determine a combination of descriptors that can be used to correlate the first node to the second node. The computing device can then perform dynamic alignment to other well logs using the combination of descriptors determined using the first well log and the second well log. This can provide more precise correlations between well logs.
2016-IPM-100867-U1-FR 1
CORRELATION OF AUTOMATED WELL LOGGING USING
DESCRIPTORS
Technical area
The present disclosure generally relates to methods and devices for well logging or borehole investigation. More specifically, but not limited to, this disclosure relates to an automated well log correlation using descriptors.
Context of the invention
A well system (eg, for oil or gas extraction) can include multiple wells drilled through an underground formation. Each wellbore may have a well logging tool which provides data in the form of a well logging back to a well operator. A well log can be an indicator record of underground strata that are penetrated to various depths by a wellbore. The well operator can manually review well logs to identify strata or other features of interest in the underground formation. For example, the well operator can manually review multiple well logs in a two-dimensional (2D) section view or a three-dimensional (3D) section view to identify structures or features of interest.
Brief description of the drawings
Figure 1 is a side sectional view of an example of a well system for obtaining well logs in certain aspects.
Figure 2 is a graph of an example of strata surfaces that have been correlated among multiple well logs via manual correlation and automated correlation in some respects.
FIG. 3A is a graph of an example of well logging before dynamic alignment according to certain aspects,
Figure 3B is a graph of an example of well logging after dynamic alignment in certain aspects.
Figure 4A is a graph of an example of a shape descriptor associated with a node in a well log according to certain aspects.
Figure 4B is a graph of an example of another descriptor of
2016-IPM-100867-U1-FR 2 form associated with another node in another well log according to certain aspects.
Figure 5 is a graph of an example of strata surfaces that have been correlated using dynamic alignment with descriptors in certain aspects,
Figure 6 is a flow diagram of an example of a process for automated well log correlation using descriptors in certain aspects.
Figure 7 is a graph of an example of manually correlated nodes in well logging in certain aspects.
Figure 8 is a flow diagram of an example of a process for determining a winning combination of descriptors, parameters, and weights in certain aspects.
FIG. 9 is a block diagram of an example of a computing device for performing an automated well log correlation according to certain aspects.
Related detailed description
Certain aspects and specifics of this disclosure relate to a correlation (eg, alignment) of well logs in performing dynamic alignment using descriptors. Dynamic alignment can include shifting, compressing, and stretching two well logs to try to align the well logs with each other. Descriptors can be information present in a pattern, shape, frequency, amplitude, or other data characteristic surrounding a node (a data point) of interest in a well log. As a particular example, a computing device can identify descriptors related to each node in a first well log and descriptors related to each node in a second well log. The computing device can then perform dynamic alignment using the descriptors to determine an alignment between the well logs. For example, the computing device can perform dynamic time alignment at least in part by comparing the descriptors linked to each node in the first well log to the descriptors linked to each node in the second well log. The resulting alignment between well logs may be more precise than alignments determined using other methods. For example, the resulting alignment can be
2016-IPM-100867-U1-EN 3 more precise than is practicable by simply comparing the amplitude of each node in the first well log to the amplitude of each node in the second well log. Performing dynamic alignment using descriptors can leverage more information from well logs to perform well log correlation, thereby providing more informed and accurate results.
In some examples, dynamic alignment can be achieved using multiple descriptors for each node. And each descriptor can present one or more configurable parameters, a configurable weighting, or both. It can be difficult and time consuming to manually identify a combination of descriptors, parameter values, and weights that result in sufficiently precise alignment between two well logs. Some exempt from this disclosure may overcome one or more of the problems mentioned above (i) by automatically generating various combinations of descriptors, parameters, and weights; (ii) automatically performing dynamic alignment on two well logs using the various combinations; and (ni) by automatically identifying which combination results in the most precise alignment between the two well logs.
For example, the computing device can receive a selection of two well logs. The computing device can then receive user input indicating a node in one of the well logs which is correlated to another node in another of the well logs. This manually determined correlation can be referred to as a manual correlation. After which, the computing device can generate (e.g., randomly) multiple different combinations of descriptors, parameters, and weights. The computing device can perform dynamic alignment on the two well logs using all of the different combinations to identify which combination results in a correlation that is most similar to manual correlation. The calculator can use this combination as a winning combination. The compute device can then correlate nodes in other well logs by performing dynamic alignment using the winning combination.
These illustrative examples are given to present to the reader the general object described here and are not intended to limit the scope of the concepts disclosed. The following sections describe various specific features and additional examples with reference to the drawings, among which identical numbers denote elements.
2016-IPM-100867-U1-EN 4 identical, and the management descriptions are used to describe the illustrative aspects but, like the illustrative aspects, they should not be used to limit the present disclosure.
Figure 1 is a side sectional view of an example of a well system 100 for obtaining well logs in certain aspects. The well system 100 includes multiple wells 102a to h drilled through an underground formation 104. The wells 102a to b extend from the surface of the well 108 in strata 106a to c of the underground formation 104. Strata 106a-c can include different materials (eg, rock, soil, oil, water, or gas) and vary in thickness and shape.
Some or all of the boreholes 102a through h may include well tools, such as logging tools 110a and b, for generation of well logs. Each of the well tools can measure properties of rocks, a fluid, or other contents of strata 106a-c and use the measured properties to generate a respective well log. For example, the logging tool 110a can measure the electrical, acoustic, radioactive, electromagnetic, or pressure properties of the strata regions near a wellbore 102a. The 110a logging tool can then use the measurements to generate a well log. A separate well log can be generated for each of the boreholes 102a through h.
The well tools can electronically communicate the well logs to a computing device 112, which can be positioned on site (as shown in Figure 1) or off site. The well tool can communicate electrically with the computing device 112 via a wired or wireless interface. In some examples, the well tools may communicate the well logs electronically to the computing device 112 indirectly, such as over the Internet or another network.
The computing device 112 can display some or all of the well logs as two-dimensional (2D) or three-dimensional (3D) figures. An example of such a figure is shown in Figure 2. In Figure 2, each vertical line represents a single wellbore. For example, a vertical line 202a represents a particular wellbore. Since there are 21 vertical lines, there are 21 different boreholes shown in this figure. The portion of line 202a near the top 206 of (a figure represents the portion of the wellbore near the well surface, with the depth of the wellbore increasing at the bottom of the page. The patterned data to the left of 'a vertical line are a well log.
2016-IPM-100867-U1-FR 5
For example, pattern data 204 is a well log corresponding to the wellbore represented by a line 202a. A peak, a trough, or a change in the shading of pattern data 204 may represent the changing composition of strata in the underground formation.
In Figure 2, the nine rightmost vertical lines have two corresponding well logs: one well log on the left and another well log on the right. For example, the vertical line 202b has a well log 210 on the left and another well log 212 on the right. These well logs come from the same wellbore (represented by a vertical line 202b), but are generated using different methods. For example, a well log 210 can be generated based on gamma radiation emitted from the strata in the underground formation, and a well log 212 can be generated based on the spontaneous potential of the strata in the underground formation. . The computing device 112 can output any number and any combination of well logs individually or at the same time.
In the example shown in Figure 2, all of the well logs come from boreholes that are part of the same well system and penetrate the same strata. So all of the well logs provide information about the same strata. But since the strata are non-uniform, your wells are in different locations within the well system, and the wells have different depths, not all well logs will align perfectly with each other. others. For example, the grooves and troughs representing a particular stratum in a well log 208 may be closer to the top 206 of Figure 2 than the grooves and troughs representing the same stratum in a well log 210. It may be desirable identify and correlate a single stratum across multiple well logs.
A geologist can manually review your well logs and identify the same stratum in each well log. The geologist can then correlate the same stratum through the well logs. Examples of manually identified and correlated strata are shown as solid lines in Figure 2. The first solid line (extending from a box "1" on the left) can represent the top surface of a first strata. The second continuous line (extending from a box "2" on the left) can represent the upper surface of a second layer. The third continuous line (extending from a box "3" on the left) can represent the upper surface of a third layer. But
2016-IPM-100867-U1-EN 6 it is costly, time consuming, difficult, and subjective to manually review and correlate strata surfaces among multiple well logs. A geologist can spend several days identifying and correlating thousands of surfaces through dozens of well logs.
In some examples, the computing device 112 can run a software application to identify and correlate strata across multiple well logs in a more automated fashion. This can be referred to as an automated correlation. For example, the computing device 112 can analyze each of the well logs, identify the various strata in the well logs, and correlate your strata to each other, all with little or no manual intervention. Examples of strata surfaces correlated in such a way by the computing device 112 are shown as dotted lines in your figure 2. The first dotted line (extending from a box "1" to the left ) can represent the upper surface of a first layer. The second dotted line (extending from a box "2" on the left) can represent the top surface of a second layer. The third dashed line (extending from a box "3" on the left) can represent the top surface of a third layer. Automated correlation can be faster, more consistent, require less processing time, and require less power than manual correlation.
In some examples, the computing device 112 can perform automated correlation using dynamic alignment. An example of dynamic alignment is dynamic time alignment. Dynamic alignment can include shifting, compressing, and stretching two well logs to try to align the well logs with each other. An example of dynamic alignment is shown in Figures 3A and 3B. Figure 3A shows an example of two well logs for two wells, Well I and Well J, before dynamic alignment. The underground depths associated with the nodes in the well logs are expressed along the Y axis. The amplitudes of the nodes are expressed along the X axis. In your Figure 3A, your two well logs are in misalignment l with each other. For example, sets of nodes 302a and b, 304a and b, and 306a and b are out of alignment with each other. Figure 3B shows an example of the well logs after dynamic alignment. As shown, the well log for a Well I has been compressed and shifted to align it with the well log for a Well J. For example, the above mentioned nodes (shown circled) are now aligned with each other. other. After an alignment
2016-IPM-100867-U1-EN 7 of the two well logs to each other, correlations between the various nodes (e.g., between a node 302a and a node 302b, between a node 304a and a node 304b, and between a node 306a and a node 306b) can be determined.
A dynamic alignment can include a comparison of the amplitudes of the nodes in the well log for a Well I to the amplitudes of the nodes in the well logs for a Well J. For example, the amplitude of a node 302a can be compared to l amplitude of each respective node in the well log for a Well J. A dynamic alignment can also include a creation of an error matrix expressing the respective difference between a node 302a and each respective node in the well log for a Well J. This process can be repeated for all nodes in the two well logs, with each repetition leading to the addition of more information to the error matrix. After creating the error matrix, a path through the error matrix with the lowest possible average error can be determined. The path can be correlated nodes between the two well logs. Each point in the path may be a correlation between a node in the well log for a Well l and another node in the well log for a Well J. The computing device 112 can perform dynamic alignment at least in part to implement a automated correlation process.
Dynamic alignment that is implemented only by comparing the amplitudes of nodes in well logs to each other, however, may not always provide ideal results. For example, returning to Figure 2, the dashed lines represent automated correlations generated using this type of dynamic alignment. And the solid lines represent manual correlations. We can see that the dotted lines deviate from the solid lines in certain locations. These deviations can represent errors between the automated correlations and the manual correlations.
Certain examples of this disclosure may provide more precise automated correlations by taking into account in addition or alternatively other information related to well logging nodes when performing dynamic alignment. For example, the computing device 112 can analyze each node in a set of well logs and label each node with additional information which describes the shape surrounding the node, the amplitudes of neighboring nodes, an average of the amplitudes of neighboring nodes, a frequency of the amplitudes of neighboring nodes, or any combination thereof. Each additional piece of information can be designated as a descriptor because it describes
2016-IPM-100867-U1-FR 8 a characteristic of the node. Then, rather than just comparing (the amplitudes of two nodes during a dynamic alignment, the computing device 112 can additionally or alternatively compare the descriptors for the nodes. This can lead to a more precise correlation between your two nodes, and thus to a more precise correlation between the set of well diagrapbies.
An example of a descriptor can be a form descriptor. A shape descriptor can indicate a form of data surrounding (before, after, or both) a node of interest in a well log. For example, the shape descriptor may indicate one or more slopes (eg, gradients) in data surrounding a node of interest in a well log. An example of a shape descriptor is shown in Figures 4A and 4B. In Figure 4A, the computing device 112 has identified a node of interest in a well log. The node of interest is surrounded. The caicul device 112 can then divide a predetermined number of nodes in the region surrounding the node of interest into a predetermined number of segments, which in this example is four segments 402a to 408a. The calculation device 112 can then determine the slope of the nodes in each segment. For example, the computing device 112 can determine the slope of all the nodes in a segment 402a by performing a linear regression through the nodes in the segment 402a. The computing device 112 can repeat this process for segments 404a to 408a. Visual representations of the slopes for your segments 402a to 408a are shown in Figure 4A to the right of segments 402a to 408a. The caicul device 112 can then associate some or all of the slopes with the node of interest. In some examples, each individual slope can be a separate shape descriptor for the node of interest. In other examples, two or more of the slopes can collectively represent a shape descriptor for the node of interest. For example, a single shape descriptor for the node of interest may include the four slopes for segments 402a to 408a. The computing device 112 can repeat the process mentioned above for some or all of the nodes in the well log 400a.
After determining the shape descriptors for the nodes in the well log 400a, the computing device 112 can perform dynamic alignment at least in part by comparing the shape descriptor for each node in a well log 400a with the shape descriptor for each node in a well log 400b in Figure 4B. For example, the computing device 112 can compare the shape descriptor for the node of interest in a well log
2016-IPM-100867-U1-FR 9
400a to the shape descriptor for the node of interest in a well log 400b. It can be seen in FIGS. 4A and B that the shape descriptors for these two nodes of interest are substantially similar, which can result in the fact that the computing device 112 identifies a correlation between these two nodes. This can lead to a more precise correlation between well diagrapbies 400a and b than if, for example, only the amplitudes of the nodes of interest were compared with each other (e.g., because the amplitudes of the nodes of interest in the figures 4A and B are significantly different from each other).
Another example of a descriptor is an average descriptor. The mean descriptor can indicate mean values of the amplitudes of nodes surrounding a node of interest in a well log. For example, the computing device 112 can divide a predetermined number of nodes in the region surrounding the node of interest into a predetermined number of segments, such as segments 402a to 408a. The computing device 112 can then determine a respective average of the amplitudes of the nodes in each segment. In some examples, each individual average can be a separate average descriptor for the node of interest. In other examples, two or more of the means may collectively represent a mean descriptor for the node of interest. For example, a single mean descriptor for the node of interest may include the four means for segments 402a to 408a.
Another example of a descriptor is an amplitude descriptor. The amplitude descriptor can indicate the amplitudes of nodes surrounding a node of interest in a well log. For example, the amplitude descriptor can indicate one or more amplitudes of a predetermined number of nodes surrounding the node of interest.
Another example of a descriptor is a histogram descriptor.
The histogram descriptor can indicate frequencies in which the slopes of nodes which surround a node of interest lie within previously designated slope ranges. For example, the computing device 112 can identify a predetermined number of nodes in the region surrounding the node of interest. The calculation device 112 can determine the slope of each of these nodes. In some examples, the computing device 112 can determine a slope of a node by determining a central gradient of the node according to the following equation:
p _ node value (+1+ value _noeu ^ _j_ ^{into an} noeud_i 2 * (distance between nodes)
2016-IPM-100867-U1-FR 10 where Pente _noeudi is the slope through a node i; _node value, ₊₁ is your value of a node i + 1; _node value, _ _t is the value of a node i-1; and Distance between nodes is the distance between a node i + 1 and a node i-1 (e.g., with a half-foot depth spacing between nodes, the distance between a node i + 1 and a node i-1 would be 1 foot). The computing device 112 can also determine categories (eg, ranges of slopes) into which to categorize the slopes of the nodes. For example, the computing device 112 can determine four categories, where a first category includes slopes between -40 degrees and -20 degrees, a second category includes slopes between -20 degrees and 0 degrees, a third category includes slopes between 0 degrees and 20 degrees, a fourth category includes slopes between 20 degrees and 40 degrees. The calculating device 112 can determine any number of categories, with each category covering a similar or different number of degrees. The computing device 112 can then categorize the slopes of the nodes in their corresponding categories, and count how many nodes are in each category. The number of nodes in each category can be designated as a category value. For example, the computing device 112 can determine the fact that three nodes are in the first category, therefore the first category can have a category value of three. The calculation device 112 can determine the fact that a node is in the second category, therefore the second category can have a category value of one. The computing device 112 can determine the fact that two nodes are in the third category, therefore the third category can have a category value of two. The histogram descriptor for the node of interest can indicate some or all of these category values.
Another example of a descriptor is an error descriptor. The error descriptor may indicate an error between the amplitudes of nodes of interest in two different well logs. The calculation device 112 can determine the error between the nodes of interest according to the following equation:
, i power error = noeudi - node) j where nodes is ie node of interest in one well log, nodej is the node of interest in the other well log, and power is a predetermined exponential value. The error descriptor can reduce or exaggerate an outlier effect.
The computing device 112 can determine and use any number and combination of descriptors for a correlation of well logs. By
2016-IPM-100867-U1-EN 11 example, the computing device 112 can use shape descriptors, amplitude descriptors, mean descriptors, histogram descriptors, error descriptors, or any combination of these (in addition or as a variant of the amplitudes of the nodes) to correlate well logs. This can lead to a more precise correlation. For example, Figure 5 shows the same well logs as Figure 2, with the dotted lines representing automated correlations generated using dynamic alignment with descriptors. Solid lines represent manual correlations. The dotted lines are significantly closer to the solid lines than in Figure 2, which may mean that the automated correlations are more precise.
In some examples, the computing device 112 can associate weights with the different descriptors to improve the accuracy of the correlation between well logs. For example, a shape descriptor match may be a more reliable indication of a correlation between nodes in well logs than an amplitude descriptor match. Therefore, the computing device 112 can assign shape descriptors a higher weighting than amplitude descriptors. The calculation device 112 can take into account the weights for the descriptors when performing a well log correlation.
It can be difficult to manually select descriptors, parameters for each descriptor, and weighting for each descriptor due to the large number of descriptors and parameters involved. For example, the shape descriptor can have a number of nodes parameter, a node location parameter, a number of segments parameter, a segment size parameter, or any combination thereof. The number of nodes setting can indicate how many surrounding nodes (e.g., nodes surrounding a node of interest) should be divided into segments. The node location setting can indicate whether the surrounding nodes should be located before the node of interest, after the node of interest, or both. The number of segments setting can indicate how many segments divide your surrounding nodes. The segment size parameter can indicate the number of nodes per segment (there may be overlapping segments). Varying each of these parameters can affect the accuracy of a well log correlation. As another example, the mean descriptor can present a number of nodes parameter, a node location parameter, a number of segments parameter, a size parameter
2016-IPM-100867-U1-EN 12 segment, or any combination thereof. In yet another example, the amplitude descriptor can present a number of nodes parameter, a node location parameter, or both. In yet another example, the histogram descriptor can present a number of nodes parameter, a node location parameter, a number of categories parameter Indicating in how many categories to categorize node slopes, a type of parameter slopes, or any combination thereof. In addition, each of these descriptors can have an associated weighting. In yet another example, the error descriptor may present an exponent parameter for the exponential value in the above-mentioned equation. It can be difficult, time consuming, and expensive for a user to attempt to manually determine appropriate parameters and weights for each descriptor to provide accurate well log correlation results.
Certain examples of this disclosure may overcome one or more of the above problems by automatically selecting one or more descriptors, parameters for the descriptor (s), and / or weights for the descriptors for use in a log correlation realization. well. An example of a process flow diagram of a process for automatically selecting such information is shown in Figure 6. In other examples, the process can be implemented using more steps, fewer steps, different steps, or a different order of steps shown in Figure 6.
Referring now to FIG. 6, in a block 602, the computing device 112 determines wells of interest. The computing device 112 can determine the wells of interest based on user input indicating the wells of interest. For example, the computing device 112 may display a graphical user interface (GUI) presenting a visual representation of multiple wellbores in a well system. The computing device 112 can receive, as user input, a selection of some or all of the wells displayed in the GUI. In some exemptions, the boreholes do not need to be selected or ordered according to a primary dip line. The computing device 112 can receive user input via a user input device. The computing device 112 can use the wells selected as wells of interest.
In a block 604, the computing device 112 receives a group of well logs associated with the boreholes of interest. For example, the
2016-IPM-100867-U1-EN 13 computation 112 can retrieve well logs from a local memory device. In another example, the computing device 112 can communicate with a remote server to retrieve well logs from a remote database.
In a block 606, the calculation device 112 pre-processes the group of well logs corresponding to the boreholes of interest. The calculation device 112 can carry out one or more operations to pre-process the well logs.
In some examples, the computing device 112 can pre-process the well logs by normalizing the well logs so that the well logs can be compared with each other. For example, all well logs may not have a uniform range of values. To make the well logs comparable, the computing device 112 can manipulate the well logs so that all of the well logs have the same range of values. As a specific example, a spontaneous polarization well log often does not have a standard range of values. Therefore, to compare two spontaneous polarization well logs, the computing device 112 can normalize the two well logs so that they both have the same range of values.
In some examples, the computing device 112 can pre-process the well logs by smoothing the well logs. For example, well logs may exhibit high frequency noise. The computing device 112 can smooth your well logs to reduce or eliminate the influence of high frequency noise. For example, the computing device 112 may divide the well logging into intervals of a predetermined amount of nodes (eg, five-node intervals). The computing device 112 can determine an average value of the nodes in an interval and assign all of your nodes in the interval the average value. Computing device 112 can repeat this process for all intervals, thereby smoothing the well log. This can help reduce the influence of high frequency noise in the well log correlation process.
In some examples, the computing device 112 can preprocess your well logs by removing a trend from the well logs. A trend can include an increase or decrease in the values of a well log with depth. A trend may occur due to a malfunction of the well tool that generated the well logging or to another non-geological phenomenon. The computing device 112 can eliminate a trend from a well log to reduce errors associated with the trend.
2016-IPM-100867-U1-FR 14
In some exemptions, the computing device 112 may not preprocess the well logs. Instead, the compute device 112 can use your well logs in their raw form to perform some or all of the remaining steps described below.
In block 608, computing device 112 receives user input indicating manual correlation between nodes in two of the well logs. In some examples, the computing device 112 may receive user input indicating a manual correlation between a first node in a first well log in the group of well logs and a second node in a second well log in the group well logs. As a particular example, the computing device 112 can display the two well logs adjacent to each other, as in Figure 7. A user can manipulate a mouse or other user input device to select the first node in a Well I and the second node in a Well J, The computing device 112 can receive user input and, based on user input, determine whether a correlation exists between the first node in a Well I and the second node in a Well J. An example of such a correlation is represented by an arrow 704a.
In a block 610, the computing device 112 determines whether there are other manual correlations to be entered. For example, automated correlation software may require a minimum number of manual correlations between well logs (eg, to provide sufficiently accurate results). If the computing device 112 determines that there are other manual correlations to be entered, the process can revert to a block 608, the user can then enter another manual correlation. This second manual correlation can be represented by an arrow 704b in FIG. 7. Blocks 608 to 610 can be repeated so that the user can enter any number and combination of manual correlations between any number and combination of well logs in the well log group. If the computing device 112 determines that there are no other manual correlations to enter, the process can continue to block 612.
In a block 612, the computing device 112 determines a winning combination of descriptors, parameters for the descriptor (s), weights for the descriptor (s), or any combination of these on the basis of manual correlations between the two well logs. The winning combination can include any number and combination of descriptors, parameters, and weights. In some exempt, the computing device 112 can determine the
2016-IPM-100867-U1-EN 15 winning combination by performing some or all of the processes shown in Figure 8. In other examples, the process in Figure 8 can be implemented using more steps , fewer steps, different steps, or a different order of steps shown in Figure 8.
Referring now to FIG. 8, in a block 802, the computing device 112 selects different combinations of descriptors from a group of descriptors. The descriptor group can include a shape descriptor, an amplitude descriptor, an average descriptor, a histogram descriptor, an error descriptor, or any combination thereof.
In some examples, the computing device 112 can select the different combinations of descriptors according to any technique or combination of techniques. For example, the computing device 112 can randomly select the different combinations of descriptors. As another example, the computing device 112 can methodically select the different combinations of descriptors to guarantee that all possible combinations of descriptors are attempted. In yet another example, certain combinations of descriptors may be previously designated as particularly precise or particularly imprecise. The calculation device 112 can select the combinations previously designated as particularly precise, pass the combinations previously designated as particularly imprecise, or both.
In some cases, the computing device 112 can determine an initial combination of descriptors. The calculation device 112 can then carry out some or all of the remaining steps in FIG. 8 to determine whether the initial combination of descriptors corresponds to one or more manual correlations or is within a predetermined tolerance range of the manual correlation (s) . If so, the initial combination of descriptors can be used as the winning combination. If not, your process can revert to block 802, where another combination of descriptors can be selected, and your process can be repeated (e.g., until a winning combination of descriptors is identified).
In a block 804, the computing device 112 determines parameter values for the descriptors in each combination of descriptors. For example, the computing device 112 can determine values for at least two parameters corresponding to at least one descriptor in a combination of descriptors. As a specific example, the computing device 112 can select
2016-IPM-100867-U1-EN 16 values for a number of nodes parameter, a node location parameter, a number of segments parameter, and a segment size parameter corresponding to a shape descriptor in a combination descriptors.
In some examples, the computing device 112 can randomly select your parameter values for your descriptors. In other examples, the computing device 112 can methodically select the different parameter values to guarantee that all possible combinations of parameter values are attempted. In some examples, certain parameter values or ranges of parameter values may be previously designated as particularly precise or particularly imprecise. The calculation device 112 can select the parameter values previously designated as particularly precise, pass your parameter values previously designated as particularly imprecise, or both.
In a block 806, the calculation device 112 determines weights for the descriptors in each combination of descriptors. In some examples, the computing device 112 can randomly select the weights for the descriptors. In other examples, the computing device 112 can methodically select the different weights to ensure that all of your possible combinations of weights are attempted. In some exemptions, certain weights may be previously designated as particularly precise or particularly imprecise. The calculation device 112 can select the weights previously designated as particularly precise, pass the combinations previously designated as particularly imprecise, or both.
In a block 808, the computing device 112 performs dynamic alignment on two well logs using the different combinations of descriptors, parameters, weights, or any combination of these to generate multiple different alignments between the two well logs. For example, the computing device 112 can perform dynamic alignment on the two well logs using all of the combinations of descriptors, parameters and weights previously identified. This can result in multiple different alignments between the two well logs.
In a block 810, the computing device 112 analyzes the multiple different alignments to determine which combination of descriptors, parameters, weights, or any combination thereof results in the correlation which is closest to the correlation (s) manual. For example, the
2016-IPM-100867-U1-EN 17 calculation device 112 can compare all of the different alignments to determine that a specific combination of descriptors, parameters, and weights results in your correlation closest to your correlation (s) manual supplied in block 608.
In a block 812, the computing device 112 selects as a winning combination the combination of descriptors, parameters, and weights which results in the closest correlation to the manual correlation or correlations. In certain examples, the calculation device 112 can identify several combinations of descriptors, parameters and weights which result in correlations which lie within a predetermined tolerance range of ta or manual correlations. The predetermined tolerance range can be provided as user input. The calculation device 112 can select, from among the several combinations, the combination which presents the results most similar to ta or to manual correlations such as the winning combination.
Returning now to FIG. 6, in a block 614, the computing device 112 can correlate nodes between other well logs in the group of well logs by performing dynamic alignment using the winning combination (e.g. , determined in block 812). For example, the computing device 112 can select a different set of well logs from the group of well logs. The computing device 112 can perform dynamic alignment on the different set of well logs using the descriptors, parameters, and weights determined in a block 612. This can lead to a correlation between the different set of well logs. well. In some examples, the computing device 112 may repeat this process, for example, to determine correlations between each possible pair of well logs in the group of well logs. These correlations can be designated as correlations by two.
After determining some or all of the correlations by two, the computing device 112 can use a regression algorithm on the correlations by two to identify an overall correlation across all the well logs. The computing device 112 can then determine a surface of a stratum (a "stratum surface") represented in your well logs and correlate the surface of stratum across all the well logs in the group of well logs.
In some exempt, some or all of the specifics
2016-IPM-100867-U1-EN 18 mentioned above can be implemented using the computing device 112 shown in FIG. 9. The computing device 112 can include a processing device 904, a bus 906, a device memory 908, a user input device 916, a display device 918, and a communication interface 920. In some examples, some or all of the components shown in Figure 9 may be integrated into a single structure, such as a single housing. In other examples, some or all of the components shown in Figure 9 can be distributed (eg, in separate packages) and in electrical communication with each other.
The processing device 904 can perform one or more operations to automatically correlate well logs using descriptors. The processing device 904 can execute instructions stored in the memory device 908 to carry out the operations. The processing device 904 may include a processing device or multiple processing devices. Non-limiting examples of the processing device 904 include a programmable pre-broadcast integrated circuit ("FPGA"), an application-specific integrated circuit ("ASiC"), a micro-processing device, etc.
The processing device 904 can be communicatively coupled to the memory device 908 via the bus 906. The non-volatile memory device 908 can include any type of memory device which retains information when it is not energized. Nonlimiting examples of the memory device 908 include an electrically erasable and erasable read only memory ("EEPROM"), a flash memory, or any other type of non-volatile memory. In some examples, at least a portion of the memory device 908 may include a medium from which the processing device 904 can read instructions. Computer readable media may include electronic, optical, magnetic, or other storage devices capable of providing the processing device 904 with computer readable instructions or other program code. Non-limiting examples of computer readable media include (but are not limited to) one or more magnetic disks, one or more memory chips, read only memory (ROM), random access memory ("RAM"), ASIC , a configured processing device, an optical storage device, or any other medium from which a computer processing device can read instructions. The instructions may include device-specific processing instructions generated by a compilation program or an interpretation program from
2016-IPM-100867-U1-FR 19 of code written in any suitable computer programming language, including, for example, C, C ++, C #, etc.
In some examples, memory device 908 can include well logs 910. Well logs 910 can be communicated to computing device 112 from one or more well tools positioned in one or more wells.
In some examples, memory device 908 may include a well log correlation engine 912. The well log correlation engine 912 may be a software application for identifying correlations among nodes in well logs 910 The 912 well log correlation engine can identify correlations using descriptors to achieve dynamic alignment. In some examples, the well log correlation engine 912 may determine the descriptors, parameters for the descriptor (s), and / or weights for the descriptors for use in dynamic alignment.
In some examples, the memory device 908 may include a correlation database 914. The correlation database 914 may be a database having correlations between nodes in well logs. For example, the correlation database 914 may include an association of nodes in the well logs 910.
In some examples, the computing device 112 includes a user input device 916. The user input device 916 can represent one or more components used to enter data. Examples of the 916 user input device may include a keyboard, mouse, touchpad, button, or touch screen display, etc.
In some examples, the computing device 112 includes a display device 918. The display device 918 can represent one or more components used to output data. Examples of the 918 display device may include a liquid crystal display (LCD), television, computer monitor, touch screen display, etc. In some examples, the user input device 916 and the display device 918 may be a single device, such as a touch screen display.
In some examples, the computing device 112 includes a communication interface 920. The communication interface 920 can represent one or more components which facilitate a network connection or otherwise facilitate a
2016-IPM-100867-U1-FR 20 communication between electronic devices. Examples include, but are not limited to, wired interfaces such as Ethernet, USB, IEEE 1394, and / or wireless interfaces such as IEEE 802.11, Bluetooth, near field communication (CCP) interfaces, interfaces RFID, or radio interfaces for access to cellular telephone networks (e.g., a transceiver / antenna for access to a CDMA, GSM, UMTS, or other mobile communication network).
In some aspects, automated well logging can be performed using descriptors according to one or more of the following examples:
Example 1: Non-transient computer-readable medium that may include a program code for automatic correlation of well logs to identify specificities of a stratum represented in well logs. The program code can be executable by a processing device. The program code may cause the processing device to receive a plurality of well logs. Each well log of the plurality of well logs may indicate subterranean strata penetrated to various depths by a respective wellbore. The program code may cause the processing device to determine a combination of descriptors used to correlate a first data point in a first well log of the plurality of well logs to a second data point in a second well log the plurality of well logs. The combination of descriptors can be determined (i) by selecting different combinations of descriptors from a plurality of descriptors, where a descriptor can be a type of information about a data point in a well log; (ü) performing dynamic alignment on the first well log and the second well log using the different descriptor combinations to generate a plurality of alignments between the first well log and the second well log, where an alignment dynamic may include manipulating a form of a well log to align the well log with another well log; and / or (tii) by analyzing the plurality of alignments to determine whether the combination of descriptors results in the first data point being correlated with the second data point. The program code can cause the processing device to correlate data points between other well logs in the plurality of well logs by performing dynamic alignment using the combination of descriptors.
Example 2: Non-transient computer-readable medium according to
2016-IPM-100867-U1-EN 21 exemption 1, which may also include a program code executable by the processing device to cause the processing device to determine the combination of descriptors by selecting a value for a parameter corresponding to at least one descriptor in the combination of descriptors.
Example 3: Non-transient computer-readable medium according to any one of Examples 1 and 2 which may also include a program code executable by the processing device to cause the processing device to determine the combination of descriptors by selecting a respective weight for each descriptor in the combination of descriptors.
Free 4; Non-transient computer-readable medium according to any of Examples 1 to 3 which may contain the combination of descriptors including a shape descriptor which is indicative of one or more slopes in data before or after a data point of interest. The non-transient computer-readable medium may further include program code executable by the processing device to cause the processing device to determine the one or more slopes: (i) by determining the data point of interest in the first well logging; (ii) dividing data points before or after the data point of interest into one or more segments; and / or (iii) determining the one or more slopes on the basis of the one or more segments. Each slope of the one or more slopes can correspond to a respective segment of the one or more segments and can be determined by performing a linear regression on respective data points in the respective segment.
Example 5: Non-transient computer-readable medium according to any one of Examples 1 to 4 which may contain the combination of descriptors including an average descriptor which is indicative of one or more data means before or after a data point of interest The non-transient computer-readable medium may further include a program code executable by the processing device to cause the processing device to determine your one or more means: (i) by determining the data point of interest in the first well log; (ii) dividing data points before or after the data point of interest into one or more segments; and / or (iii) determining the one or more averages on the basis of the one or more segments. Each average of the one or more averages can correspond to a respective segment of the one or more segments and can be determined by averaging the respective data points in the respective segment.
2016-IPM-100867-U1-FR 22
Example 6: Non-transient computer-readable medium according to any of Examples 1 to 5 which may contain the combination of descriptors including a histogram descriptor which is representative of frequencies in which slopes of a plurality of data points occur lie in a plurality of ranges of slopes. The non-transient computer-readable medium may further include program code executable by the processing device to cause the processing device to determine the frequencies in which the slopes of the plurality of data points lie in the plurality of ranges of slopes; (i) determining a plurality of slopes corresponding to the plurality of data points, each slope in the plurality of slopes corresponding to a respective data point in the plurality of data points; (ii) determining the plurality of slope ranges, each slope range of the plurality of slope ranges covering a predetermined number of degrees; (iii) associating each data point in the plurality of data points with a respective slope range of the plurality of slope ranges based on a slope associated with the data point; and / or (iv) determining a number of data points associated with each range of slopes of the plurality of ranges of slopes.
Example 7: Non-transient computer-readable medium according to any one of Examples 1 to 6, which may also include a program code executable by the processing device to cause the processing device to determine the combination of descriptors; (i) by identifying multiple combinations of descriptors which lead to the first data point being correlated to the second data point in an amount which is within a predetermined tolerance range; and / or (îi) selecting the combination of descriptors from the multiple combinations of descriptors based on the combination of descriptors resulting in a closer correlation between the first data point and the second data point than another combination descriptors in the multiple combinations of descriptors.
Example 8; System which may be for automatic correlation of well logs to identify specificities of a stratum represented in well logs. The system may include a processing device and a memory device on which instructions executable by the processing device are stored. The instructions may cause the processing device to receive a plurality of well logs. Each well log of the plurality of well logs may indicate subterranean strata penetrated to various depths by a respective wellbore. The instructions can bring the device
2016-IPM-100867-U1-EN 23 process to determine a combination of descriptors used to correlate a first data point in a first well log of the plurality of well logs to a second data point in a second well log wells from the plurality of well logs. The combination of descriptors can be determined (i) by selecting different combinations of descriptors from a plurality of descriptors, where a descriptor can be a type of information about a data point in a well log; (ii) by performing a dynamic alignment on the first well log and the second well log using the different combinations of descriptors to generate a plurality of alignments between the first well log and the second well log, where an alignment dynamic may include manipulating a form of a well log to align the well log with another well log; and / or (iii) analyzing the plurality of alignments to determine whether the combination of descriptors results in the first data point being correlated with the second data point. The instructions may cause the processing device to correlate data points between other well logs in the plurality of well logs by performing dynamic alignment using the combination of descriptors.
Example 9; System according to example 8 which can contain the memory device further including instructions executable by the processing device to cause the processing device to determine the combination of descriptors by selecting values for at least two parameters corresponding to at least one descriptor in the combination of descriptors.
Example 10: System according to any one of Examples 8 and 9 which can contain the memory device further including instructions executable by the processing device to cause the processing device to determine the combination of descriptors by selecting a respective weighting for each descriptor in the combination of descriptors.
Example 11: System according to any one of Examples 8 to 10 which may contain the combination of descriptors including a shape descriptor which is indicative of one or more slopes in data before or after a data point of interest. The memory device may further include instructions executable by the processing device to cause the processing device to determine the one or more slopes: (i) by determining the data point of interest in the first well log; (ii) dividing data points before or after the data point of interest into one or more segments; and / or (iii) in
2016-IPM-100867-U1-EN 24 determining one or more slopes on the basis of one or more segments. Each slope of the one or more slopes can correspond to a respective segment of the one or more segments and can be determined by performing a tinear regression on respective data points in the respective segment.
Example 12: System according to any one of exempt 8 to 11 which may contain the combination of descriptors including an average descriptor which is indicative of one or more data averages before or after a data point of interest. The memory device may further include instructions executable by the processing device to cause the processing device to determine the one or more means: (i) by determining the data point of interest in the first well log; (ii) dividing data points before or after the data point of interest into one or more segments; and / or (iii) determining the one or more averages on the basis of the one or more segments. Each average of the one or more averages can correspond to a respective segment of the one or more segments and can be determined by averaging the respective data points in the respective segment.
Example 13: System according to any one of Examples 8 to 12 which may contain the combination of descriptors including a histogram descriptor which is representative of the frequencies in which slopes of a plurality of data points lie in a plurality of slope beaches. The memory device may further include instructions executable by the processing device to cause the processing device to determine the frequencies in which the slopes of the plurality of data points lie in the plurality of ranges of slopes: (i) determining a plurality of slopes corresponding to the plurality of data points, each slope in the plurality of slopes corresponding to a respective data point in the plurality of data points; (ii) determining the plurality of slope ranges, each slope range of the plurality of slope ranges covering a predetermined number of degrees; (iii) associating each data point in the plurality of data points with a respective slope range of the plurality of slope ranges based on a slope associated with the data point; and / or (iv) determining a number of data points associated with each range of slopes of the plurality of ranges of slopes.
Example 14: System according to any of Examples 8 to 13 which may contain the memory device which may further include instructions executable by the processing device to bring the processing device to
2016-IPM-100867-U1-EN 25 determine the combination of descriptors: (i) by identifying multiple combinations of descriptors which lead to the first data point being correlated to the second data point in a quantity which lies within a range of predetermined tolerance; and / or (ii) selecting the combination of descriptors from the multiple combinations of descriptors based on the combination of descriptors resulting in a closer correlation between the first data point and the second data point than another combination descriptors in the multiple combinations of descriptors.
Example 15; A method may be for automatic correlation of well logs to identify specificities of a stratum represented in well logs. The method may include receiving a plurality of well logs. Each well log of the plurality of well logs may indicate subterranean strata penetrated to various depths by a respective wellbore. The method may include determining a combination of descriptors used to correlate a first data point in a first well log of the plurality of well logs to a second data point in a second well log of the plurality of logs well. The combination of descriptors can be determined (i) by selecting different combinations of descriptors from a plurality of descriptors, where a descriptor can be a type of information about a data point in a well log; (ii) performing dynamic alignment on the first well log and the second well log using the different descriptor combinations to generate a plurality of alignments between the first well log and the second well log, where an alignment dynamic may include manipulating a form of a well log to align the well log with another well log; and / or (iii) analyzing the plurality of alignments to determine whether the combination of descriptors leads to the first data point being correlated to the second data point. The method can include correlating data points between other well logs in the plurality of well logs by performing dynamic alignment using the combination of descriptors. Some or all of the steps can be performed by a processing device.
Example 16; The method according to example 15 can also include determining the combination of descriptors by selecting values for at least three parameters corresponding to at least one descriptor in the combination of descriptors.
2016-IPM-100867-U1-FR 26
Example 17: Method according to any one of Examples 15 and 16 which can also include determining the combination of descriptors by selecting a respective weighting for each descriptor in the combination of descriptors.
Example 18: Method according to any one of Examples 15 to 17 which may contain the combination of descriptors including a shape descriptor which is indicative of one or more slopes in data before or after a data point of interest. The method may further include determining one or more slopes: (i) by determining the point of data of interest in the first well log; (ii) dividing data points before or after the data point of interest into one or more segments; and / or (iii) determining the one or more slopes on the basis of the one or more segments. Each slope of the one or more slopes can correspond to a respective segment of the one or more segments and can be determined by performing a linear regression on respective data points in the respective segment.
Example 19: Method according to any one of Examples 15 to 18 which may contain the combination of descriptors including an average descriptor which is indicative of one or more data means before or after a data point of interest. The method may further include determining one or more means: (i) by determining the point of data of interest in the first well log; (ii) dividing data points before or after the data point of interest into one or more segments; and / or (iii) determining the one or more averages on the basis of the one or more segments. Each average of the one or more averages can correspond to a respective segment of the one or more segments and can be determined by averaging the respective data points in the respective segment.
Example 20: Method according to any one of Examples 15 to 19 which may contain the combination of descriptors including a histogram descriptor which is representative of frequencies in which slopes of a plurality of data points lie in a plurality of slope beaches. The method may include determining the frequencies in which the slopes of the plurality of data points lie in the plurality of ranges of slopes: (i) determining a plurality of slopes corresponding to the plurality of data points, each slope in the plurality of slopes corresponding to a respective data point in the plurality of data points; (ii) by determining the plurality of ranges of slopes, each range of
2016-IPM-100867-U1-EN 27 slopes of the plurality of ranges of slopes covering a predetermined number of degrees; (iii) associating each data point in the plurality of data points with a respective slope range of the plurality of slope ranges based on a slope associated with the data point; and / or (ïv) by determining a number of data points associated with each range of slopes of the plurality of ranges of slopes.
The foregoing description of certain examples, including illustrated examples, has been presented for purposes of illustration and description only, and is not intended to be exhaustive or to limit disclosure to the specific forms disclosed. Many modifications, adaptations and uses thereof will be apparent to those skilled in the art without departing from the scope of the disclosure.
2016-IPM-100867-U1-FR

权利要求:
Claims (15)
[1" id="c-fr-0001]
Claims
1. Non-transient computer-readable medium comprising a program code for automatically correlating well logs to identify specificities of a stratum represented in well logs, the program code being executable by a processing device to bring the device processing to:
receiving a plurality of well logs, each well log of the plurality of well logs indicating subterranean strata penetrated at various depths by a respective wellbore;
determining a combination of descriptors used to correlate a first data point in a first well log of the plurality of well logs to a second data point in a second well log of the plurality of well logs;
selecting different combinations of descriptors from a plurality of descriptors, a descriptor being a type of information about a data point in a well log;
by performing dynamic alignment on the first well log and the second well log using the different combinations of descriptors to generate a plurality of alignments between the first well log and the second well log, in which dynamic alignment includes manipulating a shape of a well log to align the well log with another well log; and analyzing the plurality of alignments to determine whether the combination of descriptors results in the first data point being correlated with the second data point; and correlating data points between other well logs in the plurality of well logs by performing dynamic alignment using the combination of descriptors.
[2" id="c-fr-0002]
The non-transient computer-readable medium of claim 1, further comprising program code executable by the processing device to cause the processing device to determine the combination of descriptors by selecting a value for a parameter corresponding to at least one descriptor in the combination of descriptors.
2016-IPM-100867-U1-FR 29
[3" id="c-fr-0003]
The non-transient computer-readable medium of claim 2, further comprising program code executable by the processing device to cause the processing device to determine the combination of descriptors by selecting a respective weight for each descriptor in the combination of descriptors.
[4" id="c-fr-0004]
The non-transient computer readable medium of claim 3, wherein the combination of descriptors comprises a shape descriptor which is indicative of one or more slopes in data before or after a data point of interest, and comprising furthermore, program code executable by the processing device to cause the processing device to determine the one or more slopes;
determining the data point of interest in the first well log;
dividing data points before or after the data point of interest into one or more segments; and determining the one or more slopes based on the one or more segments, each slope of the one or more slopes corresponding to a respective segment of the one or more segments and being determined by performing a linear regression on respective data points in the respective segment.
[5" id="c-fr-0005]
5. A non-transient computer readable medium according to claim 3, wherein the combination of descriptors comprises an average descriptor which is indicative of one or more data averages before or after the data point of interest, and comprising in addition to a program code executable by the processing device to bring the processing device to:
determining one or more averages on the basis of the one or more segments, each average of the one or more averages corresponding to a respective segment of the one or more segments and being determined by averaging respective data points in the respective segment,
[6" id="c-fr-0006]
The non-transient computer readable medium of claim 3, wherein the combination of descriptors comprises a histogram descriptor which is representative of frequencies in which slopes of a plurality of data points lie in a plurality of ranges of slopes, and further comprising
2016-IPM-100867-U1-EN 30 program code executable by the processing device to cause the processing device to determine the frequencies in which the slopes of the plurality of data points lie in the plurality of ranges of slopes :
determining a plurality of slopes corresponding to the plurality of data points, each slope in the plurality of slopes corresponding to a respective data point in the plurality of data points;
determining the plurality of slope ranges, each slope range of the plurality of slope ranges covering a predetermined number of degrees;
by associating each data point in the plurality of data points with a respective slope range of the plurality of slope ranges based on a slope associated with the data point; and determining a number of data points associated with each range of slopes of the plurality of ranges of slopes,
[7" id="c-fr-0007]
7. A non-transient computer readable medium according to any one of claims 1 to 6, further comprising a program code executable by the processing device to cause the processing device to determine the combination of descriptors:
identifying multiple combinations of descriptors that lead to the first data point being correlated to the second data point in an amount that falls within a predetermined tolerance range; and selecting the combination of descriptors from multiple combinations of descriptors based on the combination of descriptors resulting in a closer correlation between the first data point and the second data point than another combination of descriptors in the multiple combinations descriptors.
[8" id="c-fr-0008]
8. System for automatic correlation of well logs to identify specificities of a stratum represented in well logs, ie system;
a processing device; and a memory device on which instructions executable by the processing device are stored to cause the processing device to:
receiving a plurality of well logs, each of the well logs of the plurality of well logs indicating strata
2016-IPM-100867-U1-EN 31 subterranean penetrated at various depths by a respective wellbore;
determining a combination of descriptors used to correlate a first data point in a first well log of the plurality of well logs to a second data point in a second well log of the plurality of well logs:
selecting different combinations of descriptors from a plurality of descriptors, a descriptor being a type of information about a data point in a well log;
by performing dynamic alignment on the first well log and the second well log using the different combinations of descriptors to generate a plurality of alignments between the first well log and the second well log, in which dynamic alignment includes manipulating a form of a well log to align the well log with another well log; and analyzing the plurality of alignments to determine whether the combination of descriptors results in the first data point being correlated with the second data point; and correlating data points between other well logs in the plurality of well logs by performing dynamic alignment using the combination of descriptors.
[9" id="c-fr-0009]
The system of claim 8, wherein the memory device further comprises instructions executable by the processing device to cause the processing device to determine the combination of descriptors by selecting values for at least two parameters corresponding to at least a descriptor in the combination of descriptors.
[10" id="c-fr-0010]
10. The system of claims 8 and 9, wherein the combination of descriptors comprises an average descriptor which is indicative of one or more data means before or after a data point of interest, and wherein the device memory further comprises instructions executable by the processing device to cause the processing device to determine the one or more means:
determining the data point of interest in the first well log;
2016-IPM-100867-U1-FR 32 by dividing data points before or after the data point of interest into one or more segments; and determining one or more averages based on the one or more segments, each average of the one or more averages corresponding to a respective segment of the one or more segments and being determined by averaging respective data points in the respective segment.
[11" id="c-fr-0011]
The system of any of claims 8 and 9, wherein the combination of descriptors includes a histogram descriptor which is representative of frequencies in which slopes of a plurality of data points lie in a plurality of ranges of slopes, and wherein the memory device further comprises instructions executable by the processing device to cause the processing device to determine the frequencies in which the slopes of the plurality of data points lie in the plurality of ranges of slopes:
determining a plurality of slopes corresponding to the plurality of data points, each slope in the plurality of slopes corresponding to a respective data point in the plurality of data points;
determining the plurality of slope ranges, each slope range of the plurality of slope ranges covering a predetermined number of degrees;
associating each data point in the plurality of data points with a respective slope range of the plurality of slope ranges based on a slope associated with the data point; and determining a number of data points associated with each range of slopes of the plurality of ranges of slopes.
[12" id="c-fr-0012]
12. A method for automatic correlation of well logs to identify specificities of a stratum represented in well logs, the method comprising:
receiving, by a processing device, a plurality of well logs, each well log of the plurality of well logs indicating subterranean strata penetrated at various depths by a respective wellbore;
a determination, by a processing device, of a combination of descriptors used to correlate a first data point in a
2016-IPM-100867-U1-EN 33 first well log of the plurality of well logs at a second data point in a second well log of the plurality of well logs in:
selecting different combinations of descriptors from a plurality of descriptors, a descriptor being a type of information about a data point in a well log;
by performing dynamic alignment on the first well log and the second well log using the different combinations of descriptors to generate a plurality of alignments between the first well log and the second well log, in which dynamic alignment includes manipulating a form of a well log to align the well log with another well log; and analyzing the plurality of alignments to determine whether the combination of descriptors results in the first data point being correlated with the second data point; and correlating, by the processing device, data points between other well logs in the plurality of well logs by performing dynamic alignment using the combination of descriptors.
[13" id="c-fr-0013]
13. The method of claim 12, further comprising determining the combination of descriptors by selecting values for at least three parameters corresponding to at least one descriptor in the combination of descriptors.
[14" id="c-fr-0014]
14. The method according to claim 12, in which the combination of descriptors comprises a shape descriptor which is indicative of one or more slopes in data before or after a data point of interest, and further comprising a determination of one or more slopes:
determining the data point of interest in the first well log;
dividing data points before or after the data point of interest into one or more segments; and by determining the one or more slopes on the basis of the one or more segments, each slope of the one or more slopes corresponding to a respective segment of the one or more segments and being determined by carrying out a
2016-IPM-100867-U1-EN 34 linear regression on respective data points in the respective segment.
[15" id="c-fr-0015]
15. Method according to any one of claims 12 and 13, in which the combination of descriptors comprises a histogram descriptor which is
5 representative of frequencies in which slopes of a plurality of data points lie in a plurality of ranges of slopes, and further comprising determining frequencies in which the slopes of the plurality of data points lie in the plurality of slope beaches:
by determining a plurality of slopes corresponding to the plurality of
10 data points, each slope in the plurality of slopes corresponding to a respective data point in the plurality of data points;
determining the plurality of slope ranges, each slope range of the plurality of slope ranges covering a predetermined number of degrees;
by associating each data point in the plurality of data points
15 data to a respective slope range of the plurality of slope ranges based on a slope associated with the data point; and determining a number of data points associated with each range of slopes of the plurality of ranges of slopes.
2016-IPM-100867-U1-FR
1/9
1Ü2’J * 02 * '

类似技术:

公开号 | 公开日 | 专利标题

FR3064755A1|2018-10-05|AUTOMATED WELL DIAGRAPH CORRELATION USING DESCRIPTORS

EP2987003B1|2019-06-12|System and method for automatically correlating geologic tops

US20210150638A1|2021-05-20|System and Method for Automatically Correlating Geologic Tops

EP2761329B1|2016-01-13|Method for validating a training image for the multipoint geostatistical modeling of the subsoil

EP2628893A1|2013-08-21|Method for exploiting a deposit using a technique for selecting the positions of drilling wells

FR3034222A1|2016-09-30|

FR3036820A1|2016-12-02|MODELING THE SATURATION AND PERMEABILITY OF PETROLEUM FIELD RESERVOIR

FR3038338A1|2017-01-06|CORRECTION OF DEVIATION AND DISPERSION EFFECTS ON ACOUSTICAL DIAGRAM MEASUREMENTS OF WELLS DUE TO STRATIFIED FORMATIONS

WO2018020181A1|2018-02-01|Method for determining a maximum allowable volume of water that can be removed over time from an underground water source

EP3146367B1|2018-07-04|Method of determining a map of height of liquid hydrocarbon in a reservoir

FR3034894A1|2016-10-14|

FR3038408A1|2017-01-06|

FR3033432A1|2016-09-09|

EP2963235B1|2017-02-22|Method for exploiting an oil deposit based on a technique for positioning wells to be drilled

FR3063765A1|2018-09-14|CORRELATION OF STRATA SURFACES THROUGH WELL LOGS

FR3029664A1|2016-06-10|DEFINITION OF NON-LINEAR PETROFACIES FOR A RESERVOIR SIMULATION MODEL

EP3224653B1|2020-06-24|Method and device for processing well data

FR3049735A1|2017-10-06|VISUALIZATION OF ATTRIBUTES OF MULTIPLE SURFACES OF FAILURE IN REAL TIME

FR3044446A1|2017-06-02|SYSTEM AND METHOD FOR COMBINING A WORKFLOW

FR3087274A1|2020-04-17|CALIBRATION OF REPETITIVE SEISMIC IMAGES FOR PRODUCTION OPERATIONS

FR3074339A1|2019-05-31|SOURCE-WELL GEOLOGICAL DISPLAY AND ANALYSIS SYSTEM

FR3074340A1|2019-05-31|ANALYSIS OF THE PROVENANCE OF GEOLOGICAL SEDIMENTS AND DISPLAY SYSTEM

FR3110729A1|2021-11-26|Method for determining the minimum simulation, simulation method and associated device

FR3037992A1|2016-12-30|

FR3057979A1|2018-04-27|BRITISH CORRECTION IN MICROSISMIC EVENT DATA

同族专利:

公开号 | 公开日

US20190383133A1|2019-12-19|

GB2573708B|2021-10-20|

NO20191018A1|2019-08-23|

GB2573708A|2019-11-13|

GB201911265D0|2019-09-18|

US11220898B2|2022-01-11|

WO2018182691A1|2018-10-04|

引用文献:

公开号 | 申请日 | 公开日 | 申请人 | 专利标题

US7280932B2|2004-09-07|2007-10-09|Landmark Graphics Corporation|Method, systems, and computer readable media for optimizing the correlation of well log data using dynamic programming|

WO2009064732A1|2007-11-12|2009-05-22|Schlumberger Canada Limited|Wellbore depth computation|

US7873476B2|2008-07-02|2011-01-18|Chevron U.S.A. Inc.|Well log correlation|

US20140081613A1|2011-11-01|2014-03-20|Austin Geomodeling, Inc.|Method, system and computer readable medium for scenario mangement of dynamic, three-dimensional geological interpretation and modeling|

US11106185B2|2014-06-25|2021-08-31|Motive Drilling Technologies, Inc.|System and method for surface steerable drilling to provide formation mechanical analysis|

US9377547B2|2012-10-05|2016-06-28|Halliburton Energy Services, Inc.|Analyzing fracture stratigraphy|

US9664028B2|2012-12-19|2017-05-30|Halliburton Energy Services, Inc.|Systems and methods for look ahead resistivity measurement with offset well information|

US10459098B2|2013-04-17|2019-10-29|Drilling Info, Inc.|System and method for automatically correlating geologic tops|

US20150088424A1|2013-09-20|2015-03-26|Schlumberger Technology Corporation|Identifying geological formation depth structure using well log data|

CN104598705B|2013-10-31|2018-05-22|国际商业机器公司|For identifying the method and apparatus of subsurface material layer|

US10365261B2|2014-07-18|2019-07-30|Chevron U.S.A. Inc.|System and method for determining stratigraphic location and areal extent of total organic carbon using an integrated stratigraphic approach|

US10120343B2|2015-05-13|2018-11-06|Conocophillips Company|Time corrections for drilling data|

WO2017135972A1|2016-02-05|2017-08-10|Hitachi, Ltd.|System and method for well log data analysis|WO2021007271A1|2019-07-09|2021-01-14|Schlumberger Technology Corporation|Visually differentiating and merging oil-gas data using a mapping user interface|

GB2597021A|2019-08-26|2022-01-12|Landmark Graphics Corp|Performing dynamic time warping with null or missing data|

US20210238997A1|2020-01-30|2021-08-05|Landmark Graphics Corporation|Determination Of Representative Elemental Length Based On SubSurface Formation Data|

WO2022011107A1|2020-07-08|2022-01-13|Saudi Arabian Oil Company|Stochastic dynamic time warping for automated stratigraphic correlation|

法律状态:
2019-01-22| PLFP| Fee payment|Year of fee payment: 2 |

2020-08-14| PLSC| Publication of the preliminary search report|Effective date: 20200814 |

2020-11-13| ST| Notification of lapse|Effective date: 20201006 |

优先权:

申请号 | 申请日 | 专利标题

PCT/US2017/025405|WO2018182691A1|2017-03-31|2017-03-31|Automated well-log correlation using descriptors|

IBWOUS2017025405|2017-03-31|

[返回顶部]