Reservoir oil and gas prediction method and device based on multi-seismic attribute optimization

Abstract

The present invention discloses a reservoir oil and gas prediction method and device based on multi-seismic attribute optimization, the method comprising: obtaining first seismic attribute data of a study area, wherein the first seismic attribute data comprises: parameter data of multiple seismic attributes; based on the first seismic attribute data, screening second seismic attribute data that meets preset seismic attribute conditions; inputting the second seismic attribute data into a pre-trained classifier, and outputting third seismic attribute data after dimension reduction, wherein the classifier is a seismic attribute classification model obtained by machine learning training based on the relationship between known drilling data and seismic attributes; based on the third seismic attribute data, predicting reservoir oil and gas in the study area. The present invention utilizes the vertical high resolution of logging data and the horizontal high resolution of seismic attribute data, combined with the advantages of multi-seismic attribute reservoir prediction, can improve the accuracy of reservoir oil and gas prediction, and can be applicable to oil and gas prediction in tight sandstone reservoirs.

Description

Technical Field

[0001] The present invention relates to the field of reservoir oil and gas identification, and in particular to a reservoir oil and gas prediction method and device based on multi-seismic attribute optimization. Background Art

[0002] This section is intended to provide a background or context to the embodiments of the invention recited in the claims. No admission is made that the description herein is prior art by inclusion in this section.

[0003] With the advancement of seismic exploration technology, seismic attribute analysis technology is widely used in oil and gas development. Seismic attribute analysis can obtain hidden information about lithology and reservoir properties from seismic data, so that seismic interpreters can use this information to interpret underground structures, strata, lithology, oil and gas and other geological characteristics.

[0004] There is a complex relationship between seismic attributes and underground rock properties. Different regions, different lithologies and different reservoir configurations will affect the relationship between seismic attributes. Since a single seismic attribute has multiple solutions and cannot well represent geological characteristics, multiple seismic attributes are processed by dimensionality reduction. Using the seismic attributes after dimensionality reduction, more abundant reservoir geological information can be mined, thereby obtaining more accurate reservoir oil and gas prediction results.

[0005] At present, the prestack inversion technology suitable for predicting reservoirs with thin low-velocity drop layers and high seismic resolution is not ideal for predicting reservoirs with thick low-velocity drop layers and low seismic resolution due to the influence of seismic resolution, etc. (for example, the oil and gas prediction compliance rate for the Upper Paleozoic tight sandstone reservoirs in the southern part of Sulige in the Ordos Basin is only about 60%), which makes it difficult to meet the increasingly sophisticated needs of oil and gas exploration and development. Summary of the invention

[0006] In an embodiment of the present invention, a reservoir oil and gas prediction method based on multi-seismic attribute optimization is provided to solve the technical problem that the existing reservoir oil and gas prediction method based on pre-stack inversion technology is affected by the reservoir seismic resolution and is not suitable for tight sandstone reservoir oil and gas prediction. The method comprises: obtaining first seismic attribute data of a study area, wherein the first seismic attribute data comprises: parameter data of a plurality of seismic attributes; based on the first seismic attribute data, screening second seismic attribute data that meets preset seismic attribute conditions; inputting the second seismic attribute data into a pre-trained classifier, and outputting third seismic attribute data after dimensionality reduction, wherein the classifier is a seismic attribute classification model obtained by machine learning training based on the relationship between known drilling data and seismic attributes; and performing reservoir oil and gas prediction for the study area based on the third seismic attribute data.

[0007] The embodiment of the present invention also provides a reservoir oil and gas prediction device based on multi-seismic attribute optimization, which is used to solve the technical problem that the existing reservoir oil and gas prediction method based on pre-stack inversion technology is affected by the reservoir seismic resolution and is not suitable for tight sandstone reservoir oil and gas prediction. The device includes: a seismic attribute data acquisition module, which is used to acquire the first seismic attribute data of the study area, wherein the first seismic attribute data includes: parameter data of multiple seismic attributes; a seismic attribute data screening module, which is used to screen the second seismic attribute data that meets the preset seismic attribute conditions according to the first seismic attribute data; a seismic attribute data dimensionality reduction processing module, which is used to input the second seismic attribute data into a pre-trained classifier and output the third seismic attribute data after dimensionality reduction, wherein the classifier is a seismic attribute classification model obtained by machine learning training based on the relationship between known drilling data and seismic attributes; a reservoir oil and gas prediction module, which is used to perform reservoir oil and gas prediction for the study area according to the third seismic attribute data.

[0008] A computer device is also provided in an embodiment of the present invention to solve the technical problem that the existing reservoir oil and gas prediction method based on pre-stack inversion technology is affected by the reservoir seismic resolution and is not suitable for tight sandstone reservoir oil and gas prediction. The computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the reservoir oil and gas prediction method based on multi-seismic attribute optimization as claimed in the claim is implemented.

[0009] A computer-readable storage medium is also provided in an embodiment of the present invention to solve the technical problem that the existing reservoir oil and gas prediction method based on pre-stack inversion technology is affected by the reservoir seismic resolution and is not suitable for tight sandstone reservoir oil and gas prediction. The computer-readable storage medium stores a computer program for executing the above-mentioned reservoir oil and gas prediction method based on multi-seismic attribute optimization.

[0010] In an embodiment of the present invention, after obtaining seismic attribute data containing multiple seismic attributes in the study area, the seismic attribute data that meet the conditions are screened out according to preset seismic attribute conditions, and then the screened out seismic attribute data are input into a pre-trained classifier, and the seismic attribute data after dimensionality reduction are output. Finally, based on the seismic attribute data after dimensionality reduction, reservoir oil and gas prediction is performed for the study area.

[0011] Through the embodiments of the present invention, dimensionality reduction processing of seismic attributes is achieved based on drilling control, and then the seismic attribute data after dimensionality reduction are used to predict reservoir oil and gas. The vertical high resolution of logging data and the lateral high resolution of seismic attribute data can be utilized, combined with the advantages of multi-seismic attribute reservoir prediction, so as to improve the accuracy of reservoir oil and gas prediction, and can be applicable to oil and gas prediction in tight sandstone reservoirs. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0013] Figure 1 A flow chart of a reservoir oil and gas prediction method based on multi-seismic attribute optimization provided in an embodiment of the present invention;

[0014] Figure 2 It is a plane diagram of the minimum Poisson's ratio of the box 8 provided in an embodiment of the present invention;

[0015] Figure 3 The cross-plot of the minimum Poisson's ratio of the box 8 section and the effective reservoir thickness provided in the embodiment of the present invention;

[0016] Figure 4 It is a plane distribution diagram of earthquake attributes after dimensionality reduction of the box 8 segment provided in an embodiment of the present invention;

[0017] Figure 5 The cross-plot of the seismic attributes and the effective reservoir thickness after the dimensionality reduction of the box 8 segment provided in the embodiment of the present invention;

[0018] Figure 6 Schematic diagram of a reservoir oil and gas prediction device based on multi-seismic attribute optimization provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0019] To make the purpose, technical solution and advantages of the embodiments of the present invention more clear, the embodiments of the present invention are further described in detail below in conjunction with the accompanying drawings. Here, the exemplary embodiments of the present invention and their descriptions are used to explain the present invention, but are not intended to limit the present invention.

[0020] In the description of this specification, the terms "include", "including", "have", "contain", etc. are all open terms, which mean including but not limited to. The descriptions with reference to the terms "one embodiment", "a specific embodiment", "some embodiments", "for example", etc. mean that the specific features, structures or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures or characteristics described may be combined in a suitable manner in any one or more embodiments or examples. The order of steps involved in each embodiment is used to schematically illustrate the implementation of the present application, and the order of steps is not limited and can be appropriately adjusted as needed.

[0021] The present invention provides a reservoir oil and gas prediction method based on multi-seismic attribute optimization. Figure 1 Flow chart of a reservoir oil and gas prediction method based on multi-seismic attribute optimization provided in an embodiment of the present invention, such as Figure 1 As shown, the method comprises the following steps:

[0022] S101, obtaining first seismic attribute data of a study area, wherein the first seismic attribute data includes: parameter data of a plurality of seismic attributes.

[0023] It should be noted that seismic attributes are specific measurement contents that generate geometric, kinematic, and dynamic characteristics from seismic data bodies, and are inextricably linked to underground rock properties. Different regions, different lithologies, and different reservoir configurations will affect the relationship between seismic attributes. However, since a single seismic attribute has strong multi-solution properties and cannot characterize geological characteristics well, the embodiment of the present invention obtains seismic attribute data containing multiple seismic attributes in the study area to predict reservoir oil and gas in the study area. Optionally, the first seismic attribute data obtained in S101 above may be parameter data of multiple seismic attributes extracted from seismic data. It should be noted that parameter data of hundreds of seismic attributes can be extracted from seismic data.

[0024] In the embodiments of the present invention, the reservoir in the study area may be, but is not limited to, a tight sandstone reservoir. The following embodiments are described using the southern Sulige reservoir in the Ordos Basin as an example.

[0025] The surface in the southern part of Sulige, Ordos Basin is complex, with thick low-velocity drop layers and low frequency of seismic data; the Upper Paleozoic reservoir changes rapidly laterally, with thin single sand bodies, low porosity, low permeability and low abundance, and is a typical "three-low gas reservoir". Due to the small impedance difference between the reservoir and the surrounding rock in this area, it is difficult to identify the reservoir and the effective reservoir by longitudinal wave impedance; and the pre-stack inversion technology, which is suitable for areas with thin low-velocity drop layers and high seismic resolution, is not ideal for oil and gas prediction of reservoirs in this area due to the influence of seismic resolution, with the prediction compliance rate reaching only 60%, which is difficult to meet the increasingly sophisticated exploration and development needs.

[0026] S102: Screening second seismic attribute data that meets a preset seismic attribute condition based on the first seismic attribute data.

[0027] It should be noted that, since there are usually hundreds of seismic attributes extracted from seismic data, how to select appropriate dominant seismic attributes is the primary issue. The selection of dominant seismic attributes in the embodiment of the present invention is mainly based on the following three principles: first, whether the correlation between the seismic attributes and the actual drilling reservoir data is greater than a certain threshold (for example, 50%); second, whether the planar distribution of the seismic attributes conforms to the geological laws (the reservoir distribution characteristics can be preliminarily clarified through the analysis of the previous drilling data and the application of inversion technology); third, whether the correlation between the seismic attributes is within a preset range (for example, greater than 60% and less than 80%).

[0028] It should be noted that, based on the first principle and the second principle, 12 attributes can usually be selected, including root mean square amplitude, instantaneous frequency, amplitude slope, amplitude arithmetic mean, quality factor, target layer peak density, AVO change rate, smooth reflection intensity, minimum velocity ratio, texture attribute, highlight, arc length attribute, etc., and the correlation is calculated by intersecting them two by two. If the correlation between two seismic attributes is less than 60%, it means that one of the seismic attributes cannot reflect the geological law; if the correlation between two seismic attributes is greater than 80%, it means that the two seismic attributes are too similar and one of them can be selected.

[0029] As an optional implementation, the above S102 can be implemented through the following steps: based on the correlation between single seismic attribute data and drilling data, from the first seismic attribute data, select seismic attribute data with a correlation higher than a preset threshold; from the seismic attribute data with a correlation higher than the preset threshold, select seismic attribute data that conforms to the reservoir geological characteristic information of the study area; based on the correlation between seismic attributes, select second seismic attribute data from the seismic attribute data that conforms to the reservoir geological characteristic information of the study area.

[0030] Optionally, before screening seismic attribute data with a correlation higher than a preset threshold from the first seismic attribute data based on the correlation between the single seismic attribute data and the drilling data, the reservoir oil and gas prediction method based on multi-seismic attribute optimization provided in the embodiment of the present invention may also include the following steps: acquiring the drilling data of the study area; intersecting the parameter data of the single seismic attribute with the drilling data of the study area, and determining the correlation between the single seismic attribute data and the drilling data.

[0031] Optionally, before screening the seismic attribute data that conforms to the reservoir geological characteristic information of the study area from the seismic attribute data with a correlation higher than a preset threshold, the reservoir oil and gas prediction method based on multi-seismic attribute optimization provided in the embodiment of the present invention may also include the following steps: obtaining the reservoir geological characteristic information of the study area.

[0032] Optionally, before screening the second seismic attribute data from the seismic attribute data that conform to the reservoir geological characteristic information of the study area according to the correlation between the seismic attributes, the reservoir oil and gas prediction method based on multi-seismic attribute optimization provided in the embodiment of the present invention may also include the following steps: intersecting the parameter data of different seismic attributes to determine the correlation between the different seismic attribute data.

[0033] S103, inputting the second seismic attribute data into a pre-trained classifier, and outputting the third seismic attribute data after dimension reduction, wherein the classifier is a seismic attribute classification model obtained through machine learning training based on the relationship between known drilling data and seismic attributes.

[0034] It should be noted that a single seismic attribute represents the geological characteristics from a certain aspect and contains a lot of information about non-geological characteristics. Therefore, it is necessary to reduce the dimensionality of multiple seismic attributes in order to achieve the purpose of comprehensively and truly representing the geological characteristics while eliminating the false and retaining the true.

[0035] The embodiment of the present invention performs dimensionality reduction processing on the selected dominant seismic attributes under the control of drilling (referred to as "well-controlled dominant attribute dimensionality reduction technology"), uses actual drilling data as known sample data, and designs a classifier based on the known sample data. For example, sandstone thickness, porosity, etc. are used as sample data, and a classifier is designed using the change rate of sample data. The categories are divided according to the change rate, and the relationship between drilling data and seismic attributes is established based on a neural network algorithm to classify unknown samples, so that samples of the same category have certain similarities and samples of different categories show certain differences, thereby achieving the purpose of dimensionality reduction processing of seismic attributes.

[0036] It can be seen that the well control advantage attribute dimensionality reduction technology provided in the embodiment of the present invention makes good use of existing drilling data, the high vertical resolution of logging data and the high horizontal resolution of seismic attribute data, combined with the advantages of seismic multi-attribute reservoir prediction, to better realize oil and gas prediction of reservoirs, especially suitable for tight sandstone reservoirs.

[0037] Furthermore, the relationship between drilling data and seismic attributes is established based on the neural network algorithm, and the classifier is designed using the inflection point method, which can achieve good application results.

[0038] Optionally, the above classifier may be a seismic attribute classification model obtained through machine learning training based on a neural network algorithm.

[0039] S104: Predict reservoir oil and gas in the study area based on the third seismic attribute data.

[0040] It should be noted that since the seismic attribute data after dimensionality reduction processing (i.e., the third seismic attribute data) can retain the information of multiple seismic attributes before dimensionality reduction, using the seismic attribute data after dimensionality reduction processing to predict reservoir oil and gas in the study area can mine rich geological information and accurately identify and predict the distribution of reservoir oil and gas.

[0041] From the above, it can be seen that the reservoir oil and gas prediction method based on multi-seismic attribute optimization provided in the embodiment of the present invention, after obtaining the seismic attribute data containing multiple seismic attributes in the study area, screens out the seismic attribute data that meets the conditions according to preset seismic attribute conditions, and then inputs the screened seismic attribute data into a pre-trained classifier, outputs the seismic attribute data after dimensionality reduction, and finally predicts the reservoir oil and gas in the study area based on the seismic attribute data after dimensionality reduction.

[0042] Through the reservoir oil and gas prediction method based on multi-seismic attribute optimization provided in the embodiment of the present invention, seismic attribute dimensionality reduction processing is implemented based on drilling control, and then the seismic attribute data after dimensionality reduction is used to predict reservoir oil and gas. The vertical high resolution of logging data and the lateral high resolution of seismic attribute data can be utilized, combined with the advantages of multi-seismic attribute reservoir prediction, so as to improve the accuracy of reservoir oil and gas prediction, and can be applicable to oil and gas prediction in tight sandstone reservoirs.

[0043] Taking the prediction of tight sandstone gas in the Upper Paleozoic in the southern part of Sulige in the Ordos Basin as an example, the southern block of Sulige is located in the western part of the Yishan slope. The main reservoir of the Upper Paleozoic is the Permian Box 8-Shan 1. The gas reservoir enrichment law is controlled by the late Paleozoic sedimentary environment and the later tectonic evolution. The geological conditions for reservoir formation are favorable and have a certain scale of exploration potential. The Box 8-Shan 1 section is a delta plain-front deposit. The single sand body is small in scale and mostly lens-shaped; the composite sand body is large in scale and vertically stacked in multiple layers. The effective reservoir is the coarse lithofacies in the river channel sandstone, which is mainly distributed in the lower part of the river channel filling and the intersection of the river channel. The effective reservoir has large lateral changes and is isolated or in a small range of continuous sheets. The thickness of the sand layer of a single river channel sand body is 3 to 5 meters, and the thickness of the composite river channel sand layer is 10 to 20 meters. In general, the Upper Paleozoic reservoir characteristics in the southern part of Sulige are small reservoir thickness, thin effective reservoir thickness, strong heterogeneity, and it is quite difficult to predict the reservoir and gas content.

[0044] The reservoir oil and gas prediction method based on multi-seismic attribute optimization provided in the embodiment of the present invention predicts the distribution of high-quality sandstone reservoirs based on the comprehensive changes in seismic multi-attributes, forms a supporting seismic prediction method, and provides reliable guarantee for well location optimization.

[0045] Figure 2 and Figure 3 is the effective reservoir prediction result based on prestack inversion, among which, Figure 2 It is the minimum Poisson's ratio plane diagram of box 8; Figure 3This is the cross plot of the minimum Poisson's ratio and the effective reservoir thickness in Box 8. It can be seen from the cross plot that the minimum Poisson's ratio is generally negatively correlated with the effective reservoir thickness in Box 8, but the correlation is only 63.57%.

[0046] Figure 4 This is the plane distribution diagram of the He 8 segment after dimensionality reduction processing of seismic attributes, which is used to characterize the effective reservoir thickness of the He 8 segment;

[0047] Figure 5 This is the cross plot of the seismic attributes and the effective reservoir thickness after dimensionality reduction in Box 8. It can be seen from the cross plot that the seismic attributes are positively correlated with the effective reservoir thickness of Box 8, with a correlation of 86.7%.

[0048] By comparing Figure 2 and Figure 4 It can be seen that the minimum Poisson's ratio is roughly similar to the overall distribution of seismic attributes, indicating that the effective reservoir thickness of the He 8 segment predicted by prestack inversion and seismic attributes is highly reliable and can guide actual production. Figure 3 and Figure 5 It can be seen that the seismic attributes after dimensionality reduction processing have a higher correlation with the effective reservoir thickness, which can better predict the effective reservoir, guide actual production, and improve the drilling compliance rate.

[0049] Based on the same inventive concept, an embodiment of the present invention also provides a reservoir oil and gas prediction device based on multi-seismic attribute optimization, as described in the following embodiment. Since the principle of solving the problem in the embodiment of the device is similar to that of the reservoir oil and gas prediction method based on multi-seismic attribute optimization, the implementation of the embodiment of the device can refer to the implementation of the method, and the repeated parts will not be repeated.

[0050] Figure 6 Schematic diagram of a reservoir oil and gas prediction device based on multi-seismic attribute optimization provided in an embodiment of the present invention, such as Figure 6 As shown, the device may include: a seismic attribute data acquisition module 61 , a seismic attribute data screening module 62 , a seismic attribute data dimensionality reduction processing module 63 and a reservoir oil and gas prediction module 64 .

[0051] Among them, the seismic attribute data acquisition module 61 is used to obtain the first seismic attribute data of the study area, wherein the first seismic attribute data includes: parameter data of multiple seismic attributes; the seismic attribute data screening module 62 is used to screen the second seismic attribute data that meets the preset seismic attribute conditions according to the first seismic attribute data; the seismic attribute data dimensionality reduction processing module 63 is used to input the second seismic attribute data into a pre-trained classifier and output the third seismic attribute data after dimensionality reduction, wherein the classifier is a seismic attribute classification model obtained by machine learning training based on the relationship between known drilling data and seismic attributes; the reservoir oil and gas prediction module 64 is used to predict the reservoir oil and gas in the study area according to the third seismic attribute data.

[0052] As can be seen from the above, the reservoir oil and gas prediction device based on multi-seismic attribute optimization provided by the embodiment of the present invention obtains seismic attribute data containing multiple seismic attributes in the study area through the seismic attribute data acquisition module 61; screens out seismic attribute data that meet the conditions according to preset seismic attribute conditions through the seismic attribute data screening module 62; inputs the screened seismic attribute data into a pre-trained classifier through the seismic attribute data dimensionality reduction processing module 63, and outputs the seismic attribute data after dimensionality reduction; and performs reservoir oil and gas prediction on the study area according to the seismic attribute data after dimensionality reduction through the reservoir oil and gas prediction module 64.

[0053] Through the reservoir oil and gas prediction device based on multi-seismic attribute optimization provided in the embodiment of the present invention, seismic attribute dimensionality reduction processing is implemented based on drilling control, and then the seismic attribute data after dimensionality reduction is used to predict reservoir oil and gas. The vertical high resolution of logging data and the lateral high resolution of seismic attribute data can be utilized, combined with the advantages of multi-seismic attribute reservoir prediction, so as to improve the accuracy of reservoir oil and gas prediction, and can be applicable to oil and gas prediction in tight sandstone reservoirs.

[0054] Optionally, the reservoir oil and gas prediction device based on multi-seismic attribute optimization provided by the embodiment of the present invention can be used for, but not limited to, oil and gas prediction of tight sandstone reservoirs.

[0055] In one embodiment, the above-mentioned seismic attribute data screening module 62 can also be used to perform the following steps: based on the correlation between single seismic attribute data and drilling data, screening seismic attribute data with a correlation higher than a preset threshold from the first seismic attribute data; screening seismic attribute data that conforms to the reservoir geological characteristic information of the study area from the seismic attribute data with a correlation higher than the preset threshold; based on the correlation between seismic attributes, screening second seismic attribute data from the seismic attribute data that conforms to the reservoir geological characteristic information of the study area.

[0056] Furthermore, the seismic attribute data screening module 62 can also be used to perform the following steps: obtaining drilling data in the study area; intersecting the parameter data of a single seismic attribute with the drilling data in the study area, and determining the correlation between the single seismic attribute data and the drilling data.

[0057] Furthermore, the seismic attribute data screening module 62 can also be used to perform the following steps: obtaining reservoir geological characteristic information of the study area.

[0058] Furthermore, the seismic attribute data screening module 62 can also be used to perform the following steps: intersecting parameter data of different seismic attributes to determine the correlation between different seismic attribute data.

[0059] Optionally, the classifier used by the seismic attribute data dimensionality reduction processing module 63 may be a seismic attribute classification model obtained through machine learning training based on a neural network algorithm.

[0060] Based on the same inventive concept, a computer device is also provided in an embodiment of the present invention to solve the technical problem that the existing reservoir oil and gas prediction method based on pre-stack inversion technology is affected by the reservoir seismic resolution and is not suitable for tight sandstone reservoir oil and gas prediction. The computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the reservoir oil and gas prediction method based on multi-seismic attribute optimization of any one of the above claims.

[0061] A computer-readable storage medium is also provided in an embodiment of the present invention to solve the technical problem that the existing reservoir oil and gas prediction method based on pre-stack inversion technology is affected by the reservoir seismic resolution and is not suitable for tight sandstone reservoir oil and gas prediction. The computer-readable storage medium stores a computer program for executing any of the above-mentioned reservoir oil and gas prediction methods based on multi-seismic attribute optimization.

[0062] In summary, the embodiments of the present invention provide a reservoir oil and gas prediction method, apparatus, computer equipment and computer-readable storage medium based on multi-seismic attribute optimization, which utilizes existing drilling data to combine the high vertical resolution of logging data with the high horizontal resolution of seismic attribute data, and at the same time takes advantage of the comprehensive prediction of reservoir oil and gas distribution based on the changes in seismic multiple attributes, so as to better predict reservoir oil and gas in tight sandstone, form a supporting seismic prediction method, and provide reliable guarantee for well site optimization.

[0063] The reservoir oil and gas prediction method based on multi-seismic attribute optimization provided in the embodiment of the present invention has been applied in the exploration of natural gas in the tight sandstone thin reservoir of the Upper Paleozoic in the southern area of ​​Sulige in the Ordos Basin. From January 2018 to December 2018, a total of 40 development wells were selected and 30 were drilled, including 26 Class I+II wells, accounting for 86.7%. Among them, 2 wells obtained a high-yield industrial gas flow of one million cubic meters, which was an increase of 23.13% compared with the pre-stack inversion prediction. It can be seen that the reservoir oil and gas prediction method based on multi-seismic attribute optimization provided in the embodiment of the present invention can greatly boost the success rate of well site optimization, bring huge economic benefits to oil field development, and is of great significance to promoting the exploration of tight sandstones and the search for natural gas exploration successor fields.

[0064] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0065] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0066] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0067] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0068] The specific embodiments described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A reservoir oil and gas prediction method based on multi-seismic attribute optimization, characterized in that: include: Acquire first seismic attribute data of the study area, wherein the first seismic attribute data includes: parameter data of multiple seismic attributes; According to the first seismic attribute data, second seismic attribute data meeting a preset seismic attribute condition is selected; Inputting the second seismic attribute data into a pre-trained classifier, and outputting third seismic attribute data after dimension reduction, wherein the classifier is a seismic attribute classification model obtained by machine learning training based on the relationship between known drilling data and seismic attributes; Predicting reservoir oil and gas in the study area according to the third seismic attribute data; Screening second seismic attribute data that meets a preset seismic attribute condition according to the first seismic attribute data includes: According to the correlation between the single seismic attribute data and the drilling data, the seismic attribute data with a correlation higher than a preset threshold are selected from the first seismic attribute data; Selecting, from the seismic attribute data with a correlation higher than a preset threshold, seismic attribute data that conform to the reservoir geological characteristic information of the study area; According to the correlation between the seismic attributes, the second seismic attribute data is selected from the seismic attribute data that conform to the reservoir geological characteristic information of the study area; The relationship between drilling data and seismic attributes is established based on a neural network algorithm, and a classifier is designed using an inflection point method; the classifier is a seismic attribute classification model obtained through machine learning training based on a neural network algorithm; Using actual drilling data as sample data, designing a classifier based on the sample data change rate, dividing categories according to the change rate, and establishing the relationship between drilling data and seismic attributes based on the neural network algorithm to classify unknown samples; The selection of dominant seismic attributes is based on the following three principles: first, whether the correlation between the seismic attributes and the actual drilling reservoir data is greater than the preset threshold; second, whether the planar distribution of the seismic attributes conforms to the geological laws, and the reservoir distribution characteristics are preliminarily clarified through the analysis of previous drilling data and the application of inversion technology; third, whether the correlation between the seismic attributes is within the preset range.

2. The method according to claim 1, characterized in that Before screening, from the first seismic attribute data, seismic attribute data having a correlation higher than a preset threshold value according to the correlation between the single seismic attribute data and the drilling data, the method further comprises: Obtaining drilling data for the study area; The parameter data of the single seismic attribute is intersected with the drilling data of the study area to determine the correlation between the single seismic attribute data and the drilling data.

3. The method according to claim 1, characterized in that Before selecting the seismic attribute data that conforms to the reservoir geological characteristic information of the study area from the seismic attribute data with a correlation higher than a preset threshold, the method further includes: Obtain reservoir geological characteristic information of the study area.

4. The method according to claim 1, characterized in that Before selecting the second seismic attribute data from the seismic attribute data that conform to the reservoir geological characteristic information of the study area according to the correlation between the seismic attributes, the method further comprises: The parameter data of different seismic attributes are intersected to determine the correlation between the different seismic attribute data.

5. The method according to any one of claims 1 to 4, characterized in that: The reservoir in the study area is a tight sandstone reservoir.

6. A reservoir oil and gas prediction device based on multi-seismic attribute optimization, characterized in that: include: The seismic attribute data acquisition module is used to acquire first seismic attribute data of the study area, wherein the first seismic attribute data includes: parameter data of multiple seismic attributes; A seismic attribute data screening module, used for screening second seismic attribute data that meets a preset seismic attribute condition according to the first seismic attribute data; A seismic attribute data dimensionality reduction processing module, used for inputting the second seismic attribute data into a pre-trained classifier, and outputting third seismic attribute data after dimensionality reduction, wherein the classifier is a seismic attribute classification model obtained by machine learning training based on the relationship between known drilling data and seismic attributes; A reservoir oil and gas prediction module, used for predicting reservoir oil and gas in the study area according to the third seismic attribute data; The earthquake attribute data screening module is specifically used for: According to the correlation between the single seismic attribute data and the drilling data, the seismic attribute data with a correlation higher than a preset threshold are selected from the first seismic attribute data; Selecting, from the seismic attribute data with a correlation higher than a preset threshold, seismic attribute data that conform to the reservoir geological characteristic information of the study area; According to the correlation between the seismic attributes, the second seismic attribute data is selected from the seismic attribute data that conform to the reservoir geological characteristic information of the study area; The relationship between drilling data and seismic attributes is established based on a neural network algorithm, and a classifier is designed using an inflection point method; the classifier is a seismic attribute classification model obtained through machine learning training based on a neural network algorithm; Using actual drilling data as sample data, designing a classifier based on the sample data change rate, dividing categories according to the change rate, and establishing the relationship between drilling data and seismic attributes based on the neural network algorithm to classify unknown samples; The selection of dominant seismic attributes is based on the following three principles: first, whether the correlation between the seismic attributes and the actual drilling reservoir data is greater than the preset threshold; second, whether the planar distribution of the seismic attributes conforms to the geological laws, and the reservoir distribution characteristics are preliminarily clarified through the analysis of previous drilling data and the application of inversion technology; third, whether the correlation between the seismic attributes is within the preset range.

7. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the reservoir oil and gas prediction method based on multi-seismic attribute optimization as described in any one of claims 1 to 5 is implemented.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method for predicting reservoir oil and gas based on multi-seismic attribute optimization according to any one of claims 1 to 5 is implemented.

Images