CN112379440A - Method and device for identifying solid filling characteristics of fracture and hole body - Google Patents
Method and device for identifying solid filling characteristics of fracture and hole body Download PDFInfo
- Publication number
- CN112379440A CN112379440A CN202011203151.9A CN202011203151A CN112379440A CN 112379440 A CN112379440 A CN 112379440A CN 202011203151 A CN202011203151 A CN 202011203151A CN 112379440 A CN112379440 A CN 112379440A
- Authority
- CN
- China
- Prior art keywords
- data
- seismic
- target area
- seismic data
- different component
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 239000007787 solid Substances 0.000 title claims abstract description 59
- 238000000034 method Methods 0.000 title claims abstract description 41
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 53
- 238000005553 drilling Methods 0.000 claims abstract description 38
- 230000006870 function Effects 0.000 claims abstract description 36
- 230000002860 competitive effect Effects 0.000 claims abstract description 17
- 238000004364 calculation method Methods 0.000 claims abstract description 14
- 238000004590 computer program Methods 0.000 claims description 16
- 239000013598 vector Substances 0.000 claims description 15
- 238000000354 decomposition reaction Methods 0.000 claims description 8
- 238000007781 pre-processing Methods 0.000 claims description 7
- 238000003860 storage Methods 0.000 claims description 7
- 230000009467 reduction Effects 0.000 claims description 6
- 238000005457 optimization Methods 0.000 claims description 4
- 230000001105 regulatory effect Effects 0.000 claims description 4
- 238000010586 diagram Methods 0.000 description 13
- 238000004458 analytical method Methods 0.000 description 12
- 230000008569 process Effects 0.000 description 8
- 230000004044 response Effects 0.000 description 6
- 239000011324 bead Substances 0.000 description 5
- 238000011161 development Methods 0.000 description 5
- 239000012530 fluid Substances 0.000 description 5
- 238000012545 processing Methods 0.000 description 5
- 230000007547 defect Effects 0.000 description 4
- 238000004519 manufacturing process Methods 0.000 description 4
- BVKZGUZCCUSVTD-UHFFFAOYSA-L Carbonate Chemical compound [O-]C([O-])=O BVKZGUZCCUSVTD-UHFFFAOYSA-L 0.000 description 3
- 239000007789 gas Substances 0.000 description 3
- 239000003921 oil Substances 0.000 description 3
- 208000035126 Facies Diseases 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 230000003247 decreasing effect Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- VNWKTOKETHGBQD-UHFFFAOYSA-N methane Chemical compound C VNWKTOKETHGBQD-UHFFFAOYSA-N 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 239000011435 rock Substances 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000005856 abnormality Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000007621 cluster analysis Methods 0.000 description 1
- 230000001186 cumulative effect Effects 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000000556 factor analysis Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 239000003345 natural gas Substances 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000000513 principal component analysis Methods 0.000 description 1
- 230000001629 suppression Effects 0.000 description 1
- 238000001308 synthesis method Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 238000000844 transformation Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
- G01V1/00—Seismology; Seismic or acoustic prospecting or detecting
- G01V1/40—Seismology; Seismic or acoustic prospecting or detecting specially adapted for well-logging
- G01V1/44—Seismology; Seismic or acoustic prospecting or detecting specially adapted for well-logging using generators and receivers in the same well
- G01V1/48—Processing data
- G01V1/50—Analysing data
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
- G01V2210/00—Details of seismic processing or analysis
- G01V2210/60—Analysis
- G01V2210/61—Analysis by combining or comparing a seismic data set with other data
- G01V2210/616—Data from specific type of measurement
- G01V2210/6169—Data from specific type of measurement using well-logging
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
- G01V2210/00—Details of seismic processing or analysis
- G01V2210/60—Analysis
- G01V2210/62—Physical property of subsurface
- G01V2210/624—Reservoir parameters
Landscapes
- Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Acoustics & Sound (AREA)
- Environmental & Geological Engineering (AREA)
- Geology (AREA)
- Remote Sensing (AREA)
- General Life Sciences & Earth Sciences (AREA)
- General Physics & Mathematics (AREA)
- Geophysics (AREA)
- Geophysics And Detection Of Objects (AREA)
Abstract
The invention provides a method and a device for identifying solid filling characteristics of a slot body, wherein the method comprises the following steps: extracting attributes from seismic data volumes with different waveforms and different frequencies of the three-dimensional seismic data, and performing characteristic selection to obtain a sample set; continuously calculating a winning center and a punishment center of each input sample in the sample set by utilizing an opponent punishment competitive learning RPCL algorithm until a cost function is met, and finishing the continuous calculation to obtain different component combination data; and verifying the combined data of different components according to the actual drilling logging data to obtain the solid filling characteristics of the fracture-cavity body in the target area. The extracted attributes are subjected to feature selection to obtain a sample set, so that a new seismic data volume is automatically formed to avoid manual intervention; a cost function is introduced into an RPCL algorithm, convergence is accelerated, the existence of a dead zone is overcome, the solid filling characteristic of the fracture-cavity body can be effectively identified, and therefore the identification precision of the solid filling of the fracture-cavity body is improved.
Description
Technical Field
The invention relates to the technical field of natural gas exploration and development, in particular to a method and a device for identifying solid filling characteristics of a slot body.
Background
The Ordovician heterogeneous carbonate fracture-cave reservoir is one of the key areas of the top production of Tarim oil fields. The carbonate fracture-cavity body reservoir appears as a "beaded" reflection on the seismic section. Such reflection characteristics match well with high and stable producing wells. The well drilling designed aiming at the carbonate rock fracture-cave reservoir stratum has multiple developmental engineering abnormalities at a target point, such as drilling fluid overflow, leakage and even development emptying. However, with the continuous deepening of exploration and development, wells deployed in a strong bead-string-shaped reflection type are drilled in the drilling process to reveal different degrees of rock filling, and some wells do not obtain effective industrial oil and gas flow, even directly cause drilling failure.
In recent years, three-dimensional geological recognition techniques have gained increasing importance in the analysis of large complex seismic data in exploration, and the recognition results have relied primarily on the analysis of seismic interpretations. Such as root mean square amplitude attribute analysis, seismic reflection energy intensity attribute analysis, instantaneous amplitude attribute analysis, etc., as well as prestack elastic inversion, geostatistical inversion, etc., based on the original seismic data. However, as seismic exploration technology develops, the description of the geological characteristics of the three-dimensional filling crack-cave becomes more complex and more detailed, and the filling of a crack body cannot be identified through simple analysis.
The identification of the filling of the fracture and cavern body is considered as a multi-attribute seismic data clustering process, the main analysis methods at present mainly comprise a competitive learning algorithm, a principal component analysis and discriminant factor analysis, a texture analysis, a neural network and the like, and in the analysis methods, each data sample is generally defined by unique combination of physical and geometric attributes and is considered as a multi-dimensional vector. The algorithms mainly take a K-means algorithm and a self-organizing map network (SOM) as the main ones, and can be used for automatic cluster analysis and displaying the statistical characteristics of a sample set. However, there are still some drawbacks in typical competitive learning algorithms. For example, in the K-means algorithm, the appropriate number of clusters must be determined before clustering, while in the self-organizing neural network, seismic interpretation still requires manual intervention. Furthermore, due to the use of the winner take-all (WTA) learning approach, two competing learning algorithms may encounter a so-called dead zone problem: if a hub is improperly activated, it may never be modified. The adversary punishment competitive learning (RPCL) algorithm solves the dead zone problem and classifies without an accurate cluster number. However, since the convergence rate of the classical RPCL algorithm is slow, and the distance between the sample and the weight center is measured by the Euclidean distance, the defects limit the applicable field of the algorithm and cannot effectively identify the solid filling characteristics of the fracture-cavity body.
Disclosure of Invention
The embodiment of the invention provides a method for identifying solid filling characteristics of a fracture-cavity body, which is used for avoiding the defects of manual intervention and dead zones, effectively identifying the solid filling characteristics of the fracture-cavity body and improving the identification precision of the solid filling of the fracture-cavity body, and comprises the following steps:
acquiring three-dimensional seismic data of a target area;
extracting attributes from seismic data volumes with different waveforms and different frequencies of the three-dimensional seismic data of the target area, and performing feature selection on the extracted attributes to obtain a sample set;
continuously calculating a winning center and a punishment center of each input sample in the sample set by utilizing an opponent punishment competitive learning RPCL algorithm until a cost function is met, and finishing the continuous calculation to obtain different component combination data;
and verifying the different component combined data according to the actual drilling logging data, and obtaining the solid filling characteristics of the fracture-cavity body in the target area according to the verified different component combined data.
The embodiment of the invention also provides a device for identifying the solid filling characteristics of the fracture-cavity body, which is used for avoiding the defects of manual intervention and dead zones, effectively identifying the solid filling characteristics of the fracture-cavity body and improving the identification precision of the solid filling of the fracture-cavity body, and comprises the following components:
the data acquisition module is used for acquiring three-dimensional seismic data of a target area;
the sample optimization selection module is used for extracting attributes from seismic data bodies with different waveforms and different frequencies of the three-dimensional seismic data of the target area, and performing feature selection on the extracted attributes to obtain a sample set;
the competitive learning operation module is used for continuously calculating a winning center and a punishment center of each input sample in the sample set by utilizing an opponent punishment competitive learning RPCL algorithm until a cost function is met, and finishing continuous calculation to obtain different component combination data;
and the filling characteristic determining module is used for verifying the different component combined data according to the actual drilling logging data and obtaining the solid filling characteristics of the fracture-cavity body in the target area according to the verified different component combined data.
The embodiment of the invention also provides computer equipment which comprises a memory, a processor and a computer program which is stored on the memory and can run on the processor, wherein the processor executes the computer program to realize the fracture-cavity solid filling characteristic identification method.
The embodiment of the invention also provides a computer readable storage medium, which stores a computer program for executing the identification method of the solid filling characteristics of the fracture-cavity body.
In the embodiment of the invention, three-dimensional seismic data of a target area are obtained; extracting attributes from seismic data volumes with different waveforms and different frequencies of the three-dimensional seismic data of the target area, and performing feature selection on the extracted attributes to obtain a sample set; continuously calculating a winning center and a punishment center of each input sample in the sample set by utilizing an opponent punishment competitive learning RPCL algorithm until a cost function is met, and finishing the continuous calculation to obtain different component combination data; and verifying the different component combined data according to the actual drilling logging data, and obtaining the solid filling characteristics of the fracture-cavity body in the target area according to the verified different component combined data. The extracted attributes are subjected to feature selection to obtain a sample set, so that a new seismic data volume is automatically formed to avoid manual intervention; a cost function is introduced into an opponent punishment competition learning RPCL algorithm, convergence of the RPCL algorithm is accelerated, so that the existence of a dead zone is overcome, solid filling characteristics of the seam hole body can be effectively identified, and the identification precision of the solid filling of the seam hole body is improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
Fig. 1 is a schematic diagram of a solid filling characteristic identification method for a slot body in the embodiment of the invention.
Fig. 2 is a schematic diagram of a method for identifying solid filling characteristics of a slot body in an embodiment of the invention.
Fig. 3 is a schematic diagram of a specific implementation method of step 102 in an embodiment of the present invention.
Fig. 4 is a schematic diagram of a specific implementation method of step 104 in an embodiment of the present invention.
FIG. 5 is a schematic cross-sectional comparison of various seismic data and raw seismic data based on the identification of the M-RPCL algorithm in an implementation of one embodiment of the present invention.
FIG. 6 is a schematic diagram of a comparison of reservoir prediction planes based on original seismic data and data code number 8 seismic data identified by the M-RPCL algorithm and residuals thereof in an implementation of a particular application of the present invention.
FIG. 7 is a comparison of the effect of cross-well profiles of raw seismic data and data reservoir predictions identified based on the M-RPCL algorithm in an implementation of an embodiment of the present invention.
FIG. 8 is a cross-sectional comparison of raw seismic data and final component-optimized data in the practice of one embodiment of the present invention.
Fig. 9 is a schematic view of a solid filling characteristic recognition device of a slot body in the embodiment of the invention.
Fig. 10 is a schematic view of a solid filling characteristic recognition device of a slot body in an embodiment of the invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The embodiment of the invention provides a method for identifying solid filling characteristics of a fracture-cavity body, which is used for avoiding the defects of manual intervention and dead zones, effectively identifying the solid filling characteristics of the fracture-cavity body and improving the identification precision of the solid filling of the fracture-cavity body, and comprises the following steps of:
step 101: acquiring three-dimensional seismic data of a target area;
step 102: extracting attributes from seismic data volumes with different waveforms and different frequencies of the three-dimensional seismic data of the target area, and performing feature selection on the extracted attributes to obtain a sample set;
step 103: continuously calculating a winning center and a punishment center of each input sample in the sample set by utilizing an opponent punishment competitive learning RPCL algorithm until a cost function is met, and finishing the continuous calculation to obtain different component combination data;
step 104: and verifying the different component combined data according to the actual drilling logging data, and obtaining the solid filling characteristics of the fracture-cavity body in the target area according to the verified different component combined data.
As can be known from the flow shown in fig. 1, in the embodiment of the present invention, three-dimensional seismic data of a target area is obtained; extracting attributes from seismic data volumes with different waveforms and different frequencies of the three-dimensional seismic data of the target area, and performing feature selection on the extracted attributes to obtain a sample set; continuously calculating a winning center and a punishment center of each input sample in the sample set by utilizing an opponent punishment competitive learning RPCL algorithm until a cost function is met, and finishing the continuous calculation to obtain different component combination data; and verifying the different component combined data according to the actual drilling logging data, and obtaining the solid filling characteristics of the fracture-cavity body in the target area according to the verified different component combined data. The extracted attributes are subjected to feature selection to obtain a sample set, so that a new seismic data volume is automatically formed to avoid manual intervention; a cost function is introduced into an opponent punishment competition learning RPCL algorithm, convergence of the RPCL algorithm is accelerated, so that the existence of a dead zone is overcome, solid filling characteristics of the seam hole body can be effectively identified, and the identification precision of the solid filling of the seam hole body is improved.
In specific implementation, three-dimensional seismic data of a target area are acquired first. In a specific embodiment, as shown in fig. 2, after acquiring the three-dimensional seismic data of the target area based on fig. 1, the method further includes:
step 201: and carrying out noise removal preprocessing on the three-dimensional seismic data.
Pre-stack or post-stack three-dimensional seismic data is loaded into the system for pre-processing to reduce the effects of noise.
After the three-dimensional seismic data of the target area are obtained, attributes are extracted from seismic data volumes of different waveforms and different frequencies of the three-dimensional seismic data of the target area, and feature selection is carried out on the extracted attributes to obtain a sample set. The specific implementation process, as shown in fig. 3, includes:
step 301: carrying out seismic volume decomposition in seismic data volumes with different waveforms and different frequencies of the three-dimensional seismic data of the target area to obtain a plurality of components of seismic waves;
step 302: and (4) performing feature selection in a plurality of components of the seismic waves by using a dimensionality reduction algorithm to obtain a sample set.
In practice, the basic model of seismic traces in conventional processing and interpretation of seismic data is a convolution model, i.e., a seismic trace can be represented as a convolution of a seismic wavelet with a sequence of formation reflection coefficients:
S(t)＝W(t)×R(t)+N(t)
wherein S (t) represents a seismic trace; r (t) represents a reflection coefficient sequence function, W (t) represents seismic wavelets, and N (t) represents a noise term.
The seismic response of strata with different physical characteristics, such as reservoirs and non-reservoirs, oil-gas reservoirs and oil-gas-free reservoirs is different, the seismic wavelet is subjected to different transformations when passing through different strata, and the shape of the seismic wavelet is changed differently.
Seismic volume decomposition is a decomposition and synthesis method for extracting linearly similar seismic wave shapes, which can decompose an input seismic data segment into several components: the waveform of the first component represents a waveform having the greatest commonality, greatest energy, among all of the input seismic data segments; the second component is the waveform with the maximum commonality and the maximum energy in the seismic data segment after the first component is removed; the third component is the waveform … … having the greatest commonality, greatest energy, and so on in the seismic data segment after the first and second components have been removed. Generally, the higher-order components are less common and most are noise.
The same seismic component of similar amplitude reflects similar physical and seismic lithofacies characteristics of the corresponding geological interval, i.e., geological lithology or lithofacies characteristics of similar seismic response. Thus, the first component reflects the largest grade of geological lithology or lithofacies classification for the region. The second component reflects the next level of geological lithology or lithofacies classification … … for the region, and so on. Seismic volume decomposition can effectively identify geological lithology and lithofacies blocks of different seismic responses in different levels.
After the multiple components of the seismic waves are obtained through seismic volume decomposition, feature selection is carried out on the multiple components of the seismic waves through a dimensionality reduction algorithm, and a sample set containing the multi-attribute seismic wave samples is obtained. The multiple attributes may include, among others, seismic amplitude root mean square attributes, seismic amplitude gradient attributes, sweet-spot attributes, frequency attributes, and the like.
And continuously calculating the winning center and the penalty center of each input sample in the sample set by utilizing an opponent punishment competitive learning RPCL algorithm until the cost function is met, and finishing the continuous calculation to obtain different component combination data.
The main idea of the RPCL algorithm is to award the best unit weight vector to fit the input sample and make a smaller penalty on the weight vector of its competitor (second best). Based on a given learning and penalty rate, the RPCL algorithm can automatically assign the appropriate weight vectors while driving out additional units of weight vectors to the distance. However, the convergence of the classical RPCL algorithm is sensitive to the cost function and the update parameters. In addition, the cluster form is limited because the distance between the samples and the weight vector is mainly used in the form of euclidean distance. Therefore, in the embodiment of the invention, an M-RPCL algorithm is provided, a new cost function and a parameter learning method are introduced to accelerate the convergence of the RPCL algorithm on the basis of the original PPCL algorithm, and meanwhile, the Mahalanobis distance is applied to the cost function for calculation.
In specific implementation, the process of reconstructing and optimizing the component of the seismic data of the target layer by using the M-RPCL algorithm is as follows:
assuming the input multi-attribute seismic wave samples as a one-dimensional vector set:
wherein S represents a one-dimensional vector set of the multi-attribute seismic wave samples; xuRepresenting one-dimensional vector seismic wave samples; u represents the label of the seismic wave sample; n represents the number of multi-attribute seismic wave samples and d represents the number of selected attributes.
Cluster center Wi(i denotes the number of seismic facies) can also be represented by a set of one-dimensional vectors:
for a randomly given sample, the winning center vector c is given by the following formula:
c＝arg minγi||Xu-Wi||，i＝1,..,k
wherein k represents the number of seismic facies; center of weight WjRelative winning frequency gamma ofiIs determined by the following equation:
in the formula tjIs the cumulative number of times.
The second best (opponent's center) r is calculated by the following equation:
r＝arg minγi||Xu-Wj||,i＝1,2,...,k,i≠c
because of winning frequency gammaiThe chance of winning at the next calculation at the center of the last win is reduced.
A special cost function is introduced to dynamically control the number of cycles in the M-RPCL algorithm, and the cost function can be expressed as:
wherein C (u) represents a sample XuAn index of winning centers; wC(u)Representing a cluster center; η represents a regulatory factor and once the cost function is minimized, the M-RPCL algorithm will stop immediately. All weight centers can be updated and adjusted by the following formula:
wherein 0 < ac< 1 and 0 < ar< 1 are winning centers W, respectivelycAnd competition center WrLearning rate and penalty rate. It is generally assumed that these two variables have fixed values or a simple decreasing function ac(t) and ar(t), and a cycle time t. However, in the algorithm, the two areThe key value is a dynamic estimate, which can be expressed as the following function:
where λ (t) and ρ (t) are two decreasing functions of the cycle time t. The learning rate and penalty rate are not only related to the cycle time t, but also by the current weight vector
The influence of (c).
And reconstructing and optimizing components of the seismic data of the target layer by using the M-RPCL algorithm to obtain different component combined data. And verifying the different component combined data according to the actual drilling logging data, and obtaining the solid filling characteristics of the fracture-cavity body in the target area according to the verified different component combined data. As the actual drilling logging data can confirm the basic characteristics of the solid filling characteristics of the fracture-cavity body in the target area, the actual drilling logging data can be used for firstly checking the combined data of different components, and obviously wrong data are removed. The specific implementation process, as shown in fig. 4, includes:
step 401: based on the corresponding real drilling well logging data, removing error data deviating from the actual data in the different component combined data to obtain verified different component combined data;
step 402: selecting the waveform data with the highest mud filling common characteristic and energy matching degree reflected by the actual drilling well logging data from the verified different component combined data;
step 403: and determining the solid filling characteristics of the slot body in the target area according to the waveform data.
And matching the verified different component combined data with the shale filling common characteristic and energy obtained by the real drilling well logging data analysis, and selecting the waveform data of the component combined data with the highest matching degree, namely the seismic waveform which can reflect the solid filling characteristic of the fracture-cavity body of the target area most. According to the waveform data, the solid filling characteristics of the fracture-cavity body in the target area can be determined, and the method belongs to filling, non-filling or semi-filling, so that more accurate information about the fracture-cavity filling characteristics is extracted, the 'string-bead' reflection characteristics of the filled fracture-cavity body are effectively predicted and identified, the drilling risk is reduced, and the production is effectively built. And identifying multiple geological abnormal bodies such as filling and non-filling hole bodies, a river channel system and the like in a target area so as to guide the development of irregular oil and gas reservoir development well pattern deployment and drilling engineering abnormity early warning analysis.
A specific example is given below to illustrate how embodiments of the present invention identify solid-filled features of a cavity. The example is applied to the M region of the Tarim basin.
In the process of identifying and filling the crack-hole body on the three-dimensional seismic data, the pre-stack or post-stack seismic data are loaded into the system and preprocessed to reduce the influence of noise. And then extracting attributes from seismic data volumes with different frequencies and different waveforms according to actual needs, and obtaining a sample set and corresponding three-dimensional seismic data thereof through a dimension reduction algorithm. And then, continuously calculating winning and punishing centers of each input sample by using an M-RPCL algorithm, automatically finishing the algorithm when a cost function is met, and updating all weight centers by using current parameters when the algorithm is finished. In this method, each type of geological feature may be associated with a different cluster in the plurality of attribute seismic wave samples, and the filled fracture-cavity volume features may be automatically determined and a new seismic data volume formed. And reservoir inversion based on well-seismic analysis is carried out on the newly formed seismic data body, so that the identification precision of the filled cavern body is further improved.
In the M region, the earlier raw data seismic sections identified both WELL1 and WELL2 reservoir features as typical "beaded" reflections. The method comprises the steps of calculating the seismic data through an M-RPCL algorithm, automatically optimizing to form a corresponding seismic data body according to data frequency and waveform characteristics, sequentially comparing and analyzing different data differences, basically presenting the characteristic that 'bead' reflection capacity is consistent with waveform change trend through data codes 1-7, and basically presenting filled and unfilled 'bead' -shaped reflection characteristics through data codes 8-10. The typical 'bead' reflection of WELL drilling of WELL1, the actual drilling meets a large set of mudstone and fills a cavity (30.1m thick), the 'bead' reflection of WELL filling characteristics (according to WELL logging lithology information and WELL logging information) of WELL2 is not seen, and the combined data code 8 can effectively identify the filling characteristics of the WELL1 WELL (see figure 5).
In M areas, a reservoir prediction plan based on original seismic data reflects the comprehensive response of a target interval geologic body, and the characteristics of a river channel flowing from east to south to west to north are obvious, but the scope of river channel control is unclear. And the seismic data of the data code 8 is obtained after M-RPCL algorithm operation, and then reservoir prediction is carried out, so that the comprehensive response of the hole body with the target interval not filled with the seam can be reflected, and the prediction result has high goodness of fit with the drilling fluid loss, overflow position and loss scale generated in the drilling process. The reservoir prediction plan based on the residual between the original seismic data and the data code 8 seismic data reflects the comprehensive response of the target interval filling the cavern body, the river channel characteristics of the plan and the characteristics of the high-mud-content strips controlled along the river channel system are clear, and the gamma logging data are expressed as high values, and are shown in fig. 6.
In addition, as shown in fig. 7, performing well-seismic reservoir inversion analysis based on the data code 8 can obtain a result more consistent with drilling data. Compared with the inversion result based on the original seismic data, the geostatistical inversion result based on the well-to-seismic combination of the data code 8 seismic data has obvious difference of filling characteristic identification capability. The geostatistical inversion wave impedance profile of seismic data identified based on the M-RPCL algorithm is more consistent with drilling and logging information, the scale of an effective reservoir of a WELL1 WELL is limited, only 126 sides of drilling fluid leaked from the reservoir are filled under a cave body, no pressure is needed for closing the WELL after nearly thousand sides of fluid are produced, no benefit can be obtained, the WELL2 WELL develops the effective reservoir of a certain scale, 1775 sides of the drilling fluid is leaked during drilling, and the current accumulated fluid production is 7.8 times that of the WELL1 WELL.
As shown in fig. 8, which is a comparison of the raw seismic data and the final component-optimized data profile for this embodiment, the final component-optimized data profile shows significant suppression of WELL-filled characteristics of WELL1, while WELL-unfilled characteristics of WELL2 are effectively preserved.
The implementation of the above specific application is only an example, and the rest of the embodiments are not described in detail.
Based on the same inventive concept, embodiments of the present invention further provide a fracture-cavity solid filling feature recognition apparatus, and as the principle of the problem solved by the fracture-cavity solid filling feature recognition apparatus is similar to the fracture-cavity solid filling feature recognition method, the implementation of the fracture-cavity solid filling feature recognition apparatus may refer to the implementation of the fracture-cavity solid filling feature recognition method, and the repeated parts are not repeated, and the specific structure is as shown in fig. 9:
a data obtaining module 901, configured to obtain three-dimensional seismic data of a target area;
a sample optimization selection module 902, configured to extract attributes from seismic data volumes with different waveforms and different frequencies of three-dimensional seismic data in a target area, and perform feature selection on the extracted attributes to obtain a sample set;
a competitive learning operation module 903, configured to perform continuous calculation on a winning center and a penalty center of each input sample in the sample set by using an opponent punishment competitive learning RPCL algorithm, and end the continuous calculation until a cost function is satisfied, so as to obtain different component combination data;
and the filling characteristic determining module 904 is used for verifying the different component combined data according to the actual drilling logging data, and obtaining the solid filling characteristics of the fracture-cavity body in the target area according to the verified different component combined data.
In a specific embodiment, the sample optimization selecting module 902 is specifically configured to:
carrying out seismic volume decomposition in seismic data volumes with different waveforms and different frequencies of the three-dimensional seismic data of the target area to obtain a plurality of components of seismic waves;
and (4) performing feature selection in a plurality of components of the seismic waves by using a dimensionality reduction algorithm to obtain a sample set.
In a specific embodiment, the cost function is:
wherein E (W) represents a cost function; xuRepresenting one-dimensional vector seismic wave samples; c (u) represents a sample XuAn index of winning centers; wC(u)Representing a cluster center; c represents the amount of winning centers; eta represents a regulatory factor; u represents the index of the seismic wave sample.
In a specific embodiment, the filling characteristic determining module 904 is specifically configured to:
based on the corresponding real drilling well logging data, removing error data deviating from the actual data in the different component combined data to obtain verified different component combined data;
selecting the waveform data with the highest mud filling common characteristic and energy matching degree reflected by the actual drilling well logging data from the verified different component combined data;
and determining the solid filling characteristics of the slot body in the target area according to the waveform data.
In a specific embodiment, a device for identifying solid filling characteristics of a slot body is further provided, as shown in fig. 10, on the basis of fig. 9, the device further includes:
the preprocessing module 1001 is configured to perform preprocessing for removing noise on the three-dimensional seismic data after acquiring the three-dimensional seismic data of the target area.
The embodiment of the invention also provides computer equipment which comprises a memory, a processor and a computer program which is stored on the memory and can run on the processor, wherein the processor executes the computer program to realize the fracture-cavity solid filling characteristic identification method.
The embodiment of the invention also provides a computer readable storage medium, which stores a computer program for executing the identification method of the solid filling characteristics of the slot body.
In summary, the method and the device for identifying the solid filling characteristics of the slot body provided by the embodiment of the invention have the following advantages:
acquiring three-dimensional seismic data of a target area; extracting attributes from seismic data volumes with different waveforms and different frequencies in the three-dimensional seismic data of the target area, and performing feature selection on the extracted attributes to obtain a sample set; continuously calculating a winning center and a punishment center of each input sample in the sample set by utilizing an opponent punishment competitive learning RPCL algorithm until a cost function is met, and finishing the continuous calculation to obtain different component combination data; and verifying the different component combined data according to the actual drilling logging data, and obtaining the solid filling characteristics of the fracture-cavity body in the target area according to the verified different component combined data. The extracted attributes are subjected to feature selection to obtain a sample set, so that a new seismic data volume is automatically formed to avoid manual intervention; a cost function is introduced into an opponent punishment competition learning RPCL algorithm, convergence of the RPCL algorithm is accelerated, so that the existence of a dead zone is overcome, solid filling characteristics of the seam hole body can be effectively identified, and the identification precision of the solid filling of the seam hole body is improved.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, apparatus, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The above description is only a preferred embodiment of the present invention, and is not intended to limit the present invention, and various modifications and changes may be made to the embodiment of the present invention by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (12)
1. A method for identifying solid filling characteristics of a slot body is characterized by comprising the following steps:
acquiring three-dimensional seismic data of a target area;
extracting attributes from seismic data volumes with different waveforms and different frequencies of the three-dimensional seismic data of the target area, and performing feature selection on the extracted attributes to obtain a sample set;
continuously calculating a winning center and a punishment center of each input sample in the sample set by utilizing an opponent punishment competitive learning RPCL algorithm until a cost function is met, and finishing the continuous calculation to obtain different component combination data;
and verifying the different component combined data according to the actual drilling logging data, and obtaining the solid filling characteristics of the fracture-cavity body in the target area according to the verified different component combined data.
2. The method of claim 1, wherein extracting attributes from seismic data volumes of different waveforms, different frequencies, of the three-dimensional seismic data of the target area, and performing feature selection on the extracted attributes to obtain a sample set comprises:
carrying out seismic volume decomposition in seismic data volumes with different waveforms and different frequencies of the three-dimensional seismic data of the target area to obtain a plurality of components of seismic waves;
and (4) performing feature selection in a plurality of components of the seismic waves by using a dimensionality reduction algorithm to obtain a sample set.
3. The method of claim 1, wherein the cost function is:
wherein E (W) represents a cost function; xuRepresenting one-dimensional vector seismic wave samples; c (u) represents a sample XuAn index of winning centers; wC(u)Representing a cluster center; c represents the amount of winning centers; eta represents a regulatory factor; u represents the index of the seismic wave sample.
4. The method of claim 1, wherein the verifying the different component combined data according to the actual drilling well logging data, and the obtaining the solid filling characteristics of the fracture-cavity body of the target area according to the verified different component combined data comprises:
based on the corresponding real drilling well logging data, removing error data deviating from the actual data in the different component combined data to obtain verified different component combined data;
selecting the waveform data with the highest mud filling common characteristic and energy matching degree reflected by the actual drilling well logging data from the verified different component combined data;
and determining the solid filling characteristics of the slot body in the target area according to the waveform data.
5. The method of claim 1, further comprising:
and after the three-dimensional seismic data of the target area are obtained, preprocessing for removing noise is carried out on the three-dimensional seismic data.
6. A slot solid filling characteristic recognition device is characterized by comprising:
the data acquisition module is used for acquiring three-dimensional seismic data of a target area;
the sample optimization selection module is used for extracting attributes from seismic data bodies with different waveforms and different frequencies of the three-dimensional seismic data of the target area, and performing feature selection on the extracted attributes to obtain a sample set;
the competitive learning operation module is used for continuously calculating a winning center and a punishment center of each input sample in the sample set by utilizing an opponent punishment competitive learning RPCL algorithm until a cost function is met, and finishing continuous calculation to obtain different component combination data;
and the filling characteristic determining module is used for verifying the different component combined data according to the actual drilling logging data and obtaining the solid filling characteristics of the fracture-cavity body in the target area according to the verified different component combined data.
7. The apparatus of claim 6, wherein the sample-optimized selection module is specifically configured to:
carrying out seismic volume decomposition in seismic data volumes with different waveforms and different frequencies of the three-dimensional seismic data of the target area to obtain a plurality of components of seismic waves;
and (4) performing feature selection in a plurality of components of the seismic waves by using a dimensionality reduction algorithm to obtain a sample set.
8. The apparatus of claim 6, wherein the cost function is:
wherein E (W) represents a cost function; xuRepresenting one-dimensional vector seismic wave samples; c (u) represents a sample XuAn index of winning centers; wC(u)Representing a cluster center; c represents the amount of winning centers; eta represents a regulatory factor; u represents the index of the seismic wave sample.
9. The apparatus of claim 6, wherein the fill characteristic determination module is specifically configured to:
based on the corresponding real drilling well logging data, removing error data deviating from the actual data in the different component combined data to obtain verified different component combined data;
selecting the waveform data with the highest mud filling common characteristic and energy matching degree reflected by the actual drilling well logging data from the verified different component combined data;
and determining the solid filling characteristics of the slot body in the target area according to the waveform data.
10. The apparatus of claim 6, further comprising:
and the preprocessing module is used for preprocessing the three-dimensional seismic data to remove noise after acquiring the three-dimensional seismic data of the target area.
11. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the method of any one of claims 1 to 5 when executing the computer program.
12. A computer-readable storage medium, characterized in that the computer-readable storage medium stores a computer program for executing the method of any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011203151.9A CN112379440A (en) | 2020-11-02 | 2020-11-02 | Method and device for identifying solid filling characteristics of fracture and hole body |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011203151.9A CN112379440A (en) | 2020-11-02 | 2020-11-02 | Method and device for identifying solid filling characteristics of fracture and hole body |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112379440A true CN112379440A (en) | 2021-02-19 |
Family
ID=74576913
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011203151.9A Pending CN112379440A (en) | 2020-11-02 | 2020-11-02 | Method and device for identifying solid filling characteristics of fracture and hole body |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112379440A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115822563A (en) * | 2022-12-30 | 2023-03-21 | 中国石油化工股份有限公司 | Design method of ultra-deep layer fracture control fracture-cave type oil and gas reservoir high-yield well |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040156548A1 (en) * | 2002-11-20 | 2004-08-12 | Taku Kodama | Image processing apparatus, image compressing apparatus, image decompressing apparatus, image processing method, image compressing method, image decompressing method, information processing apparatus, information processing method, program and recording medium |
| JP2007028439A (en) * | 2005-07-20 | 2007-02-01 | Ricoh Co Ltd | Image processing apparatus and image processing method |
| CN108344360A (en) * | 2017-11-15 | 2018-07-31 | 北京航空航天大学 | A kind of the laser scan type overall situation calibrating installation and method of vision measurement system |
| CN110570289A (en) * | 2019-09-13 | 2019-12-13 | 腾讯科技（深圳）有限公司 | service processing method, device, equipment and storage medium based on block chain |
-
2020
- 2020-11-02 CN CN202011203151.9A patent/CN112379440A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040156548A1 (en) * | 2002-11-20 | 2004-08-12 | Taku Kodama | Image processing apparatus, image compressing apparatus, image decompressing apparatus, image processing method, image compressing method, image decompressing method, information processing apparatus, information processing method, program and recording medium |
| JP2007028439A (en) * | 2005-07-20 | 2007-02-01 | Ricoh Co Ltd | Image processing apparatus and image processing method |
| CN108344360A (en) * | 2017-11-15 | 2018-07-31 | 北京航空航天大学 | A kind of the laser scan type overall situation calibrating installation and method of vision measurement system |
| CN110570289A (en) * | 2019-09-13 | 2019-12-13 | 腾讯科技（深圳）有限公司 | service processing method, device, equipment and storage medium based on block chain |
Non-Patent Citations (1)
| Title |
|---|
| XIANGWEN LI ET AL.: "Automatic filled fractured-vuggy detection using the competitive learning algorithm", SEG INTERNATIONAL EXPOSITION AND 89TH ANNUAL MEETING, pages 3369 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115822563A (en) * | 2022-12-30 | 2023-03-21 | 中国石油化工股份有限公司 | Design method of ultra-deep layer fracture control fracture-cave type oil and gas reservoir high-yield well |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10942293B2 (en) | Rock physics based method of integrated subsurface reservoir characterization for use in optimized stimulation design of horizontal wells | |
| AU603521B2 (en) | Method for determining the depth of a hydraulic fracture zone in the earth | |
| US8639443B2 (en) | Microseismic event monitoring technical field | |
| US7873476B2 (en) | Well log correlation | |
| US11740372B1 (en) | Method and system for intelligently identifying carbon storage box based on GAN network | |
| EP3362640B1 (en) | History matching of hydrocarbon production from heterogenous reservoirs | |
| US8255165B2 (en) | Method for predicting differences in subsurface conditions | |
| CN110658556A (en) | Seismic technology combination method for identifying and evaluating broken zone of sliding fracture of carbonate rock | |
| NO20190214A1 (en) | Classifying well data using a support vector machine | |
| CN108254783A (en) | A kind of poststack earthquake fluid recognition methods based on time frequency analysis | |
| US20190331812A1 (en) | Methods and Systems for Reference-Based Inversion of Seismic Image Volumes | |
| CN112379440A (en) | Method and device for identifying solid filling characteristics of fracture and hole body | |
| CN103278849A (en) | Method and system for performing wavelet estimation on the basis of seismic data and logging information | |
| CN113534247B (en) | Crack quantitative prediction method and device based on post-stack seismic data | |
| CN110069797B (en) | Method and system for judging connectivity between fracture-cavity type oil reservoir wells | |
| Hoyle | COMPUTER TECHNIQUES FOR THE ZONING AND CORRELATION OF WELL‐LOGS | |
| CN112147677A (en) | Oil and gas reservoir parameter tag data generation method and device | |
| CN112379435A (en) | Phase-controlled karst type seam hole aggregate carving method and device | |
| CN109581489B (en) | Nested seismic facies extraction method and system | |
| CN112346121A (en) | Reservoir stratum separation processing method based on full waveform | |
| Li et al. | Automatic filled fractured-vuggy detection using the competitive learning algorithm | |
| Topór et al. | An advanced ensemble modeling approach for predicting carbonate reservoir porosity from seismic attributes | |
| US11719844B2 (en) | System and method for analyzing reservoir changes during production | |
| CN112241578B (en) | Salt structure style analysis method, device and system | |
| Man et al. | Hydraulic Flow Unit Classification and Prediction Using Machine Learning Techniques: A Case Study from the Nam Con Son Basin, Offshore Vietnam. Energies 2021, 14, 7714 |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |