CN116201535B

CN116201535B - Automatic dividing method, device and equipment for oil and gas reservoir target well sign stratum

Info

Publication number: CN116201535B
Application number: CN202310113990.9A
Authority: CN
Inventors: 任钰; 刘晓庆; 申瑞彩; 方杰; 张兴聪; 石任
Original assignee: Beijing Yuexin Times Technology Co ltd
Current assignee: Beijing Yuexin Times Technology Co ltd
Priority date: 2023-02-06
Filing date: 2023-02-06
Publication date: 2024-02-09
Anticipated expiration: 2043-02-06
Also published as: CN116201535A

Abstract

The application relates to an automatic dividing method, device and equipment for a target well sign stratum of a petroleum reservoir. Determining a reference well for the target well from well data for the known well and the target well based on well trajectories in the well data; then determining a selected mark stratum and a selected public curve according to the logging curves of the reference wells, stratum partition data and the logging curves of the target wells; then determining the weight of the selected public curve and the corresponding direction of the weight of each reference well and each target well to each selected public curve; the filter synthesis curves of the reference wells and the target wells are obtained through calculation, so that the strong and important characteristics of the selected public curves are better reserved, the redundant characteristics of the curves are eliminated, and the dividing result is more accurate; then according to the filter synthesis curve of each reference well, the method realizes the automatic division of the target well mark stratum by the morphological similarity of the filter synthesis curve of the target well and the selected mark stratum; the dividing efficiency of the target well sign stratum is improved.

Description

Automatic dividing method, device and equipment for oil and gas reservoir target well sign stratum

Technical Field

The application relates to the field of oil and gas reservoir stratum data processing, in particular to an automatic dividing method, device and equipment for oil and gas reservoir target well sign stratum.

Background

The division and comparison of stratum are used as the core technology of fine reservoir description, are the basis for developing the geological research of the oil and gas reservoirs, and are the key preconditions for the subsequent formulation of scientific and efficient oil and gas reservoir development schemes. The main working process is as follows: firstly, selecting known wells in a work area, then determining a structural grid of a key geological horizon through the combination of the known wells and seismic data, and carrying out stratum division work of each well on the basis.

The traditional stratum division mode is to observe the data of the target well manually, so that the division of the marked stratum is realized. When the stratum division comparison work of a plurality of scattered oil and gas reservoirs and thousands of wells in each block is faced, the stratum division mode is seriously dependent on the experience knowledge of business personnel on specific blocks, has the problems of high subjectivity, long time consumption, uneven results and the like, and brings huge working pressure to oil and gas reservoir evaluation personnel, oil reservoir researchers and logging interpreters.

Therefore, an intelligent and automatic mode is urgently needed to replace the traditional manual mode so as to reduce the workload of personnel and improve the efficiency of stratum division of the target well sign.

Disclosure of Invention

In order to solve the problems that the current stratum division efficiency is low and the experience knowledge of business personnel on specific blocks is seriously relied on, the application provides an automatic stratum division method, device and equipment for a target well sign of a petroleum reservoir.

In a first aspect, the method for automatically dividing the oil and gas reservoir target well sign stratum provided by the application adopts the following technical scheme:

an automatic demarcation method for a target well mark stratum of a hydrocarbon reservoir, comprising the following steps:

acquiring well data of a known well and a target well; the well data for the known well includes a first well trajectory, a first log, and first stratigraphic data; the well data of the target well includes a second well trajectory and a second log;

screening a reference well of the target well from the known wells according to a preset screening rule according to the first well track of the known well and the second well track of the target well;

acquiring a selected common curve and a selected mark stratum according to the third logging curve of each reference well, the third stratum partition data and the second logging curve of the target well; the third well logging curve is the same as the first well logging curve of the known well corresponding to the reference well; the third stratigraphic division data is the same as the first stratigraphic division data of the known well corresponding to the reference well;

Determining the weight of each selected public curve and the corresponding direction of the weight of each target well and each reference well to the selected public curve;

according to the weight corresponding direction of the reference well to each selected public curve and the weight of each selected public curve, carrying out weighted summation to obtain a third filtering synthetic curve of the reference well;

according to the weight corresponding direction of the target well to each selected public curve and the weight of each selected public curve, carrying out weighted summation to obtain a second filtering synthetic curve of the target well;

dividing the third filter synthesis curve according to the selected mark stratum corresponding to the reference well to obtain a reference curve segment corresponding to the selected mark stratum; selecting a sliding window curve segment by adopting a variable length sliding window detection method aiming at the second filtering synthesis curve; calculating a first curve form similarity between each sliding window curve segment and a reference curve segment; taking the stratum where the sliding window curve section meeting the first preset similarity threshold value is located as a candidate mark stratum of the target well;

and determining a final division result of the marker strata of the target well based on each candidate marker stratum.

By adopting the technical scheme, firstly, determining a reference well of a target well according to well data of a known well and the target well and well tracks in the well data; then determining a selected mark stratum according to the logging curves and stratum division data of each reference well; according to the public logging curves between each reference well and the target well, determining a selected public curve; then determining the weight of the selected public curve and the corresponding direction of the weight of each reference well and each target well to each selected public curve; the filter synthesis curves of the reference wells and the target wells are obtained through calculation, so that the strong and important characteristics of the selected public curves are better reserved, the redundant characteristics of the curves are eliminated, and the dividing result is more accurate; then according to the filter synthesis curve of each reference well, the method realizes the automatic division of the target well mark stratum by the morphological similarity of the filter synthesis curve of the target well and the selected mark stratum; the dividing efficiency of the target well sign stratum is improved.

Optionally, the acquiring well data of the known well and the target well includes acquiring the second well trajectory as follows: acquiring wellhead data and well deviation data of the target well; the wellhead data comprise wellhead XY coordinates, and the well deviation data comprise XY offset corresponding to different vertical depths;

According to the XY offset corresponding to the different vertical depths, the XY coordinates corresponding to the different vertical depths are obtained through calculation;

and sequentially connecting the XY coordinates corresponding to the different vertical depths to obtain a second well track of the target well.

By adopting the technical scheme, the well head data and the well deviation data of the target well are utilized to acquire the well track of the target well.

Optionally, the step of selecting the filter according to a preset screening rule includes:

segmenting the second well track according to the vertical depth direction to obtain a plurality of well sections;

finding m adjacent wells corresponding to each well section from the known wells; m is a natural number greater than or equal to 1;

calculating a set of adjacent wells corresponding to each well section;

and taking the adjacent wells in the collection as reference wells of the target well.

By adopting the technical scheme, the well track of the target well is segmented, adjacent wells are respectively determined for each segment, and then the adjacent well set corresponding to each segment is used as a reference well of the target well; when the target well is an inclined well, because the horizontal distances at the two ends of the well may be far apart, the problem that the error is large or an important reference well is omitted when the reference well is directly determined by a certain section or a certain point can be solved.

Optionally, the selecting mark stratum is obtained by the following method:

calculating a third curve morphology similarity between every two reference wells on the same logging curve for the same stratum according to the third logging curve of each reference well and third stratum division data;

counting a third curve form similarity mean value of the same stratum on each logging curve to be used as a third comprehensive similarity of the stratum; sequencing the third comprehensive similarity of each stratum according to the sequence from the big similarity to the small similarity;

selecting n stratum with the maximum similarity as the selected mark stratum; and n is a natural number greater than or equal to 1.

By adopting the technical scheme, the stratum with strong and obvious consistency in each reference well can be screened, and the reference well is relatively close to the target well in space distance, so that the reference well possibly has larger similarity in geological stratum, the selected marker stratum can be accurately used as the basis for dividing the subsequent marker stratum of the target well, and the favorable dividing effect can be verified based on actual test data.

Optionally, the selected public curve is obtained by the following method:

calculating a fourth curve morphology similarity between every two reference wells on the same logging curve for the same stratum according to the third logging curve and third stratum partition data of each reference well;

Determining a common log between the target well and the reference well;

counting a fourth curve form similarity mean value of the public logging curve on each stratum to be used as a fourth comprehensive similarity of the public logging curve;

sequencing the fourth comprehensive similarity of each public logging curve according to the sequence from the big similarity to the small similarity;

selecting k public logging curves with the maximum similarity as the selected public curves; and k is a natural number greater than or equal to 1.

By adopting the technical scheme, the public logging curves between the target well and the reference well are firstly determined, then the comprehensive similarity of the public logging curves on each stratum is counted, and then k public logging curves with the largest similarity are selected as the selected public curves, so that the selected public curves can better reflect stratum characteristics in the current area range (the selected reference well is usually near the target well), and the curve form consistency or similarity of the curves of each reference well on each stratum is higher, therefore, the selected public curves are used as one of the follow-up mark stratum division basis of the target well, and the accuracy of mark stratum division is facilitated.

Optionally, the determining the weight of each selected common curve includes:

and taking the fourth comprehensive similarity corresponding to the selected public curve as the weight of the fourth comprehensive similarity.

By adopting the technical scheme, the arithmetic average value of the form similarity of the fourth curve of the selected mark stratum on all stratum of each reference well is used as the corresponding fourth comprehensive similarity, and the fourth comprehensive similarity is used as the weight during filtering synthesis, so that the importance and the significance of the selected public curve on the stratum characteristics of the target well and the reference well can be accurately reflected.

Optionally, the weight corresponding directions of the target well and the reference well for the selected common curve are determined by the following modes:

determining the weight corresponding direction of the target well to each selected public curve and the weight corresponding direction of the reference well to each selected public curve by adopting a method based on signal-to-noise ratio energy maximization; the signal-to-noise ratio energy maximization-based method adopts the following objective function:

wherein k represents the number of the selected common curves, L represents the number of observation points of each selected common curve, and y _lk Data representing the first observation point on the kth selected common curve, w _k Representing the weighting factor; y is _k Representing the data average value of the kth selected public curve at each observation point;

by adjusting w _k The objective function tends to be maximum, and a weighting factor w of the objective well or the reference well to the selected public curve is obtained _k ，w _k And the positive and negative values are the directions of the target well or the reference well to the selected public curve.

By adopting the technical scheme, the corresponding directions of the weights of the target well and each reference well are accurately determined by using the signal-to-noise ratio energy maximization method, and the problem that the generated filter synthesis curve has opposite directions with the actual situation is avoided.

Optionally, the determining, based on each candidate marker formation, a final division of the marker formation for the target well includes:

the target well has at least one candidate mark stratum on each selected mark stratum; selecting one candidate mark stratum from each selected mark stratum to be combined to serve as an initial mark stratum dividing scheme of the target well; any two candidate marker strata in the initial partitioning scheme for each of the marker strata; calculating the distance between any two candidate mark strata; judging whether the distance meets stratum constraint conditions or not;

Calculating first comprehensive similarity of each candidate mark stratum in the mark stratum initial dividing scheme, and judging whether the first comprehensive similarity reaches a second preset similarity threshold; the first comprehensive similarity is calculated by the following method: determining a sliding window curve segment corresponding to each candidate mark stratum in the mark stratum initial dividing scheme, and obtaining a first curve form similarity between the sliding window curve segment and the reference curve segment; carrying out arithmetic average calculation on the first curve form similarity corresponding to each candidate mark stratum in the mark stratum initial division scheme to obtain the first comprehensive similarity of the mark stratum initial division scheme;

if the distance meets the stratum constraint condition and the first comprehensive similarity reaches the second preset similarity threshold value, reserving the initial dividing scheme of the marking stratum; otherwise, deleting the initial dividing scheme of the marking stratum;

and selecting the marker stratum with the highest first comprehensive similarity from the reserved initial dividing scheme of the marker stratum as a final dividing result of the marker stratum of the target well.

By adopting the technical scheme, the initial division result with the best division effect is screened out by adopting the global optimal thought and the formation constraint condition for the initial division scheme of each marking stratum and is used as the final division result of the marking stratum of the target well, so that the accuracy of marking stratum division is improved.

In a second aspect, the present application provides an automatic demarcation device for a target well mark stratum of a hydrocarbon reservoir, which adopts the following technical scheme:

an automatic demarcation device for a reservoir target well sign formation, comprising:

an acquisition module for acquiring well data of the known well and the target well; the well data for the known well includes a first well trajectory, a first log, and first stratigraphic data; the well data of the target well includes a second well trajectory and a second log;

the first processing module is used for screening the reference well of the target well from the known wells according to a preset screening rule according to the first well track of the known well and the second well track of the target well;

the second processing module is used for acquiring a selected public curve and a selected mark stratum according to the third logging curve of each reference well, the third stratum partition data and the second logging curve of the target well; the third well logging curve is the same as the first well logging curve of the known well corresponding to the reference well; the third stratigraphic division data is the same as the first stratigraphic division data of the known well corresponding to the reference well;

the determining module is used for determining the weight of each selected public curve and the corresponding direction of the weight of each target well and the reference well to the selected public curve;

The third processing module is used for carrying out weighted summation on the weight corresponding direction of each selected public curve and the weight of each selected public curve according to the reference well to obtain a third filtering synthetic curve of the reference well;

the fourth processing module is used for carrying out weighted summation according to the weight corresponding direction of the target well to each selected public curve and the weight of each selected public curve to obtain a second filtering synthetic curve of the target well;

the fifth processing module is used for dividing the third filtering synthesis curve according to the selected mark stratum corresponding to the reference well to obtain a reference curve segment corresponding to the selected mark stratum; selecting a sliding window curve segment by adopting a variable length sliding window detection method aiming at the second filtering synthesis curve; calculating a first curve form similarity between each sliding window curve segment and a reference curve segment; taking the stratum where the sliding window curve section meeting the first preset similarity threshold value is located as a candidate mark stratum of the target well;

and a sixth processing module, configured to determine a final division result of the marker strata of the target well based on each candidate marker stratum.

In a third aspect, the present application provides an apparatus for automatically dividing a target well mark stratum of a hydrocarbon reservoir, which adopts the following technical scheme:

The automatic dividing device for the oil and gas reservoir target well sign stratum comprises a processor, a memory and a computer program which is stored in the memory and can run on the processor, wherein the automatic dividing method for the oil and gas reservoir target well sign stratum is realized when the processor executes the computer program.

In summary, the present application includes at least the following beneficial technical effects:

1. the automatic division of the target well marking stratum is realized, and the division efficiency of the target well marking stratum is improved.

2. The arithmetic average value of the form similarity of the fourth curve of the selected mark stratum on all stratum of each reference well is used as the corresponding fourth comprehensive similarity, and the fourth comprehensive similarity is used as the weight in the filter synthesis process, so that the importance and the significance of the selected public curve in the aspect of stratum characteristics of the target well and the reference well can be accurately reflected.

3. The corresponding directions of the weights of the target well and each reference well are accurately determined by using a signal-to-noise ratio energy maximization method, so that the problem that the generated filter synthesis curve has opposite directions to the actual situation is avoided.

Drawings

FIG. 1 is a flow chart of a method for automatically partitioning a target well mark stratum of a hydrocarbon reservoir in an embodiment of the present application;

FIG. 2 is a schematic illustration of a well trajectory in an embodiment of the present application;

FIG. 3 is a schematic diagram of log and formation partition data between two reference wells according to an embodiment of the present application;

FIG. 4 is a block diagram of an apparatus for automatically dividing a target well mark stratum of a hydrocarbon reservoir in an embodiment of the present application;

FIG. 5 is a block diagram of an apparatus for automatically dividing a target well mark formation of a hydrocarbon reservoir in an embodiment of the present application.

Detailed Description

In order to make the objects, technical solutions and advantages of the present application more apparent, the present application will be further described in detail with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the present application.

Hydrocarbon reservoirs are traps that accumulate a certain amount of oil and gas. If only oil gathers in the trap, the pure oil reservoir (or oil reservoir) is called, and only natural gas gathers, the pure gas reservoir (or gas reservoir) is called. Well drilling surveys and data analysis (including but not limited to formation data) are required to investigate subsurface hydrocarbon containing conditions. Drilling exploration and data analysis are an important link in the development of oil and gas resources.

The embodiment of the application discloses an automatic dividing method for a target well mark stratum of a petroleum reservoir. Referring to fig. 1, the method comprises the steps of:

S101, acquiring well data of a known well and a target well.

A known well is a well whose associated well data is known. In embodiments of the present application, well data for known wells includes at least well trajectory (referred to herein as first well trajectory for ease of differentiation from other wells), log (referred to herein as first log), and stratigraphic data (referred to herein as first stratigraphic data).

And the target well is a well with unknown formation partition data. The target well is a well that requires or waits for stratigraphic division to be performed to obtain stratigraphic division data. In an embodiment of the present application, the well data of the target well includes a second well trajectory and a second log; but does not contain stratigraphic division data.

The well trajectory of the well may be represented by a continuous curve or broken line containing data in two dimensions, one being the vertical depth (i.e., the vertical depth of the well from the surface) and one being the XY coordinates corresponding to the vertical depth. The vertical depth and the XY coordinates are in one-to-one correspondence, and all the XY coordinates are sequentially connected according to the vertical depth order, so that the well track corresponding to the well can be obtained. Referring to fig. 2, the coordinate Z at the observation point indicates the vertical depth.

Assuming that the depth of a well is 1000 meters, an observation point is arranged at each interval of 1 meter, 1001 observation points (the head of the well calculates an observation point) exist in the well, and the XY coordinates of each observation point are sequentially connected to obtain a well track corresponding to the well.

Since the second well trajectory of the target well may not exist, the second well trajectory may be calculated by acquiring wellhead and well deviation data of the target well.

Specifically, the XY coordinates of the wellhead of the target well and the XY offset in the well deviation data are obtained, and the XY coordinates (X=X) corresponding to the current depth are calculated according to the XY offset of each depth in the well deviation data relative to the wellhead _{Wellhead for well} +X _{Offset of} ；Y＝Y _{Wellhead for well} +Y _{Offset of} ) And sequentially obtaining XY coordinates corresponding to each depth, and sequentially connecting the XY coordinates corresponding to each depth by adopting a straight line or a smooth curve to obtain a second well track of the target well.

The well logging curve comprises characteristic values corresponding to different depths, is obtained by measuring the geological properties of the well by adopting a related measurement technology, and is used for reflecting the curve that a certain characteristic changes along with the change of the geological type. The corresponding measurement technique can be flexibly selected to obtain the logging curve of the well based on the actual geological condition or exploration requirement of the well or the experience of measurement personnel.

Well logs include, but are not limited to, AC (for reflecting sonic moveout changes), C1 (for reflecting well diameter 1 changes), C2 (for reflecting well diameter 2 changes), CALI (for reflecting well diameter changes), CNL (for reflecting compensating neutron changes), DEN (for reflecting compensating density changes), droo (for reflecting density correction changes), DT (for reflecting sonic moveout changes, with AC), GR (for reflecting natural gamma changes), ILD (for reflecting deep induced resistivity changes), ILM (for reflecting medium induced resistivity changes), LLD (for reflecting deep lateral resistivity changes), LLS (for reflecting shallow lateral resistivity changes), MSFL (for reflecting microsphere focused resistivity changes), PEF (for reflecting photoelectric absorption section index changes), PHID (for reflecting density porosity changes), PHIN (for reflecting neutron porosity changes), PHIS (for reflecting sonic porosity changes), PHOB (for reflecting density changes), R1 (for reflecting 1 meter bottom gradient resistivity changes), RT (for reflecting 4 meter bottom gradient resistivity changes), RXO (for reflecting flushing zone resistivity changes).

The stratigraphic data comprises a stratigraphic name and a corresponding stratigraphic depth, wherein the stratigraphic depth is represented by a stratigraphic top depth and a stratigraphic bottom depth. For example, the formation split data is (a, 200, 300), then "a" may represent the formation name, "200" may represent the formation top depth to a depth of 200 meters from the surface, and "300" may represent the formation bottom depth to a depth of 300 meters from the surface. Based on this, a thickness of 100 meters (i.e., 300-200=100) of the formation may also be obtained.

In an alternative embodiment of the present application, after acquiring the well data, pre-processing of the well data is also included, including but not limited to longitudinal stitching, deletion of invalid values, curve attribution, structuring, log taking processing (i.e., log processing), and normalization processing. The pretreatment sequence can be longitudinal splicing, invalid value deleting, curve attribution, structuring, logarithmic processing and normalization processing in sequence. The specific process of pretreatment can be in the existing manner, and will not be described in detail here.

S102, screening a reference well of the target well from the known wells according to a preset screening rule according to the first well track of the known well and the second well track of the target well.

The purpose of reference well screening is to serve as a basis for subsequent marking of the formation division of the target well. Because of the complex geological conditions, the logging curve morphology of the adjacent well is often closer to that of the target well, so that when the reference well is screened, a spatial well trajectory analysis method is adopted, and the closest well is selected as the reference well of the target well based on spatial distance analysis.

In an alternative embodiment of the present application, when the target well is an inclined well, because horizontal distances between two ends of the well or between observation points may be far apart, directly determining a reference well by a certain section or a certain point may have a problem of large error or missing an important reference well, segmenting a well track of the target well, determining an adjacent well for each segment respectively, and then taking an adjacent well set corresponding to each segment as a reference well of the target well; thereby improving the accuracy of determining the reference well and avoiding the problem of missing important reference wells.

For example, the target well is cut into three sections according to the average depth, each section is analyzed according to the space well track, 3 nearest neighbor reference wells corresponding to each section are found, and finally, the corresponding reference wells of each section are combined to obtain the reference well C= { C1, C2, & gt. The distance between each section and the reference well is based on the shortest horizontal distance or average horizontal distance between the two.

In an alternative embodiment of the application, the reference well can be selected according to the curve integrity, the formation integrity and the formation thickness uniformity while the reference well is selected, and the wells with incomplete curve, incomplete formation and uneven formation thickness are filtered.

S103, acquiring a selected public curve and a selected mark stratum according to the third logging curve of each reference well, the third stratum partition data and the second logging curve of the target well.

Wherein, because the reference well is screened from the known wells, the third log is the same as the first log of the known well corresponding to the reference well; the third stratigraphic data is the same as the first stratigraphic data of the known well corresponding to the reference well.

In the embodiment of the application, the selected mark stratum is obtained by the following method:

calculating a third curve form similarity between every two reference wells on the same logging curve for the same stratum according to the third logging curve of each reference well and third stratum partition data; counting a third curve form similarity mean value of the same stratum on each logging curve to be used as a third comprehensive similarity of the stratum; sequencing the third comprehensive similarity of each stratum according to the sequence from the big similarity to the small similarity; selecting n stratum with the highest similarity as a selected mark stratum; wherein n is a natural number of 1 or more.

Referring to fig. 3, taking two reference wells as reference well B1 and reference well B2, respectively, firstly, performing curve morphology similarity calculation on the same log (i.e. the log names are the same), and assuming AC;

For the log curves AC of the reference well B1 and the log curves AC of the reference well B2, before the calculation of the similarity of the curve morphology, the log curves are segmented according to respective stratigraphic division data, for example, the log curves AC of the reference well B1 are segmented according to the stratigraphic division data thereof to obtain: a curved section s11 corresponding to the stratum (a, h11, h 12), a curved section s12 corresponding to the stratum (b, h12, h 13), and a curved section s13 corresponding to the stratum (c, h13, h 14);

for a logging curve AC of the reference well B2, segmenting according to stratum division data of the logging curve AC to obtain: a curved section s21 corresponding to the stratum (a, h21, h 22), a curved section s22 corresponding to the stratum (b, h22, h 23), a curved section s23 corresponding to the stratum (c, h23, h 24), and a curved section s24 corresponding to the stratum (d, h24, h 25);

similarity calculations are performed on curve segments of the same formation name for reference well B1 and reference well B2 on the same log, for example: calculating curve section s11 corresponding to stratum a on logging curve AC of reference well B1, and curve morphological similarity between curve section s21 corresponding to stratum a on logging curve AC of reference well B2; and a curve morphology similarity between curve segment s12 (i.e., the curve segment corresponding to formation B of reference well B1 on log AC) and curve segment s22 (the curve segment corresponding to formation B of reference well B2 on log AC), and a curve morphology similarity between curve segment s13 (i.e., the curve segment corresponding to formation c of reference well B1 on log AC) and curve segment s23 (the curve segment corresponding to formation c of reference well B2 on log AC); since the same stratum d does not exist between the reference well B1 and the reference well B2, calculation is not required;

Similarly, the third curve morphology similarity of the same stratum on the same logging curve can be calculated in the mode for the reference well B1 and the reference well B2 on other logging curves;

similarly, for the third curve form of the same stratum on the same logging curve between the other two reference wells, the calculation can be performed in the same way;

assume that the list of reference wells is: [ B1, B2], stratigraphic list: [ a, b, c ], log list: [ AC, CAL ], the third curve morphology similarity calculation results shown in Table 1 below were obtained:

TABLE 1

And counting a third curve morphology similarity mean value of the same stratum on each logging curve to be used as a third comprehensive similarity of the stratum. Taking the stratum a as an example, the third comprehensive similarity of the stratum is that the third curve form similarity of the reference wells B1 and B2 on the logging curve AC is 0.4, the third curve form similarity on the logging curve CAL is 0.6, and the average value of the third curve form similarity and the third curve form similarity is 0.5 (namely (0.4+0.6)/2); the third overall similarity for the similarly obtainable formation b is (0.3+0.4)/2=0.35; the third overall similarity for formation c is (0.5+0.6)/2=0.55.

In an alternative embodiment of the present application, the selected public curve is obtained by:

According to the third log curve and third stratum division data of each reference well, calculating a fourth curve form similarity (the fourth curve form similarity is substantially the same as the third curve form similarity and only is used for distinguishing two processes of selecting a mark stratum and selecting a public curve) on the same log curve for the same stratum between every two reference wells, wherein the data in the table 1 can be directly used in an actual data processing process without separate calculation); determining a common log between the target well and the reference well; counting a fourth curve form similarity mean value of the public logging curve on each stratum, and taking the fourth curve form similarity mean value as a fourth comprehensive similarity of the public logging curve; sequencing the fourth comprehensive similarity of each public logging curve according to the sequence from the big similarity to the small similarity; selecting k public logging curves with the maximum similarity as selected public curves; k is a natural number of 1 or more.

The common log between the target well and the reference well, i.e., the log where both exist. For example, the target well comprises a log AC, and the reference well B1 also comprises a log AC, then AC is the common curve of both.

Assuming that the target well corresponds to the reference well b= { B1, B2}, then the common curves of the target well and the reference well B1 are calculated, assuming that AC and CAL are included, the common curves AC of the target well and the reference well B2 are calculated, assuming that AC and CAL are included, and the common curves c=cal AC of the respective reference well and the target well are obtained by summing.

According to the calculated similarity of the third curve form (or the similarity of the fourth curve form) of the same stratum on the same logging curve, the average value of the similarity of the fourth curve form of the public logging curve on each stratum is counted (taking AC as an example, according to the data calculated in the table 1, the similarity of the fourth curve form of the AC on each stratum is counted, the similarity of B1 and B2 on the stratum a is 0.4, the similarity of B1 and B2 on the stratum B is 0.3, the similarity of B1 and B2 on the stratum c is 0.5, and the average value is calculated to obtain (0.4+0.3+0.5)/3=0.4) as the fourth comprehensive similarity of the public logging curve.

Sequencing the fourth comprehensive similarity of each public logging curve according to the sequence from the big similarity to the small similarity; and selecting k public logging curves with the maximum similarity as selected public curves. K is set to 5, for example. If the number of the common curves is less than 5, selecting all the curves as the selected curves.

In order to ensure that each reference well at least comprises one selected public curve, judging whether each reference well at least comprises one curve in the selected public curves, if a certain reference well does not comprise the selected public curve, continuing to expand the number of the selected curves according to the morphological similarity of the fourth curve until each reference well at least comprises one selected public curve.

S104, determining the weight of each selected public curve and the corresponding direction of the weight of each target well and the reference well to the selected public curve.

The vertical change of the logging curve is closely related to stratum phenomena and non-stratum factors, and the filtering synthesis treatment is to eliminate the local influence of individual singular data in the observed data and highlight the consistent and important characteristics in multiple observed data. And carrying out weighted fusion according to the principle of stable characteristic enhancement and interference characteristic weakening.

In order to better divide the marking stratum and obtain a more accurate dividing result, before dividing the marking stratum, filtering and synthesizing the target well and each reference well respectively, namely, the target well obtains a filtering and synthesizing curve, the reference well obtains a filtering and synthesizing curve, the filtering and synthesizing curve keeps consistent and important characteristics in each selected curve, redundant characteristics in the curve are eliminated, and the marking stratum dividing result can be obtained by further analyzing the form based on the filtering and synthesizing curve.

In the embodiment of the present application, before curve filtering synthesis is performed, the weight of the target machine for each selected common curve and the corresponding direction of the weight, and the weight of each reference well for each selected common curve and the corresponding direction of the weight need to be determined first.

In the embodiment of the application, the determination of the weight and the two dimensional information of the corresponding direction of the weight is realized by adopting different methods respectively. The weight values of the target well and each reference well for each selected public curve are the same, and the difference is that the corresponding directions of the weights are different.

The weight of the selected public curve is obtained by the following method: and calculating a fourth comprehensive similarity of the selected public curve, and taking the fourth comprehensive similarity as the weight of the selected public curve. The fourth comprehensive similarity is obtained by calculating an arithmetic average value according to the curve form similarity of the selected public curve on different stratum. Taking table 1 as an example, assuming that AC is a selected public curve, calculating the similarity of B1 and B2 on the a stratum to be 0.4, the similarity of B1 and B2 on the B stratum to be 0.3, and the similarity of B1 and B2 on the c stratum to be 0.5, and calculating the arithmetic average of the three to obtain (0.4+0.3+0.5)/3=0.4, namely, the fourth comprehensive similarity of the selected public curve AC.

The weight corresponding direction is determined by a method based on signal-to-noise ratio energy maximization. The signal-to-noise ratio energy maximization method adopts the following objective function:

where k represents the number of selected common curves, L represents the number of observation points of the selected common curves in the target well or the reference well, and y _lk Data representing the kth observation point of the target well or the reference well on the kth selected common curve, w _k Representing the weighting factor;representing the data average value of the kth selected public curve of the target well or the reference well at each observation point;

by adjusting w _k The objective function tends to the maximum value, and the weighting factor w of the objective well or the reference well to the selected common curve is obtained _k ，w _k And the positive and negative values are the directions of the target well or the reference well to the selected public curve.

Through the mode, the weight and the weight corresponding direction of the target well and the reference well to each selected public curve can be obtained respectively; and further obtaining respective filter synthesis curves of the target well and the reference well.

S105, carrying out weighted summation according to the weight corresponding direction of the reference well to each selected public curve and the weight of each selected public curve to obtain a third filtering synthetic curve of the reference well.

And S106, carrying out weighted summation according to the weight corresponding direction of the target well to each selected public curve and the weight of each selected public curve to obtain a second filter synthetic curve of the target well.

For example, the common curves are selected [ GR, CAL, RT ], the weights of the common curves are selected [0.1,0.8,0.1], the corresponding directions of the weights of the target wells are +, +, +, ], and the filter synthesis curves of the target wells are obtained by weighted summation calculation, wherein the filter synthesis curves are 0.1×gr+0.8cal+0.1rt.

In order to prevent the opposite directions of the two filtering synthesis curves from affecting the stratum division result, in an alternative embodiment of the present application, before curve filtering synthesis is performed, a judgment adjustment is performed on the determined direction corresponding to the weight.

For example, the weight corresponding direction of the target well is [ +, +, + ], and the weight corresponding direction of the reference well B1 is [ -, +, - ], indicating that the resulting filtered composite curves for the target well and the reference well B1 are opposite. For this situation, whether the two curves are opposite needs to be judged according to the curve weight directions corresponding to the target well and the reference well, and the specific method is to judge the number of opposite signs of the directions corresponding to the weights of the target well and the reference well, and if the opposite numbers exceed half of the number of the selected public curves, all the directions corresponding to the weights of the reference well need to be reversed, so that the final direction corresponding to the weights of the reference well is obtained. The direction of the weight correspondence of the reference well B1 is adjusted to +, -, + ], thereby obtaining a filter synthesis curve of 0.1 x gr-0.8cal+0.1rt for the reference well.

And S107, dividing the third filter synthesis curve according to the selected mark stratum corresponding to the reference well to obtain a reference curve segment corresponding to the selected mark stratum.

It should be appreciated that the selected marker formation contains the formation name and the formation depth (including the formation top depth and the formation bottom depth) and thus the third filtered synthetic curve may be segmented.

Assuming that the selected mark stratum comprises AC and CAL, determining stratum depth corresponding to the AC, and then intercepting a curve segment corresponding to the depth on the third filtering synthetic curve to be used as one of the reference curve segments; determining the stratum depth corresponding to the CAL, and then intercepting a curve section corresponding to the depth on the third filtering synthetic curve as another reference curve section; and sequentially obtaining corresponding reference curve segments on the third filtering synthetic curve for each selected mark stratum.

S108, selecting a sliding window curve segment by adopting a variable length sliding window detection method aiming at the second filtering synthesis curve.

The sliding window size of the sliding window can be changed by a variable length sliding window detection method. In the embodiment of the application, when the sliding window interception is performed on the second filtering synthesis curve of the target well, the current sliding window size can be determined according to the size of the reference curve segment, so that the current sliding window size is 3/4-5/4 of the size of the reference curve segment, the interval is 1/10 of the length of the reference curve segment, curve morphology similarity calculation can be better performed, and compared with a mode of randomly determining the sliding window size, the sliding window size determination method is beneficial to finding the sliding window which is more similar to the reference curve segment more efficiently, and therefore the efficiency of marking stratum division is improved.

It should be appreciated that there may be multiple reference wells, there may be multiple reference curve segments for each reference well, and the size of each reference curve segment may be different, and thus the sliding window size may be dynamically variable.

S109, calculating the similarity of the first curve morphology between each sliding window curve segment and the reference curve segment.

The similarity of the shapes of the two curves can be calculated in any existing mode, and the embodiment is not limited to the method.

S110, taking the stratum where the sliding window curve section meeting the first preset similarity threshold value is located as a candidate mark stratum of the target well.

It should be understood that the first preset similarity threshold may be flexibly set according to actual requirements, which is not limited in this embodiment. In an alternative embodiment of the present application, when the similarity between all sliding windows and a reference curve segment is smaller than a first preset similarity threshold after the sliding window detection is completed between the second filtering synthetic curve and the reference curve segment, one sliding window with the largest similarity value may be selected as a candidate sliding window corresponding to the reference curve segment, and the stratum where the candidate sliding window is located may be used as a candidate mark stratum of the target well. So as to ensure that at least one candidate mark stratum exists on each selected mark stratum of the target well, thereby realizing the division work of the mark stratum.

And S111, determining a final division result of the marking stratum of the target well based on each candidate marking stratum.

In an alternative embodiment of the present application, the target well has at least one candidate marker formation on each selected marker formation; selecting one candidate mark stratum from each selected mark stratum to be combined to serve as an initial mark stratum dividing scheme of the target well;

any two candidate marker strata in the scheme are initially divided for each marker stratum; calculating the distance between any two candidate mark strata; judging whether the distance meets stratum constraint conditions or not;

calculating first comprehensive similarity of each candidate mark stratum in the mark stratum initial dividing scheme, and judging whether the first comprehensive similarity reaches a second preset similarity threshold;

if the distance meets the stratum constraint condition and the first comprehensive similarity reaches a second preset similarity threshold, reserving an initial dividing scheme of the marked stratum; otherwise, deleting the initial dividing scheme of the marking stratum;

and selecting the marker stratum with the highest first comprehensive similarity as a final dividing result of the target well from the initial dividing schemes of all the maintained marker strata.

The first comprehensive similarity is calculated by the following steps: determining a sliding window curve section corresponding to each candidate mark stratum in an initial mark stratum dividing scheme, and obtaining a first curve form similarity between the sliding window curve section and a reference curve section; and carrying out arithmetic average calculation on the first curve form similarity corresponding to each candidate mark stratum in the mark stratum initial dividing scheme to obtain the first comprehensive similarity of the mark stratum initial dividing scheme.

In addition, it should be understood that the second preset similarity threshold may be flexibly set according to actual requirements, which is not limited in this embodiment.

Assume that a flag layer is selected: [ A, B, C ], after the variable length sliding window detection, a candidate window [ A1: [100,103,0.7], A2 ] of A is obtained: [101,104.5,0.8]; candidate window of B [ B1: [133.5,135,0.84], B2: [133,136,0.78]; candidate window of C [ C1: [145,147,0.9], C2: [143,145,0.8]; the three values in the candidate window are the depth corresponding to the upper boundary and the depth corresponding to the lower boundary respectively, and the similarity of the data in the candidate window and the reference curve segment.

There are now 2 x 2 = 8 marker formation initial partitioning schemes:

【A1,B1,C1】、【A1,B1,C2】、【A1,B2,C1】、【A1,B2,C2】、【A2,B1,C1】、【A2,B1,C2】、【A2,B2,C1】、【A2,B2,C2】；

firstly, screening stratum meeting stratum constraint conditions, taking [ A1, B1 and C1 ] as an example, wherein A1 is [100,103,0.7], B1 is [133.5,135,0.84] and C1 is [145,147,0.9], firstly calculating the distance D1=133.5-103=30.5 between A1 and B1, then calculating the distance L1 between corresponding stratum B and A on a corresponding reference well, and if L1 is 6/5> D1> L1 is 4/5, the scheme meets the stratum constraint conditions, and reserving the scheme; then, calculating whether the distance between C1 and B1 meets the requirement, if so, reserving [ A1, B1, C1 ]; then, a first comprehensive similarity = (0.7+0.84+0.9)/3 of [ A1, B1, C1 ] is calculated, and if a second preset similarity threshold is reached, the scheme is reserved; otherwise, the scheme is deleted. By the method based on the global optimization idea and the stratum constraint conditions, all the marker stratum initial dividing schemes meeting the requirements are calculated, and one of the marker stratum final dividing results with the highest first comprehensive similarity is selected as the target well.

According to the embodiment of the application, the target well is taken as the center, the reference well found through the target well is used for pointedly selecting the corresponding selected public curve and the selected mark stratum, so that the consistency of the selected public curve and the selected mark stratum on the target well and the reference well can be improved, and the division effect of the mark stratum can be improved.

According to the embodiment of the application, through Python programming, experiments are carried out by using the existing 5 blocks, self-adaptive parameter adjustment can be carried out on different blocks, and well division results with higher curve form consistency are better.

In order to better implement the above method, the embodiment of the application also provides an intelligent marker stratum positioning and dividing device, which can be specifically integrated in an intelligent marker stratum positioning and dividing device, such as a terminal or a server, and the terminal can include, but is not limited to, a mobile phone, a tablet computer, a desktop computer or the like.

Fig. 4 is a block diagram of an automatic dividing device for a target well sign stratum of a hydrocarbon reservoir according to an embodiment of the present application, where the device mainly includes:

an acquisition module 41 for acquiring well data of the known well and the target well; well data for a known well includes a first well trajectory, a first log, and first stratigraphic data; the well data for the target well includes a second well trajectory and a second log.

A first processing module 42, configured to screen a reference well of the target well from the known wells according to a preset screening rule based on the first well trajectory of the known well and the second well trajectory of the target well.

A second processing module 43, configured to obtain an option public curve and an option mark stratum according to the third log curves of the reference wells, the third formation partition data and the second log curves of the target wells; the third log is the same as the first log of the known well corresponding to the reference well; the third stratigraphic data is the same as the first stratigraphic data of the known well corresponding to the reference well.

The determining module 44 is configured to determine a weight of each selected common curve, and a direction corresponding to the weight of each of the target well and the reference well with respect to the selected common curve.

And the third processing module 45 is configured to perform weighted summation according to the direction of the reference well corresponding to the weight of each selected public curve and the weight of each selected public curve to obtain a third filter synthesis curve of the reference well.

A fourth processing module 46, configured to perform weighted summation according to the direction corresponding to the weight of the target well for each selected common curve and the weight of each selected common curve to obtain a second filter synthesis curve of the target well;

A fifth processing module 47, configured to divide the third filter synthesis curve according to the selected flag stratum corresponding to the reference well to obtain a reference curve segment corresponding to the selected flag stratum; selecting a sliding window curve segment by adopting a variable length sliding window detection method aiming at the second filtering synthesis curve; calculating the similarity of a first curve form between each sliding window curve segment and a reference curve segment; and taking the stratum where the sliding window curve section meeting the first preset similarity threshold value is located as a candidate mark stratum of the target well.

A sixth processing module 48 is configured to determine a final demarcation result of the target well's formation based on each candidate formation.

The various changes and specific embodiments in the method provided in the foregoing embodiments are also applicable to the automatic oil and gas reservoir target well marking stratum partitioning device in this embodiment, and by the foregoing detailed description of the method for automatically partitioning an oil and gas reservoir target well marking stratum, those skilled in the art may clearly know the implementation method of the automatic oil and gas reservoir target well marking stratum partitioning device in this embodiment, which is not described in detail herein for brevity of description.

In order to better execute the program of the method, the embodiment of the application further provides an automatic oil and gas reservoir target well sign stratum partitioning device, as shown in fig. 5, which comprises a processor 51, a memory 52 and a computer program stored in the memory 52 and capable of running on the processor 51, wherein the automatic oil and gas reservoir target well sign stratum partitioning method is implemented when the processor 51 executes the computer program.

The automatic demarcation device for the oil and gas reservoir target well sign stratum can be implemented in various forms, including but not limited to mobile phones, tablet computers, palm top computers, notebook computers, desktop computers and the like.

Wherein the memory 52 may be used to store instructions, programs, code sets, or instruction sets. The memory 52 may include a stored program area and a stored data area, where the stored program area may store instructions for implementing an operating system, instructions for at least one function (such as determining a selected common curve, determining a selected flag formation or determining a weight and a weight corresponding direction of the selected common curve, implementing curve filter synthesis, etc.), and instructions for implementing the method for automatically partitioning a target well flag formation of a hydrocarbon reservoir provided by the above embodiments; the storage data area can store data and the like related to the automatic dividing method of the oil and gas reservoir target well sign stratum.

Processor 51 may include one or more processing cores. The processor 51 performs various functions of the present application and processes data by executing or executing instructions, programs, code sets, or instruction sets stored in the memory 52, calling data stored in the memory 52. The processor 51 may be at least one of an application specific integrated circuit (Application Specific Integrated Circuit, ASIC), a digital signal processor (Digital Signal Processor, DSP), a digital signal processing device (Digital Signal Processing Device, DSPD), a programmable logic device (Programmable Logic Device, PLD), a field programmable gate array (Field Programmable Gate Array, FPGA), a central processing unit (Central Processing Unit, CPU), a controller, a microcontroller, and a microprocessor. It will be appreciated that the electronic device for implementing the functions of the processor 51 may be other for different apparatuses, and the embodiments of the present application are not specifically limited.

Embodiments of the present application also provide a computer-readable storage medium, for example, including: a U-disk, a removable hard disk, a Read Only Memory (ROM), a random access Memory (Random Access Memory, RAM), a magnetic disk, or an optical disk, or other various media capable of storing program codes. The computer readable storage medium stores a computer program that can be loaded by a processor and that performs the method of automatically partitioning a reservoir target well marker formation of the above-described embodiments.

The foregoing embodiments are only used for describing the technical solution of the present application in detail, but the descriptions of the foregoing embodiments are only used for helping to understand the method and the core idea of the present application, and should not be construed as limiting the present application. Variations or alternatives that are readily contemplated by those skilled in the art within the scope of the present disclosure are intended to be encompassed within the scope of the present disclosure.

Claims

1. The automatic dividing method for the oil and gas reservoir target well sign stratum is characterized by comprising the following steps of:

the selected public curve is obtained by the following steps: calculating a fourth curve morphology similarity between every two reference wells on the same logging curve for the same stratum according to the third logging curve and third stratum partition data of each reference well; determining a common log between the target well and the reference well; counting a fourth curve form similarity mean value of the public logging curve on each stratum to be used as a fourth comprehensive similarity of the public logging curve; sequencing the fourth comprehensive similarity of each public logging curve according to the sequence from the big similarity to the small similarity; selecting k public logging curves with the maximum similarity as the selected public curves; k is a natural number greater than or equal to 1;

Determining the weight of each selected public curve and the corresponding direction of the weight of each target well and each reference well to the selected public curve; wherein the determining the weight of each selected public curve comprises: taking the fourth comprehensive similarity corresponding to the selected public curve as the weight of the fourth comprehensive similarity;

2. The method of automatically demarcating a reservoir target well sign formation of claim 1, wherein the acquiring well data for known wells and target wells comprises acquiring the second well trajectory as follows:

acquiring wellhead data and well deviation data of the target well; the wellhead data comprise wellhead XY coordinates, and the well deviation data comprise XY offset corresponding to different vertical depths;

3. The method for automatically partitioning a hydrocarbon reservoir target well marking formation according to claim 1, wherein the step of selecting according to a preset screening rule comprises:

calculating a set of adjacent wells corresponding to each well section;

4. A method for automatically partitioning a target well marking formation for a hydrocarbon reservoir according to any one of claims 1 to 3, wherein the selected marking formation is obtained by:

counting a third curve form similarity mean value of the same stratum on each logging curve to be used as a third comprehensive similarity of the stratum;

sequencing the third comprehensive similarity of each stratum according to the sequence from the big similarity to the small similarity;

5. The method for automatically partitioning a hydrocarbon reservoir target well sign formation according to claim 1, wherein the weight correspondence direction of each of the target well and the reference well for the selected common curve is determined by:

Wherein k represents the number of the selected common curves, L represents the number of observation points of each selected common curve, and y _lk Data representing the first observation point on the kth selected common curve, w _k Representing the weighting factor;representing the data average value of the kth selected public curve at each observation point;

6. The method of automatically partitioning a target well marking formation for a hydrocarbon reservoir of claim 1, wherein said determining a final partitioning result of the marking formation for the target well based on each of said candidate marking formations comprises:

the target well has at least one candidate mark stratum on each selected mark stratum; selecting one candidate mark stratum from each selected mark stratum to be combined to serve as an initial mark stratum dividing scheme of the target well;

any two candidate marker strata in the initial partitioning scheme for each of the marker strata; calculating the distance between any two candidate mark strata; judging whether the distance meets stratum constraint conditions or not;

7. An automatic demarcation device for a reservoir target well sign stratum, comprising:

the second processing module is used for acquiring a selected public curve and a selected mark stratum according to the third logging curve of each reference well, the third stratum partition data and the second logging curve of the target well; the third well logging curve is the same as the first well logging curve of the known well corresponding to the reference well; the third stratigraphic division data is the same as the first stratigraphic division data of the known well corresponding to the reference well; the selected public curve is obtained by the following steps: calculating a fourth curve morphology similarity between every two reference wells on the same logging curve for the same stratum according to the third logging curve and third stratum partition data of each reference well; determining a common log between the target well and the reference well; counting a fourth curve form similarity mean value of the public logging curve on each stratum to be used as a fourth comprehensive similarity of the public logging curve; sequencing the fourth comprehensive similarity of each public logging curve according to the sequence from the big similarity to the small similarity; selecting k public logging curves with the maximum similarity as the selected public curves; k is a natural number greater than or equal to 1;

The determining module is used for determining the weight of each selected public curve and the corresponding direction of the weight of each target well and the reference well to the selected public curve; wherein the determining the weight of each selected public curve comprises: taking the fourth comprehensive similarity corresponding to the selected public curve as the weight of the fourth comprehensive similarity;

8. An automatic demarcation device for a reservoir target well sign formation, comprising a processor, a memory and a computer program stored in the memory and operable on the processor, wherein the processor, when executing the computer program, implements the automatic demarcation method for a reservoir target well sign formation according to any one of claims 1 to 6.