Earthquake reservoir prediction method and system
By constructing synthetic seismic records and inversion logging curve characteristics, and building a geological framework model, the low frequency deficiency and multi-solvency problems of seismic reservoir prediction in drilling-free areas are solved, the reservoir prediction accuracy is improved, and the development of oil and gas exploration technology is promoted. .
Patent Information
- Application Number
- CN202310508618.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-05-08
- Publication Date
- 2024-11-08
AI Technical Summary
In the drilling-free zone, seismic reservoir prediction has insufficient low frequency, multi-solvency and uncertainty, making it difficult to quantitatively characterize the reservoir distribution. Especially when there are multiple drilling wells but not drilling to the target layer, the existing technology is difficult to be effective. Predict reservoir distribution.
By constructing synthetic seismic records of the target area, determining the well-seismic relationship, using probability neural networks to invert the logging curve characteristics, constructing a geological framework model, and then predicting reservoir distribution. Specific steps include constructing synthetic seismic records, determining the well-seismic relationship, inverting the logging curve, and building a geological framework model to predict reservoir distribution.
It improves the accuracy of reservoir prediction, can effectively predict reservoir distribution, and provides a basis for oil and gas exploration selection and well location deployment, promotes the development of oil and gas exploration technology, and guides oil and gas exploration and increase storage and production.
Smart Images

Figure CN118915146A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of oil and gas layer exploration and development, and in particular to a seismic reservoir prediction method and system without drilling into a formation. Background Art
[0002] With the continuous deepening of exploration, research work such as the expansion of risk areas, zone evaluation and risk well location deployment continues to face new technical challenges, especially the evaluation and optimization of favorable zones for strata not encountered by drilling. How to use geophysical technology to predict the reservoir distribution of strata not encountered by drilling has become the key to breakthroughs in the risk field in the future.
[0003] Seismic reservoir prediction is the core technology of reservoir prediction. It is a process that quantitatively predicts the spatial geometry and parameter distribution characteristics of reservoirs, based on seismic technology, and constrained by seismic, geological laws, logging and drilling data. Seismic reservoir prediction in areas without drilling has always been a difficult point in reservoir prediction. Due to the lack of drilling data constraints, seismic reservoir prediction lacks low frequency, and the prediction results are not only multi-solution and uncertain, but also the reservoirs are often unable to be quantitatively characterized. Therefore, how to break through the difficulty of no well data constraints and accurately predict reservoir distribution has become the key to the success of expanding new areas or multi-layer three-dimensional exploration.
[0004] Previous researchers have explored and studied the seismic reservoir prediction and evaluation of strata in areas without well drilling. In summary, the following methods are mainly used: (1) Seismic inversion method for constructing virtual wells. This method constructs the low-frequency information of the P-wave velocity virtual well based on the stacking velocity field formed during the seismic processing process, and uses the seismic amplitude information after 90-degree phase shift to supplement the high-frequency component of the P-wave velocity curve of the well point. This curve is then used for seismic inversion reservoir prediction, and finally the spatial distribution characteristics of the reservoir are obtained. (2) Seismic multi-attribute joint reservoir prediction. This method makes full use of the seismic attributes of the neighboring areas and the analysis of actual drilling data to study the seismic response characteristics of the reservoir, determine the sensitive attributes of the reservoir, and then calibrate the reservoir in this area by combining multiple seismic attributes, and finally achieve reservoir prediction for the entire area. (3) Sparse pulse inversion based on prestack depth migration velocity field. This method constructs a low-frequency model with a high-precision prestack depth migration velocity field, and fuses the sparse pulse inversion results with the low-frequency information to obtain the wave impedance attributes.
[0005] The above research methods provide ideas for reservoir prediction in well-free areas and have achieved certain results, but these methods are mainly aimed at the situation where there are no wells, and have never involved the situation where there are wells in the study area but no target layer is encountered. In the case of multiple wells in the study area, make full use of the existing drilling data, extend the existing logging curve downward through well-seismic combination, and use the extended logging curve to perform well-seismic calibration, geological modeling and seismic inversion on the undrilled strata, and finally predict the spatial distribution law of the reservoir. This method is of great practical significance for guiding oil and gas exploration and development, expanding new exploration areas, conducting multi-layer stereoscopic exploration, and promoting the discovery of large oil and gas fields and increasing reserves and production. Summary of the invention
[0006] The present invention aims to solve at least one of the technical problems in the above-mentioned technology to a certain extent. To this end, the first aspect of the present invention proposes a seismic reservoir prediction method, comprising:
[0007] Construct synthetic seismic records of historical exploration wells in the target area;
[0008] Determining the well-seismic relationship of the target area according to the synthetic seismic record;
[0009] Inverting the well logging curve characteristics based on the probabilistic neural network and the well-seismic relationship;
[0010] Constructing an inversion logging curve of a target area according to the inversion logging curve characteristics;
[0011] Constructing a geological framework model based on the inversion logging curve;
[0012] The reservoir distribution in the target area is predicted based on the geological framework model.
[0013] Preferably, the step of constructing synthetic seismic records of historical exploration wells in the target area includes:
[0014] Based on the three-dimensional seismic data of the target area and the logging curves of the historical exploration wells in the target area, synthetic seismic records of the historical exploration wells in the target area are constructed.
[0015] Preferably, determining the well-seismic relationship of the target area according to the synthetic seismic records includes:
[0016] Extracting several seismic attributes sensitive to the reservoir response of the target area based on the seismic data of the target area;
[0017] The well-seismic relationship between each seismic attribute and the logging curves of historical exploration wells in the target area is used as the well-seismic relationship of the target area.
[0018] Preferably, before determining the well-seismic relationship between each seismic attribute and the well logging curve of the historical exploration wells in the target area, the method further includes:
[0019] Performing seismic attribute analysis on the plurality of seismic attributes, and removing abnormal values in data corresponding to the plurality of seismic attributes.
[0020] Preferably, the step of constructing an inversion logging curve of a target area according to the inversion logging curve characteristics includes:
[0021] Determine the compaction trend of historical exploration wells in the target area based on the well logging curve;
[0022] The compaction trend of the historical exploration wells in the target area is integrated with the characteristics of the inversion logging curve to obtain the inversion logging curve of the target area.
[0023] Preferably, the step of constructing an inversion logging curve of a target area according to the inversion logging curve characteristics includes:
[0024] Based on a probabilistic neural network, the compaction trend is taken as a low-frequency trend and fused with the inversion logging curve feature to obtain the inversion logging curve of the target area.
[0025] Preferably, constructing a geological framework model based on the inversion logging curve includes:
[0026] Determine the wave impedance response characteristics of the undrilled area of the target area according to the inversion logging curve and the three-dimensional seismic data of the target area, and generate a wave impedance curve;
[0027] A geological framework model is constructed according to the wave impedance curve.
[0028] A second aspect of the present invention discloses a seismic reservoir prediction system, comprising:
[0029] A record construction module is used to construct synthetic seismic records of historical exploration wells in the target area;
[0030] A relationship building module for determining the well-seismic relationship of the target area based on the synthetic seismic records;
[0031] Feature inversion module, used to invert the logging curve features based on the probabilistic neural network and the well-seismic relationship of the target area;
[0032] An inversion curve module is used to construct an inversion logging curve of a target area according to the characteristics of the inversion logging curve;
[0033] A model building module, used for building a geological framework model based on the inversion logging curve;
[0034] The reservoir prediction module is used to predict the reservoir distribution of the target area according to the geological framework model.
[0035] Preferably, the relationship building module executes the following steps:
[0036] Extracting a number of seismic attributes that are sensitive to reservoir response in the target area based on seismic data in the target area, performing seismic attribute analysis on the several seismic attributes, and removing abnormal values in data corresponding to the several seismic attributes;
[0037] The well-seismic relationship between each seismic attribute and the logging curve of the historical exploration wells in the target area is determined as the well-seismic relationship of the target area.
[0038] Preferably, the inversion curve module performs the following steps:
[0039] Determine the compaction trend of historical exploration wells in the target area based on the well logging curve;
[0040] The compaction trend of the historical exploration wells in the target area is integrated with the characteristics of the inversion logging curve to obtain the inversion logging curve of the target area.
[0041] Compared with the prior art, the present invention has the following beneficial effects:
[0042] The present invention makes full use of the existing drilling data in the study area. On the basis of fine calibration of wells and seismic data, the existing logging curves are extended downward through the combination of wells and seismic data to extend to strata that have not been drilled. Then, the extended logging curves are used to perform geological modeling and seismic inversion on strata that have not been drilled. This can not only effectively predict reservoir distribution, but also improve the accuracy of reservoir prediction, and provide a basis for the evaluation of favorable exploration zones and target optimization in the next step. The present invention further expands and improves the research methods of reservoir prediction in well-free areas, promotes the development of oil and gas exploration technology, guides oil and gas exploration area selection and well site deployment; provides technical support for the exploration and expansion of discovered oil and gas fields and the increase of reserves and production; the formed field evaluation technology has an extremely important guiding role in exploration and production, and provides important theoretical and technical support for planning support and exploration decision-making of major oil field companies.
[0043] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description and the accompanying drawings.
[0044] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS
[0045] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:
[0046] Figure 1A flow chart of a seismic reservoir prediction method provided in an embodiment;
[0047] Figure 2 A schematic diagram of the distribution of drilling wells in a certain area is provided for an embodiment;
[0048] Figure 3 A schematic diagram of a synthetic seismic record provided for an embodiment;
[0049] Figure 4 A schematic diagram of an earthquake attribute in a certain region provided in an embodiment;
[0050] Figure 5 It is a schematic diagram of compaction trend of AC curve given in the embodiment;
[0051] Figure 6 A schematic diagram of the relationship between the seismic amplitude attribute and the AC curve given in the embodiment;
[0052] Figure 7 A schematic diagram of an inverted AC curve provided in an embodiment;
[0053] Figure 8 A schematic diagram of the comparison of AC curves before and after inversion given in the embodiment;
[0054] Fig. 9 A schematic diagram of a synthetic seismic record based on an inverted AC curve is provided in an embodiment;
[0055] Fig.10 A schematic diagram of a low-frequency geological model of an undrilled stratum provided in an embodiment;
[0056] Fig.11 A schematic diagram of the wave impedance profile of an undrilled formation provided in an embodiment;
[0057] Fig.12 A schematic diagram of reservoir plane distribution provided for an embodiment;
[0058] Fig.13 The flowchart of the seismic reservoir prediction system given in the embodiment is shown in FIG. DETAILED DESCRIPTION
[0059] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.
[0060] In order to predict seismic reservoirs for strata that have not been drilled, the present invention provides a method for performing seismic reservoir prediction after extending the existing well logging curve downward. The method includes: using high-precision three-dimensional seismic data to finely calibrate the existing wells in the target area; preferably statistically analyzing the seismic attributes such as amplitude, phase, frequency and relative wave impedance that are relatively sensitive to reservoir characterization, and removing outliers; performing compaction trend analysis on the calibrated well logging curves to clarify the compaction trend of each well; establishing the relationship between seismic attributes and well logging curves; predicting the characteristics of the well logging curves of undrilled strata through probabilistic neural network inversion, and constructing well logging curves based on the predicted well logging curve characteristics and the compaction trend of each well; calibrating the undrilled strata through the predicted well logging curves; using the calibrated well logging curves to perform geological framework modeling; based on the establishment of the framework model, predicting the reservoir distribution through waveform indication inversion; extracting the reservoir thickness on the inversion data body, pointing out the favorable reservoir development area, and providing support for oil and gas exploration area selection and well site deployment. The above process is as follows: Figure 1 shown.
[0061] Figure 2 A schematic diagram of the distribution of drilling well locations in a certain area where the distribution of reservoirs in undrilled formations to be studied is provided for the embodiments of the present invention. The area is structurally located in the gentle belt in the eastern part of a basin, and is a monocline structure with high in the south and low in the north. The overall tectonic activity in the area is relatively weak, the strata are slightly deformed, and have the characteristics of small strata dip, gentle occurrence, and underdeveloped faults. The overall exploration level of the Lower Cambrian Canglangpu Formation in the basin is low. At present, several wells have been found in the northern slope area with good oil and gas displays. Among them, the Jiaotan 1 well has obtained a high-yield industrial gas flow, especially the nearby Heshen 1 well has rich oil and gas displays in the Cang section, showing the good exploration prospects of the Canglangpu Formation. Geological studies have shown that the Canglangpu Formation inherits the paleogeographic pattern of the Qiongzhusi Formation, and has an overall sedimentary pattern of one uplift, three depressions and two high belts. Controlled by sedimentary paleo-geomorphology, the area has developed large-scale granular beaches and is close to the source rocks of the Cambrian Qiongzhusi Formation, with efficient drainage and accumulation conditions, making it a favorable area for risk exploration. Although there are 7 exploration wells ( Figure 2 As shown by the dots in the figure, the character string corresponding to the dots is the name of the exploration well, L and T represent the line number and the track number respectively), but no well has encountered the Canglangpu Formation. It is impossible to accurately and quantitatively characterize the reservoir in this area by relying solely on seismic configuration or seismic attribute prediction, nor is it possible to finely characterize the distribution range of favorable reservoirs, and it is impossible to promote the deployment of risk wells in a timely manner. It also seriously hinders the fine exploration of the Canglangpu Formation in this area, so the technical method of the present invention is used for research.
[0062] The process of predicting the undrilled seismic reservoirs in the above-mentioned area based on the seismic reservoir prediction method provided by the present invention comprises the following steps:
[0063] S1. Based on the high-precision three-dimensional seismic data and well logging curves in the area, the synthetic seismic records of the seven existing exploration wells in the area were calibrated by combining well-seismic and well-seismic data. With reference to the bottom boundary of a group of stable strong wave peaks in the whole area, the synthetic records of each well were finely adjusted to match the seismic data, ensuring a good correspondence between the well logging curve and the seismic waveform. Figure 3 This is a synthetic seismic record of one of the exploration wells, which is synthesized based on the AC logging curve (acoustic time difference logging curve). TVD represents the vertical depth. P2ch, P2l, P1m, P1l and Cm2x are all formation codes. Among them, P2ch represents the Changxing Formation of the Permian System, and P2l represents the Longtan Formation of the Permian System.
[0064] S2. Using the three-dimensional seismic data of the area, extract seismic attributes such as amplitude, phase, frequency and impedance that are sensitive to reservoir reflection, and perform statistical analysis on various attributes to remove outliers, so as to facilitate the subsequent step S4 to analyze the relationship between attributes and logging curves. Figure 4 It is a schematic diagram of the instantaneous energy properties of three-dimensional earthquakes in the area. The ordinate of the figure represents the vertical depth of the stratum, and the abscissa represents the line number. The scale L:603T:660 in the figure corresponds to the line number 603 and the track number 660, and the same below. The grayscale represents the instantaneous energy intensity.
[0065] S3. Perform compaction trend analysis on the AC curves of the 7 calibrated wells and establish a unified compaction trend. This compaction trend is used as a low-frequency trend together with the neural network inversion to extend the AC curve of the undrilled formation. Figure 5 This is the compaction trend analysis chart of the AC curve of the above seven exploration wells. The horizontal axis represents the compaction trend and the vertical axis represents time.
[0066] S4. Relationship between statistical seismic single attributes and well logging curves ( Figure 6 This is a relationship diagram between the seismic amplitude attribute and the AC curve. The horizontal axis represents the acoustic wave time difference and the vertical axis represents the seismic amplitude), which lays the foundation for multi-attribute neural network prediction.
[0067] S5. The logging curve characteristics of the undrilled strata (Canglangpu Formation) are predicted by the probabilistic neural network inversion method, and the compaction trend in S3 is used as a low-frequency component to merge with the logging curve characteristics to form an inverted logging curve. Figure 7 The AC curve is obtained based on the probabilistic neural network inversion. The ordinate represents the vertical depth, and the values after the L and T on the abscissa represent the line number of the corresponding scale. The same below. Figure 8 Comparison of AC curves before and after inversion. Gray represents the original AC curve, and black represents the inverted AC curve.
[0068] S6. Recreate synthetic seismic records using the inverted predicted AC log curves ( Fig. 9This is a synthetic seismic record based on the inverted AC curve. The meanings of the characters in the figure are as follows: Figure 2 Corresponding text description) to determine the wave impedance response characteristics of the undrilled formation.
[0069] S7. Generate a new wave impedance curve based on the inverted predicted AC logging curve, and on this basis, establish a geological framework model. The geological framework model constructed based on the above inverted AC curve is as follows: Fig.10 As shown, the longitudinal broken line in the figure is the AC curve, the horizontal broken line with the same grayscale represents the stratum, and E1l and E1c are the stratum codes respectively.
[0070] S8. Based on the geological framework model in S7 and high-precision three-dimensional seismic data, the inversion parameters are debugged through the waveform indication inversion method, and the inversion data body is analyzed to obtain the reservoir thickness and predict the reservoir distribution. Fig.11 Based on Fig.10 The wave impedance well-connected profile of the Canglangpu Formation, an undrilled stratum, obtained from the corresponding geological model. E1l and E1c are the stratum codes, corresponding to the white horizontal broken lines; Fig.12 This is the predicted reservoir plane distribution map of the Canglangpu Formation in this area. The grayscale in the figure represents the E1c2 lithology thickness, and L and T represent the line numbers respectively.
[0071] Fig.13 A seismic reservoir prediction system provided by the present invention comprises:
[0072] A record construction module is used to construct synthetic seismic records of historical exploration wells in the target area;
[0073] A relationship building module for determining the well-seismic relationship of the target area based on the synthetic seismic records;
[0074] Feature inversion module, used to invert the logging curve features based on the probabilistic neural network and the well-seismic relationship of the target area;
[0075] An inversion curve module is used to construct an inversion logging curve of a target area according to the characteristics of the inversion logging curve;
[0076] Model building module, used to build geological framework model based on inversion logging curve;
[0077] The reservoir prediction module is used to predict the reservoir distribution in the target area based on the geological framework model.
[0078] In addition, the present invention also provides an electronic device, which is used to implement at least one of the methods given in the above embodiments; the present invention also provides a computer storage medium, which stores computer instructions for implementing at least one of the methods given in the above embodiments.
[0079] Beneficial effects of the above technical solutions: The present invention makes full use of the existing drilling data in the study area. On the basis of fine calibration of wells and seismic, the existing logging curves are extended downward through the combination of wells and seismic, and can be extended to the strata that have not been drilled. Then, the extended logging curves are used to perform geological modeling and seismic inversion on the strata that have not been drilled. This can not only effectively predict the distribution of reservoirs, but also improve the accuracy of reservoir prediction, and provide a basis for the evaluation of favorable exploration zones and target optimization in the next step. The present invention further expands and improves the research methods of reservoir prediction in wellless areas, promotes the development of oil and gas exploration technology, guides the selection of oil and gas exploration areas and the deployment of well sites; provides technical support for the exploration and expansion of discovered oil and gas fields and the increase of reserves and production; the formed field evaluation technology has an extremely important guiding role in exploration and production, and provides important theoretical and technical support for the planning support and exploration decision-making of major oil field companies.
[0080] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.
Claims
1. A seismic reservoir prediction method, characterized in that: include: Construct synthetic seismic records from historical exploration wells in the target area; determining a well-seismic relationship in a target area based on the synthetic seismic record; Obtaining inversion logging curve characteristics based on the probabilistic neural network and the well-seismic relationship inversion; Constructing an inversion logging curve of a target area according to the characteristics of the inversion logging curve; Constructing a geological framework model based on the inversion logging curve; The reservoir distribution in the target area is predicted based on the geological framework model.
2. The seismic reservoir prediction method according to claim 1, wherein: The method of constructing synthetic seismic records of historical exploration wells in the target area includes: Synthetic seismic records of historical exploration wells in the target area are constructed based on the three-dimensional seismic data of the target area and the logging curves of historical exploration wells in the target area.
3. The seismic reservoir prediction method according to claim 1, wherein: Determining the well-seismic relationship of the target area based on the synthetic seismic records includes: Extracting several seismic attributes that are sensitive to the reservoir response of the target area based on the seismic data of the target area; The well-seismic relationship between each seismic attribute and the logging curves of historical exploration wells in the target area is used as the well-seismic relationship of the target area.
4. The seismic reservoir prediction method according to claim 3, wherein: Before determining the well-seismic relationship between each seismic attribute and the well logging curves of historical exploration wells in the target area, it also includes: Performing seismic attribute analysis on the plurality of seismic attributes and removing abnormal values in data corresponding to the plurality of seismic attributes.
5. The seismic reservoir prediction method according to claim 1, wherein: The step of constructing an inversion logging curve of a target area according to the inversion logging curve characteristics includes: Determine the compaction trend of historical exploration wells in the target area based on well logging curves; The compaction trend of historical exploration wells in the target area is integrated with the characteristics of the inversion logging curve to obtain the inversion logging curve of the target area.
6. The seismic reservoir prediction method according to claim 5, characterized in that: The step of constructing an inversion logging curve of a target area according to the inversion logging curve characteristics includes: Based on a probabilistic neural network, the compaction trend is taken as a low-frequency trend and fused with the inversion logging curve feature to obtain the inversion logging curve of the target area.
7. The seismic reservoir prediction method according to any one of claims 1 to 6, characterized in that: The step of constructing a geological framework model based on the inversion logging curve includes: determining the wave impedance response characteristics of an undrilled area of the target area based on the inversion logging curve and the three-dimensional seismic data of the target area, and generating a wave impedance curve; A geological framework model is constructed according to the wave impedance curve.
8. A seismic reservoir prediction system, characterized in that: include: A record construction module is used to construct synthetic seismic records of historical exploration wells in the target area; a relationship building module for determining the well-seismic relationship in the target area based on synthetic seismic records; Feature inversion module, used to invert the logging curve characteristics based on the probabilistic neural network and the well-seismic relationship of the target area; Inversion curve module, used to construct the inversion logging curve of the target area according to the characteristics of the inversion logging curve; A model building module, used for building a geological framework model based on the inversion logging curve; The reservoir prediction module is used to predict the reservoir distribution of the target area based on the geological framework model.
9. The seismic reservoir prediction system according to claim 8, wherein: The relationship building module executes the following steps: Extracting abnormal values from data corresponding to several seismic attributes that are sensitive to reservoir response in the target area based on seismic data in the target area; The well-seismic relationship between each seismic attribute and the logging curves of historical exploration wells in the target area is determined as the well-seismic relationship of the target area.
10. The seismic reservoir prediction system according to claim 8, wherein: The inversion curve module performs the following steps: Determine the compaction trend of historical exploration wells in the target area based on well logging curves; The compaction trend of historical exploration wells in the target area is integrated with the characteristics of the inversion logging curve to obtain the inversion logging curve of the target area.
Citation Information
Patent Citations
Formation pressure prediction method and lithologic oil reservoir evaluation method
CN111060986A
Method and device for predicting thin reservoir in lacustrine facies beach bar sand bodies
CN111856566A
Reservoir gas content prediction method based on depth domain seismic wave frequency dispersion inversion
CN115047525A
Pre-stack and post-stack inversion method for fractured and porous reservoir
CN115508890A
