Method for improving prediction accuracy of clastic rock thin reservoir, application, device and medium

Abstract

The present application belongs to the technical field of oil and gas seismic exploration interpretation, and discloses a method and application for improving prediction accuracy of clastic rock thin reservoir, a device and a medium. The method is to first identify the thin sand body plane distribution profile through seismic waveform decomposition, then identify the thin sand body boundary through seismic waveform indication simulation, and finally depict the thin sand body details through post-stack geostatistical inversion, that is, complete the prediction of clastic rock thin reservoir. The three post-stack reservoir prediction methods are combined in a specific order after improvement, and the combination application is realized layer by layer and gradually from the thin sand body plane distribution profile to the thin sand body boundary and the thin sand body internal details. The device and the computer readable storage medium according to the above method can be applied to improve the prediction accuracy of clastic rock thin reservoir, and the present application is suitable for clastic rock thin reservoir lithologic trap and oil and gas reservoir depiction.

Description

Technical Field

[0001] This invention belongs to the field of oil and gas seismic exploration and interpretation technology, and relates to a reservoir prediction method, specifically a method, application, device and medium for improving the prediction accuracy of thin clastic reservoirs. Background Technology

[0002] Today, geological exploration targets have shifted from large structural traps in clastic rocks to concealed traps based on lithology and other factors. Among these, reservoir prediction is one of the key technologies for interpreting lithological traps.

[0003] Currently, there are three main methods for reservoir prediction in clastic rocks, each with its own advantages and disadvantages: First, the seismic waveform decomposition method, which selects waveform information closely related to reservoir changes to highlight the seismic response characteristics of the reservoir, thereby achieving planar identification of the reservoir. However, its results are entirely driven by seismic activity and cannot achieve vertical differentiation; furthermore, its precision needs improvement. Second, the seismic waveform indication simulation method, which mainly compares waveforms with those of known points. Based on waveform similarity and distance, statistical algorithms are used to statistically calculate target curves that can distinguish reservoirs from non-reservoirs at the predicted points. However, this method is less effective for predicting seismic response characteristics. For thin sand bodies with unclear characteristics, the prediction accuracy will be greatly reduced, and the prediction results are also greatly affected by the geological framework model; thirdly, post-stack geostatistical inversion fully integrates prior information such as geological information, seismic information, rock physics information, and well logging information. Through Markov chain and Monte Carlo algorithm, a series of geological models that meet various soft and hard constraints are generated. This geological model has both descriptive properties that match the measured well logging data and predictive properties that are compatible with seismic data. However, this method requires that the P-wave impedance can distinguish between reservoirs and non-reservoirs, and that there must be a sufficient number of wells with relatively uniform distribution in the study area.

[0004] The aforementioned reservoir prediction methods for clastic rocks can leverage their respective advantages for predicting general clastic rock reservoirs. However, when dealing with clastic rock reservoirs in basin areas where the reservoirs are generally extremely deep and thin, the seismic data suffers from low vertical resolution due to the low frequency (around 30 Hz) and narrow bandwidth (approximately 8-65 Hz) of these thin clastic rock reservoirs, caused by background reflections masking the seismic reflection information of the thin sand layers. Furthermore, the seismic response characteristics of the thin sand layers are unclear. Therefore, simply applying a single reservoir prediction method to thin clastic rock reservoirs cannot achieve accurate predictions, and the resulting predictions suffer from problems such as weak planar geological regularity and unclear detail characterization. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention aims to provide a method for improving the prediction accuracy of thin clastic reservoirs. By connecting and combining three post-stack reservoir prediction methods—seismic waveform decomposition, seismic waveform indication simulation, and post-stack geostatistical inversion—in a specific order, the method progressively achieves the goal of improving the spatial prediction accuracy of thin clastic reservoirs, from the planar distribution outline of thin sand bodies to the boundary of thin sand bodies and even the detailed internal features of thin sand bodies.

[0006] Another objective of this invention is to provide applications, apparatus, and computer-readable storage media based on the above-described method for improving the prediction accuracy of thin clastic reservoirs.

[0007] To achieve the above objectives, the technical solution adopted by the present invention is as follows: A method for improving the prediction accuracy of thin clastic reservoirs includes the following steps: First, the planar distribution outline of the thin sand body is identified by seismic waveform decomposition. Then, the boundary of the thin sand body is identified by seismic waveform indication simulation. Finally, the details of the thin sand body are characterized by post-stack geostatistical inversion, thus completing the prediction of clastic thin reservoirs.

[0008] As a limitation of the present invention, the identification of the planar distribution contour of thin sand bodies through seismic waveform decomposition includes the following steps performed sequentially: By using well-seismic calibration synthetic records, well-seismic calibration results are obtained, and the longitudinal position of the target layer and the seismic reflection layer are determined. Based on the well-seismic calibration results, the seismic reflection horizon interpretation of the target layer was completed, and the seismic waveform decomposition of the target layer was performed. Based on the seismic reflection horizon interpretation, the seismic waveform components that identify the planar distribution profile of the thin sand body were obtained.

[0009] As another limitation of the present invention, the method of identifying thin sand body boundaries by seismic waveform indication simulation includes the following steps performed sequentially: By combining seismic waveform components with single-well sedimentary facies delineation results, a sedimentary facies planar map of the target layer is depicted; Using the well-seismic calibration results and the interpretation of the seismic reflection horizons, time difference corrections were performed on the horizons according to the geological strata to generate new horizons that correspond one-to-one with the geological strata. A fine geological framework model was then established based on the new horizons. Using the gamma curve that can distinguish between reservoirs and non-reservoirs as the target curve, an initial gamma model is established under the constraint of the sedimentary facies plane map of the target layer through an interpolation algorithm. Under the constraints of the initial gamma model, seismic waveform indication simulation is carried out to obtain a three-dimensional spatial gamma body that determines the boundary of the thin sand body.

[0010] As a further limitation of the present invention, the description of thin sandstone body details through post-stack geostatistical inversion includes the following steps performed sequentially: Based on the analysis results of the rock physics of the target layer, a piecewise calculation formula for the probability of sandstone or mudstone is determined. This formula is used to transform the three-dimensional spatial gamma body into a lithofacies probability body. The three-dimensional lithofacies probability model is used to replace the zero-dimensional constant sandstone-mudstone ratio model. The probability density distribution function of different lithofacies is set, and post-stack geostatistical inversion is carried out to characterize the details of thin sandstone bodies and complete the prediction of thin clastic reservoirs.

[0011] As a further limitation of the present invention, the target layer depositional phase planar diagram is characterized at the subphase level.

[0012] As another limitation of the present invention, the post-stack geostatistical inversion is a post-stack geostatistical inversion based on longitudinal wave impedance.

[0013] As a further limitation of the present invention, the gamma curve that can distinguish between reservoirs and non-reservoirs is based on the rock physical analysis of the target layer.

[0014] As a further limitation of the present invention, when the upper limit A of the gamma value of sandstone is determined to be less than or equal to the lower limit B of the gamma value of mudstone based on the analysis results of rock physics, the piecewise calculation formula for the probability of sandstone or mudstone is as follows: If the gamma value is less than or equal to A, then the probability of sandstone is 1 and the probability of mudstone is 0. If the gamma value is greater than or equal to B, then the probability of sandstone is 0 and the probability of mudstone is 1. If A < gamma value < B, then the probability of sandstone is (B-GR) / (BA), and the probability of mudstone is (GR-A) / (BA), where GR is the gamma value.

[0015] The present invention also provides an apparatus, which is an electronic device, including a memory and a processor, wherein the memory stores executable instructions; the processor executes the executable instructions in the memory to implement the above-described method for improving the prediction accuracy of clastic thin reservoirs.

[0016] Furthermore, the device is a computer device.

[0017] The present invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method for improving the prediction accuracy of thin clastic reservoirs.

[0018] By adopting the above-described technical solution, the beneficial effects achieved by this invention compared to the prior art are as follows: (1) The method of improving the prediction accuracy of thin clastic reservoirs of the present invention utilizes high-fidelity and high-amplitude three-dimensional seismic data. Based on fine synthetic record calibration, geological framework model establishment, and rock physical analysis, the planar distribution contour of thin sand bodies is identified by seismic waveform decomposition, the boundary of thin sand bodies is identified by seismic waveform indication simulation, and the results of rock physical analysis are combined to obtain the lithofacies probability volume. Finally, the post-stack geostatistical inversion of lithofacies probability volume constraint is performed to characterize the details of thin sand bodies. Through the series and combined application of these three reservoir prediction methods, the method progresses step by step to gradually achieve the goal of characterizing the details from the sand body contour to the sand body boundary and even the interior of the sand body.

[0019] (2) The method of improving the prediction accuracy of clastic thin reservoirs of the present invention produces a synergistic effect by improving three reservoir prediction methods and combining them in a specific way. The method of establishing a new stratigraphic fine geological framework model obtained through improvement and innovation makes the modeling more refined and more applicable to the prediction of thinner reservoirs.

[0020] (3) The method of improving the prediction accuracy of clastic thin reservoirs of the present invention only needs to identify the subfacies level during the sedimentary facies characterization process, so as to avoid the problem of excessive characterization scale level leading to serious modeling of subsequent prediction results. At the same time, the smaller the sedimentary facies characterization scale, the higher the accuracy requirement of the result, and the more difficult it is to achieve.

[0021] (4) The method of improving the prediction accuracy of thin clastic reservoirs of the present invention was applied in Block H of the clastic oilfield in the Tarim Basin. Compared with any one of the following methods: seismic waveform decomposition, seismic waveform indicator simulation and post-stack geostatistical inversion, the method has the effect of gradually improving the prediction accuracy of thin sand layers, making the final predicted sand body boundary clear and natural, and the details inside the sand body are also clearly depicted. The well matching accuracy is also improved to 93%.

[0022] The device and computer-readable storage medium of the present invention can realize a method to improve the prediction accuracy of thin clastic reservoirs. It has the advantages of simple operation, high efficiency and high prediction accuracy. The present invention is applicable to the characterization of lithological traps and oil and gas reservoirs in clastic strata. Attached Figure Description

[0023] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments.

[0024] Figure 1 This is a planar diagram of the seismic waveform components of the planar distribution profile of the thin sand body in Embodiment 1 of the present invention; Figure 2 This is a plan view of the sedimentary phase of the target layer in Embodiment 1 of the present invention; Figure 3 This is a detailed geological framework model diagram from Embodiment 1 of the present invention; Figure 4This is the initial gamma model diagram in Embodiment 1 of the present invention; Figure 5 This is a gamma-ray planar view of the thin sand layer in Embodiment 1 of the present invention; Figure 6 This is a spatial visualization diagram of the probabilistic volume of lithofacies in Embodiment 1 of the present invention; Figure 7 This is a planar diagram of the wave impedance of the thin sand layer in Embodiment 1 of the present invention; Figure 8 This is a geological framework model diagram of Comparative Example 1 of the present invention; Figure 9 This is a diagram of the initial gamma model in Comparative Example 2 of the present invention; Figure 10 This is a gamma-ray planar view of the thin sand layer in Comparative Example 2 of the present invention; Figure 11 This is a geological framework model diagram of Comparative Example 3 of the present invention. Detailed Implementation

[0025] The present invention will be further described in detail below with reference to specific embodiments and accompanying drawings. It should be understood that the described embodiments are only used to explain the present invention and do not limit the present invention.

[0026] Example 1: A method and application for improving the prediction accuracy of thin clastic reservoirs. This embodiment utilizes a method to improve the prediction accuracy of thin clastic rock reservoirs to predict reservoirs in a certain location, specifically including the following steps: S1. Identifying the planar distribution profile of thin sand bodies through seismic waveform decomposition. By utilizing drilling, logging, and geological data, well-seismic calibration and synthesis records are made on 3D seismic data to obtain well-seismic calibration results and determine the longitudinal position of the target layer and the seismic reflection layer. Based on the well seismic calibration results, the interpretation of the seismic reflection horizon of the target layer was completed; Based on prior geological understanding and well completion data, seismic waveform decomposition was performed on the target layer. Using the interpretation of the obtained seismic reflection horizons as a foundation, appropriate time windows were opened above and below, and the analysis time windows were adjusted and optimized multiple times until a seismic waveform component that could relatively accurately reflect the planar distribution profile and pinch-out point location of the thin sand body, and had a certain degree of consistency with the well completion data, was obtained. This seismic waveform component identifying the planar distribution profile of the thin sand body is shown in the planar diagram of the obtained thin sand body planar distribution profile, as follows: Figure 1 ; S2. Identifying thin sand body boundaries using seismic waveform indicators. By combining seismic waveform components with single-well sedimentary facies classification results, the preliminary 3D seismic prediction results are determined to reflect the scale of sedimentary facies at the subfacies level, thus characterizing the sedimentary facies planar map of the target layer, such as... Figure 2 ; Using the well-seismic calibration results and the obtained seismic reflection horizon interpretation, time difference corrections were performed on the horizons according to geological strata, and time difference corrections were also performed on the seismically interpreted horizons, generating a series of new horizons that correspond one-to-one with each geological stratum. A refined geological framework model was then established based on these new horizons, such as... Figure 3 ; For the target layer, rock physical analysis, excluding geophysical elastic parameters such as P-wave impedance and S-wave impedance, is used to select the target curve that can distinguish between reservoirs and non-reservoirs. Since the gamma curve has a good ability to identify sandstone and mudstone, the target curve is selected as the gamma curve. Using the above gamma curves, a sedimentary facies plane diagram of the target layer ( Figure 2 Under the constraints of ), an initial gamma model is established by well-to-well interpolation using an interpolation algorithm. The profile of this model is as follows: Figure 4 ; Based on the well-seismic calibration results in step S1 and the refined geological framework model in S2, and constrained by the obtained initial gamma model, a well-seismic combination is performed. For the target layer, using the gamma curve as the target curve, seismic waveform indication simulation is conducted to predict the spatial distribution characteristics of thin sand bodies, obtain the three-dimensional spatial gamma body that defines the boundaries of the thin sand bodies, and generate... Figure 5 The gamma plane diagram of the thin sand layer, through the above processing, further improves the well-well fit of the thin sand body prediction and refines the sand body boundary; S3. Characterizing the details of thin sand bodies through post-stack geostatistical inversion. Based on the rock physics analysis results of the target layer in step S2, if the upper limit A of the gamma value of sandstone is determined to be less than or equal to the lower limit B of the gamma value of mudstone, then the piecewise calculation formula for the probability of sandstone or mudstone is as follows: If the gamma value is less than or equal to A, then the probability of sandstone is 1 and the probability of mudstone is 0. If the gamma value is greater than or equal to B, then the probability of sandstone is 0 and the probability of mudstone is 1. If A < gamma value < B, then the probability of sandstone is (B-GR) / (BA), and the probability of mudstone is (GR-A) / (BA), where GR is the gamma value; The three-dimensional spatial gamma body obtained in step S2 is transformed into a lithofacies probability volume using a piecewise calculation formula, and then visualized as follows: Figure 6 A three-dimensional lithofacies probability model was used to replace the zero-dimensional constant sandstone-mudstone ratio model. Probability density distribution functions for different lithofacies were set, and post-stack geostatistical inversion based on P-wave impedance was performed to characterize the details of thin sand bodies and complete the prediction of thin clastic reservoirs. The resulting thin sand layer impedance planar diagram is shown below. Figure 7 .

[0027] Depend on Figure 7 It can be seen that by using the method of the present invention to improve the prediction accuracy of thin clastic reservoirs, the final predicted sand body boundary is clear and natural, and the details inside the sand body are also clearly depicted, with high well-to-well consistency.

[0028] Comparative Example 1: Method for Predicting Thin Clastic Reservoirs via Seismic Waveform Decomposition This comparative example uses seismic waveform decomposition to predict thin clastic reservoirs. The difference from Example 1 is that step S2 does not utilize well-seismic calibration results and seismic reflection horizons to perform time-difference correction on the seismic interpretation horizons; instead, it directly establishes a geological framework model. The resulting geological framework model is as follows: Figure 8 .

[0029] Depend on Figure 3 and Figure 8 The comparison shows that, Figure 8 The previous method could only control the top and bottom of large layers, while the present invention... Figure 3 It not only achieved control over the top and bottom of the large layer, but also provided a more detailed characterization of the internal strata. Furthermore, the internal strata and well data, such as the two internal stratigraphic lines, coincided well with LP_ch61 and LP_ch62 of well YM461, respectively, characterizing at least four strata and refining the geological framework model.

[0030] Comparative Example 2: A method for predicting thin clastic reservoirs without being constrained by the sedimentary facies plan of the target layer. The difference between this comparative example and Example 1 is that step S2 is not constrained by the target layer depositional facies planar diagram, but directly establishes an initial gamma model, as shown in the example. Figure 9 The obtained gamma-ray planar diagram of the thin sand layer is as follows: Figure 10 .

[0031] Depend on Figure 4 and Figure 9 A comparison reveals that the curves indicated by the arrows in the two figures are significantly different. Figure 4 The curve indicated by the middle arrow shows the pinch-out and thinning of the sand body at the circular mark on the left, which can accurately and truly reflect the actual situation underground. Figure 9 The curve indicated by the middle arrow extends far to the left, but common sense dictates that actual underground sand bodies cannot extend indefinitely. Therefore... Figure 9 This does not align with common sense.

[0032] The above results show that the clastic thin reservoir prediction method of the present invention can more accurately reflect the actual thin reservoir situation and can achieve accurate prediction compared with methods that are not constrained by the sedimentary facies plan of the target layer.

[0033] Depend on Figure 5 and Figure 10 The comparison shows that, compared to Figure 10 The present invention Figure 5 In the middle, the conclusion that the sand body comes from the source direction in the upper right is presented more clearly with less interference, indicating that the prediction effect of the method of the present invention is better.

[0034] Comparative Example 3: Method for Predicting Thin Clastic Reservoirs Using Post-Stack Geostatistical Inversion Compared with Example 1, in this comparative example, step S3 does not utilize lithofacies probability volume constraints, but instead directly performs inversion, resulting in a geological framework model as follows: Figure 11 .

[0035] Depend on Figure 7 and Figure 11 As can be seen from the comparison, the present invention Figure 7 In the middle, the distribution of sand bodies is more regular, while Figure 11 The irregular and dispersed distribution of the medium sand bodies indicates that the geological framework model established by the method of the present invention is more accurate, which is beneficial for the accurate prediction of thin clastic reservoirs.

[0036] Example 2: Apparatus for improving the prediction accuracy of thin clastic reservoirs This embodiment is an electronic device for improving the prediction accuracy of thin clastic reservoirs, including a memory and a processor.

[0037] The memory stores executable instructions; the processor runs the executable instructions in the memory to implement the method for improving the prediction accuracy of clastic thin reservoirs in Example 1.

[0038] This memory is used to store non-transitory computer-readable instructions. Specifically, the memory may include one or more computer program products, which may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may, for example, include random access memory (RAM) and / or cache memory. The non-volatile memory may, for example, include read-only memory (ROM), hard disk, flash memory, etc.

[0039] The processor may be a central processing unit (CPU) or other processing unit with data processing and / or instruction execution capabilities, and may control other components in the electronic device to perform desired functions. In one embodiment disclosed in this application, the processor is used to execute computer-readable instructions stored in the memory.

[0040] Those skilled in the art will understand that, in order to solve the technical problem of how to achieve a good user experience, this embodiment may also include well-known structures such as communication buses and interfaces, and these well-known structures should also be included within the protection scope of this disclosure.

[0041] Example 3: A computer-readable storage medium This embodiment provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method for improving the prediction accuracy of clastic thin reservoirs in Embodiment 1.

[0042] The computer-readable storage medium stores non-transitory computer-readable instructions thereon. When the non-transitory computer-readable instructions are executed by a processor, all or part of the steps of the methods of the foregoing embodiments are performed.

[0043] The aforementioned computer-readable storage media include, but are not limited to: optical storage media (e.g., CD-ROM and DVD), magneto-optical storage media (e.g., MO), magnetic storage media (e.g., magnetic tape or portable hard drive), media with built-in rewritable non-volatile memory (e.g., memory card), and media with built-in ROM (e.g., ROM cartridge).

[0044] It should be noted that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art can still modify the technical solutions described in the above embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the claims of the present invention.

Claims

1. A method for improving the prediction accuracy of thin clastic reservoirs, characterized in that, Includes the following steps: First, the planar distribution outline of the thin sand body is identified by seismic waveform decomposition. Then, the boundary of the thin sand body is identified by seismic waveform indication simulation. Finally, the details of the thin sand body are characterized by post-stack geostatistical inversion, thus completing the prediction of clastic thin reservoirs. The method of identifying the planar distribution profile of thin sand bodies through seismic waveform decomposition includes the following steps performed sequentially: By using well-seismic calibration synthetic records, well-seismic calibration results are obtained, and the longitudinal position of the target layer and the seismic reflection layer are determined. Based on the well-seismic calibration results, the seismic reflection horizon interpretation of the target layer is completed, and the seismic waveform decomposition of the target layer is performed. Based on the seismic reflection horizon interpretation, the seismic waveform components that identify the planar distribution profile of the thin sand body are obtained. The method of identifying thin sand body boundaries through seismic waveform indication simulation includes the following steps performed sequentially: By combining seismic waveform components with single-well sedimentary facies delineation results, a sedimentary facies planar map of the target layer is depicted; Using the well-seismic calibration results and the interpretation of the seismic reflection horizons, time difference corrections were performed on the horizons according to the geological strata to generate new horizons that correspond one-to-one with the geological strata. A fine geological framework model was then established based on the new horizons. Using the gamma curve that can distinguish between reservoirs and non-reservoirs as the target curve, an initial gamma model is established under the constraint of the sedimentary facies plane map of the target layer through an interpolation algorithm. Under the constraints of the initial gamma model, seismic waveform indication simulation was carried out to obtain a three-dimensional spatial gamma body that determines the boundary of the thin sand body; The method of characterizing thin sand bodies through post-stack geostatistical inversion includes the following steps performed sequentially: Based on the analysis results of the rock physics of the target layer, a piecewise calculation formula for the probability of sandstone or mudstone is determined. This formula is used to transform the three-dimensional spatial gamma body into a lithofacies probability body. The three-dimensional lithofacies probability model is used to replace the zero-dimensional constant sandstone-mudstone ratio model. The probability density distribution function of different lithofacies is set, and post-stack geostatistical inversion is carried out to characterize the details of thin sandstone bodies and complete the prediction of thin clastic reservoirs.

2. The method for improving the prediction accuracy of thin clastic reservoirs according to claim 1, characterized in that, The target layer sedimentary facies planar diagram is characterized at the subfacies level.

3. The method for improving the prediction accuracy of thin clastic reservoirs according to claim 1, characterized in that, The post-stack geostatistical inversion is a post-stack geostatistical inversion based on longitudinal wave impedance.

4. The method for improving the prediction accuracy of thin clastic reservoirs according to claim 1 or 2, characterized in that, The gamma curves that can distinguish between reservoirs and non-reservoirs are based on the petrophysical analysis of the target layer.

5. The method for improving the prediction accuracy of thin clastic reservoirs according to claim 4, characterized in that, When, based on rock physics analysis, the upper limit A of the gamma value of sandstone is determined to be less than or equal to the lower limit B of the gamma value of mudstone, the piecewise calculation formula for the probability of the sandstone or mudstone is as follows: If the gamma value is less than or equal to A, then the probability of sandstone is 1 and the probability of mudstone is 0. If the gamma value is greater than or equal to B, then the probability of sandstone is 0 and the probability of mudstone is 1. If A < gamma value < B, then the probability of sandstone is (B-GR) / (BA), and the probability of mudstone is (GR-A) / (BA), where GR is the gamma value.

6. A device for improving the prediction accuracy of thin clastic reservoirs, characterized in that, An electronic device includes a memory and a processor, wherein the memory stores executable instructions; the processor executes the executable instructions in the memory to implement the method for improving the prediction accuracy of clastic thin reservoirs as described in any one of claims 1 to 5.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method for improving the prediction accuracy of clastic thin reservoirs as described in any one of claims 1 to 5.

Images