Statistics-based beach facies carbonate reservoir quantitative prediction method

Abstract

The invention discloses a beach facies carbonate reservoir quantitative prediction method based on statistics. The method comprises the following steps: 1, identifying a reservoir development favorable facies belt; 2, establishing a reservoir seismic response mode; 3, carrying out waveform classification, and qualitatively identifying a low-resistance reservoir development region and a high-resistance reservoir development region in the reservoir development regions; 4, respectively solving the development thickness of the low-resistance reservoir development area and the development thickness of the high-resistance reservoir development area; 5, respectively solving reservoir parameters of the low-resistance reservoir development area and the high-resistance reservoir development area; and 6, calculating the average porosity according to the reservoir parameters and the development thickness to realize quantitative prediction of the beach facies carbonate reservoir. According to the method, the whole beach facies carbonate reservoir is equivalent to a thick impedance body, the complex reservoir is integrally identified through relative impedance change, and identification can be carried out through channel integral data waveform change as long as the impedance body and the surrounding rock have impedance difference; the technical problem that accurate reservoir prediction of an existing complex beach facies carbonate reservoir cannot be achieved is solved.

Description

Technical Field

[0001] The invention belongs to the technical field of seismic exploration reservoir identification, and in particular relates to a quantitative prediction method for beach-phase carbonate reservoirs based on statistics, which is specifically used for quantitative prediction of beach-phase carbonate reservoirs. Background Art

[0002] As oilfield exploration and development continues to deepen and become more difficult, due to the complexity of carbonate diagenesis, the beach body has a strong degree of local cloudification, or local karst or faults are more developed, resulting in complex reservoir lithology and physical properties and strong heterogeneity. It is necessary to accurately identify the degree of reservoir development so that oilfield exploration and development can avoid risks and further tap potential.

[0003] As we all know, beach-facies carbonate reservoirs are formed when a relatively stable and thin beach-facies stratum is developed inside the carbonate sedimentary background. After being transformed by one or more factors such as dissolution and hydrothermal fluid, complex carbonate reservoirs are formed. The reservoirs are mixed with dolomite and limestone, and caves, pores, and fractures together constitute a complex reservoir space, and the reservoir properties in different regions vary greatly. This type of reservoir is very difficult to predict because the reservoir is thin, the pre-stack and post-stack reservoir response characteristics are not obvious, and the target layer is thin. The sidelobe effect of the lithology interface of the overlying and underlying strata will also affect the reservoir identification. In addition, the vertical reservoir is often mixed with high and low resistance, and the lateral heterogeneity is strong, which makes seismic identification very difficult. Therefore, beach-facies carbonate reservoirs are typical complex carbonate reservoirs that are difficult to identify.

[0004] The current quantitative reservoir prediction mainly adopts various inversion technologies, including:

[0005] Phase-controlled inversion: This technology adds phase control to conventional inversion, and an additional constraint condition can further improve the prediction accuracy. However, this technology is relatively easy to use for reservoirs of single origin, but it still has limitations for beach-facies carbonate reservoirs of complex origin.

[0006] Waveform indication inversion: This technology makes full use of the waveform variation law of seismic data, but the specific inversion effect depends on the calibration of the well. For some areas with few wells and complex geological conditions, the application effect is also limited;

[0007] Geostatistical inversion, this technology makes full use of various elastic parameters and geological information, but this method has high requirements for newly added control parameters. It requires as many accurate geological parameters as possible, and some parameters are difficult to obtain. In the actual production process, it is often impossible to obtain enough and accurate data, which still leads to the inability to accurately predict beach-phase carbonate reservoirs. At the same time, this method has a very large amount of calculations and a slow calculation speed, resulting in low prediction efficiency. Summary of the invention

[0008] The purpose of the present invention is to overcome the above-mentioned problems existing in the prior art, and to provide a quantitative prediction method for beach-phase carbonate reservoirs based on statistics. The present invention establishes a reservoir response model of channel integral data, and treats the complex beach-phase carbonate reservoir as a whole as a thicker impedance body. The complex reservoir is identified as a whole through relative impedance changes. As long as the impedance body has an impedance difference with the surrounding rock, macroscopic identification can be performed through the waveform change of channel integral data. In the prediction method of the present invention, the channel integral data is simple, fast and easy to obtain, and has the advantages of waveform indication and phase-controlled reservoir prediction. It can fully integrate geological information into reservoir prediction on the basis of making full use of seismic data to achieve accurate prediction of reservoirs, and effectively solve the technical problem that the existing complex beach-phase carbonate reservoirs cannot achieve accurate reservoir prediction.

[0009] To achieve the above purpose, the technical solution adopted by the present invention is as follows:

[0010] A quantitative prediction method for beach-facies carbonate reservoirs based on statistics, characterized by comprising the following steps:

[0011] Step 1: Describe the pre-depositional paleo-geomorphology based on the track integral data of the study area, and identify favorable phase zones for reservoir development based on the pre-depositional paleo-geomorphology;

[0012] Step 2: Establish reservoir seismic response model through fine calibration of well seismic data in the study area;

[0013] Step 3: Classify the waveform of favorable phase zones for reservoir development according to the reservoir seismic response mode, and qualitatively identify the low-resistivity reservoir development zone and the high-resistivity reservoir development zone in the reservoir development zone;

[0014] Step 4: Calculate the development thickness of the low-resistance reservoir development area and the high-resistance reservoir development area respectively according to statistical rules;

[0015] Step 5: Obtain the reservoir parameters of the low-resistance reservoir development area and the high-resistance reservoir development area respectively according to statistical laws;

[0016] Step 6: Calculate the average porosity based on reservoir parameters and development thickness to achieve quantitative prediction of beach-facies carbonate reservoirs.

[0017] In step 1, the specific method for depicting the paleo-geomorphology map before deposition is as follows: first, perform trace integral calculation based on the conventional seismic data of the study area to obtain trace integral data; then, identify the interface between the upper high-speed stratum and the lower mudstone stratum of the target layer on the trace integral data, and depict the thickness of the mudstone stratum to obtain the paleo-geomorphology map before deposition of the study area.

[0018] In step 1, conventional seismic data refers to prestack time migration seismic data.

[0019] In step 2, the reservoir seismic response mode established is: the low-resistance reservoir shows a double-peak reflection response mode in the trace integral data, the high-resistance reservoir shows a strong complex wave reflection response mode in the trace integral data, and the non-reservoir shows a uniaxial strong peak reflection response mode.

[0020] In step 3, the waveform classification refers to: using the channel integral data, taking the top and bottom positions of the target layer as the upper and lower interfaces of the time window respectively, using the unsupervised waveform clustering analysis method based on the adaptive neural network, dividing the waveforms with similar seismic data in the favorable phase belt of reservoir development into one category, and extracting the typical waveform of each category, comparing the typical waveform with the corresponding response waveform in different reservoir response modes, identifying the reservoir type corresponding to each typical waveform, and finally identifying the high-resistance reservoir development area, low-resistance reservoir development area and no reservoir area.

[0021] In step 4, the method for calculating the development thickness is:

[0022] S1: Track the maximum amplitude position of the high-speed formation in the overlying strata in the reservoir development area on the trace integral data as the reference layer, and calculate the time thickness between the reference layer and the top interface of the mud layer;

[0023] S2: The actual reservoir thickness of all wells drilled in the study area is analyzed with the time thickness, and the linear relationship between the actual reservoir thickness and time thickness of different types of reservoirs is statistically obtained;

[0024] S3: The calculation formula of development thickness is obtained based on the statistical linear relationship;

[0025] S4: According to the formula obtained in S3, the development thickness of the low-resistivity reservoir development area and the high-resistivity reservoir development area are calculated respectively.

[0026] In step S3, the calculation formula for the thickness of the low-resistivity reservoir development zone is:

[0027] H 发 =f 低阻 (h)

[0028] The calculation formula for the thickness of the high-resistance reservoir development zone is:

[0029] H 发 =f 高阻 (h)

[0030] Where h is the time thickness, H 发 For development thickness.

[0031] In step 5, the method for obtaining reservoir parameters is:

[0032] S11: Using the trace integral data and seismic attribute analysis, the maximum amplitude value within a 10ms window upward from the top of the mudstone formation is extracted;

[0033] S22: Calculating actual reservoir parameters of the well according to the actual reservoir thickness and average porosity of the well;

[0034] S33: Perform intersection analysis on the actual reservoir parameters of the wells that have been drilled and the extracted maximum amplitude value, statistically obtain the linear relationship between the actual reservoir parameters and the maximum amplitude value, and then obtain the calculation formula of the reservoir parameters of the study area based on the statistical linear relationship;

[0035] S44: According to the formula obtained in S33, the reservoir parameters of the low-resistance reservoir development area and the high-resistance reservoir development area are respectively calculated.

[0036] In step S33, the linear relationship between the actual reservoir parameters of the drilled well and the extracted maximum amplitude value is:

[0037] When the underlying mud layer thickness of the reservoir is less than the set value,

[0038] a 实 =g 高地 (x);

[0039] When the underlying mud layer of the reservoir has a time thickness greater than the set value,

[0040] a 实 =g 斜坡 (x);

[0041] According to the statistical linear relationship, the calculation formula of reservoir parameters in the study area is:

[0042] When the underlying mud layer thickness of the reservoir is less than the set value,

[0043] a 研究区 =g 高地 (x)

[0044] When the underlying mud layer of the reservoir has a time thickness greater than the set value,

[0045] a 研究区 =g 斜坡 (x)

[0046] In the formula, a 实 is the actual reservoir parameter of the well, a 研究区 are the reservoir parameters of the low-resistivity reservoir development area and the high-resistivity reservoir development area to be obtained in the study area, and x is the maximum peak amplitude value below the top boundary of the target layer.

[0047] In step 6, the average porosity is calculated as:

[0048] Por=a研究区 / H 发

[0049] Where Por is the average porosity of the reservoir.

[0050] By adopting the above technical solution, the beneficial technical effects of the present invention are:

[0051] The method of the present invention is specifically used for quantitative prediction of beach-phase carbonate reservoirs, and specifically includes six steps, wherein:

[0052] Step 1 uses Dow integral data to depict the pre-deposition paleo-geomorphology and identify favorable phase zones for reservoir development. Since Dow integral data is very sensitive to changes in lithology, it can more accurately depict lithology interfaces that are difficult to identify on conventional seismic data, thus improving the accuracy of paleo-geomorphology depiction. This step achieves a basic phase-controlled identification of reservoir development areas through the depiction of paleo-geomorphology, providing an accurate basis for subsequent quantitative prediction.

[0053] Step 2 establishes the reservoir seismic response model through fine calibration of well seismic data. It should be noted that the difficulty in identifying complex carbonate reservoirs lies in that, first of all, the reservoir is composed of multiple layers of high-resistance and low-resistance reservoirs, and it is difficult to form an effective identification and measurement of reservoir response on conventional seismic data, so that conventional technology is difficult to accurately predict the reservoir; at the same time, the reservoir changes rapidly laterally, and the lithology and physical properties between different wells are quite different. The prediction using various inversion methods can often only achieve the prediction of part of the reservoir. This method treats the complex reservoir as a whole as a thicker impedance body, and identifies the complex reservoir as a whole through relative impedance changes. Whether it is a high-resistance or low-resistance reservoir, as long as the impedance body has an impedance difference with the surrounding rock, it can be macroscopically identified through the waveform changes of the integrated data. Therefore, by establishing the reservoir response model of the integrated data, the difficulty of reservoir identification can be effectively reduced.

[0054] Step 3: Based on the favorable phase belt predicted in step 1, the reservoir response model established in step 2 is used to further identify the reservoir more accurately through waveform classification, and the reservoir type (high resistivity, low resistivity) can be divided to provide a basis for subsequent quantitative prediction.

[0055] Furthermore, this step adopts an unsupervised waveform clustering analysis method based on an adaptive neural network, which can avoid interference from human factors, ensure that each type of waveform has good similarity, and the typical waveform is more representative. Only in this way can we ensure that this type of waveform conforms to the reservoir response pattern after comparison.

[0056] Step 4 uses statistical rules to obtain the thickness of low-resistance reservoir development areas and high-resistance reservoir development areas respectively. Since the reservoir is often mixed with high-resistance and low-resistance in the vertical direction, it is very complex and difficult to obtain the thickness of the reservoir. Therefore, the complex reservoir is equivalent to an impedance body. According to the channel integral data, the maximum envelope thickness of the impedance change inside the target layer is extracted, and the correlation formula of different types of reservoirs is obtained by intersection analysis with the known reservoir thickness of the well, which is conducive to the accurate calculation of the reservoir thickness change.

[0057] Step 5 introduces reservoir parameters according to statistical laws. The difficulty of complex reservoir prediction lies in the large changes in lithology and physical properties. It is difficult to establish an effective relationship between the geophysical parameters of the reservoir and the physical properties of the reservoir. This step introduces geological information. According to the differences in reservoir genesis under different landforms, the reservoirs are artificially divided into two categories according to the changes in landforms, so as to achieve the linkage between reservoir parameters and seismic reflection, thereby achieving accurate prediction of the reservoir.

[0058] In summary, the present invention provides a practical and effective solution to the problem of complex reservoir prediction of beach-phase carbonate. In addition, the prediction method uses simple, fast and easy-to-obtain channel integral data, and has the advantages of waveform indication and phase-controlled reservoir prediction. It can fully integrate geological information into reservoir prediction based on full use of seismic data to achieve accurate reservoir prediction, thus effectively solving the technical problem that the existing complex beach-phase carbonate reservoir cannot achieve accurate reservoir prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 It is a flowchart of the present invention;

[0060] Figure 2 This is a schematic diagram of flattening the top boundary of the target layer in a conventional seismic well-connected section;

[0061] Figure 3 This is a schematic diagram of flattening the top boundary of the target layer in the well section of the integrated data;

[0062] Figure 4 The ancient landform map before the deposition is depicted;

[0063] Figure 5 Plan view of the reservoir development area identified qualitatively;

[0064] Figure 6 It is a plane diagram of the time thickness between the reference layer and the top interface of the mud layer;

[0065] Figure 7 It is the plane diagram of the maximum peak amplitude attribute (amplitude x) inside the channel integral. DETAILED DESCRIPTION

[0066] Example 1

[0067] This embodiment provides a quantitative prediction method for beach-facies carbonate reservoirs based on statistics, such as Figure 1 As shown, it includes the following steps:

[0068] Step 1: Describe the pre-depositional paleomorphology based on the track integral data of the study area, and identify favorable phase zones for reservoir development based on the pre-depositional paleomorphology.

[0069] Among them, the development of beach-facies carbonate reservoirs is controlled by the degree of beach development. The areas with higher paleo-geomorphology before deposition are favorable areas for beach development. The thinner the thickness of mudstone strata at the bottom of the target layer, the higher the paleo-geomorphology before deposition. Therefore, the specific method for depicting the paleo-geomorphology map before deposition is:

[0070] First, obtain conventional seismic data in the study area, mainly pre-stack time migration seismic data in the study area. Then, perform channel integral calculation based on the conventional seismic data to obtain channel integral data. Then, the interface between the upper high-speed stratum and the lower mudstone stratum of the target layer can be clearly identified on the channel integral data, and the thickness of the mudstone stratum can be depicted, that is, the pre-deposition paleo-geomorphology map of the study area can be accurately obtained.

[0071] Step 2: Based on the trace integral data of the study area, the reservoir seismic response model is established through fine calibration of well seismic data.

[0072] Among them, due to the thin reservoir thickness, complex reservoir lithology and physical properties, and the internal high-resistance and low-resistance mixture, the reservoir response is not obvious in conventional seismic, and the various geophysical parameters have a general degree of differentiation between reservoirs and non-reservoirs. Therefore, the pre-stack and post-stack inversion methods are generally effective and can only identify part of the reservoir. However, the trace integral data represents the relative impedance, and the identification is more intuitive. The high-resistance and low-resistance reservoirs have a certain response in the trace integral data, which can be used as the basis for qualitative identification of reservoirs. The method is simple and effective.

[0073] Based on this, this method divides the reservoir into two categories according to the different impedances of the target layer, namely: low-resistance reservoir - with a clearly identifiable thick low-resistance reservoir section; high-resistance reservoir - the impedance is not significantly reduced compared with the surrounding rock, and the impedance of some layers is higher than that of the surrounding rock. Therefore, the established reservoir seismic response mode is: low-resistance reservoirs show a double-peak reflection response mode in the trace integral data, high-resistance reservoirs show a strong complex wave reflection response mode in the trace integral, and no reservoir shows a uniaxial strong peak reflection response mode.

[0074] Step 3: According to the reservoir seismic response mode, waveform classification is performed on the favorable phase belts for reservoir development, and the low-resistivity reservoir development areas and high-resistivity reservoir development areas in the reservoir development areas are qualitatively identified.

[0075] Among them, the waveform classification refers to: using channel integral data, taking the top and bottom positions of the target layer as the upper and lower interfaces of the time window respectively, using an unsupervised waveform clustering analysis method based on an adaptive neural network, dividing the waveforms with similar seismic data in the favorable phase belt of reservoir development into one category, and extracting the typical waveform of each category, comparing the typical waveform with the corresponding response waveform in different reservoir response modes, identifying the reservoir type corresponding to each typical waveform, and finally identifying the high-resistance reservoir development area, low-resistance reservoir development area and no reservoir area.

[0076] Step 4: Calculate the development thickness of low-resistivity reservoir development area and high-resistivity reservoir development area respectively according to statistical rules.

[0077] Among them, the method for calculating the development thickness is:

[0078] S1: On the trace integral data, the maximum amplitude position of the high-speed formation in the overlying strata of the reservoir development area (the upper axis in the biaxial and complex wave reflection) is tracked as the reference layer, and the time thickness between the reference layer and the top interface of the mud formation is calculated.

[0079] S2: The actual reservoir thickness of all wells drilled in the study area is analyzed with the time thickness. According to statistics, it can be concluded that there is a linear relationship between the actual reservoir thickness and time thickness of different types of reservoirs, as follows:

[0080] H 实 =f 低阻 (h)

[0081] H 实 =f 高阻 (h)

[0082] Where h is the time thickness, H 实 is the actual reservoir thickness of the well that has been drilled.

[0083] S3: According to the linear relationship statistically obtained in step S2, the calculation formula for the development thickness can be obtained, which is as follows:

[0084] The calculation formula for the thickness of low-resistivity reservoir development area is:

[0085] H 发 =f 低阻 (h)

[0086] The calculation formula for the thickness of the high-resistance reservoir development zone is:

[0087] H 发 =f 高阻 (h)

[0088] Where h is the time thickness, H 发 For development thickness.

[0089] S4: According to the formula obtained in S3, the thickness of the low-resistivity reservoir development area and the high-resistivity reservoir development area can be calculated respectively.

[0090] Step 5: Obtain the reservoir parameters of the low-resistivity reservoir development area and the high-resistivity reservoir development area respectively according to statistical laws.

[0091] The method for obtaining reservoir parameters is:

[0092] S11: Using the trace integral data and seismic attribute analysis, the maximum amplitude value within a 10ms window from the top of the mudstone formation is extracted.

[0093] S22: Calculate the actual reservoir parameters of the well according to the product of the actual reservoir thickness of the well and the average porosity.

[0094] S33: The actual reservoir parameters of the wells that have been drilled are intersected and analyzed with the extracted maximum amplitude value. According to statistics, it can be concluded that the actual reservoir parameters and the maximum amplitude value have the following linear relationship:

[0095] When the underlying mud layer thickness of the reservoir is less than the set value,

[0096] a 实 =g 高地 (x);

[0097] When the underlying mud layer of the reservoir has a time thickness greater than the set value,

[0098] a 实 =g 斜坡 (x).

[0099] In the formula, a 实 is the actual reservoir parameter of the well, and x is the maximum peak amplitude below the top boundary of the target layer.

[0100] Then, based on the statistical linear relationship, the calculation formula for the reservoir parameters in the study area can be obtained, as follows:

[0101] When the underlying mud layer thickness of the reservoir is less than the set value,

[0102] a 研究区 =g 高地 (x)

[0103] When the underlying mud layer of the reservoir has a time thickness greater than the set value,

[0104] a 研究区 =g 斜坡 (x)

[0105] In the formula, a 研究区 These are the reservoir parameters of the low-resistivity reservoir development area and the high-resistivity reservoir development area to be obtained in the study area.

[0106] S44: According to the formula obtained in S33, the reservoir parameters of the low-resistance reservoir development area and the high-resistance reservoir development area can be obtained respectively.

[0107] It should be noted that the key to quantitative description of reservoirs is to associate reservoir parameters with geophysical parameters. Drift integral can respond to both high-resistance and low-resistance reservoirs. The fundamental reason is that there is an impedance difference between the reservoir and the surrounding rock, which causes the drift integral to change. Although the reservoir is very complex and it is difficult to obtain a unified law, the reservoir can be simplified and regarded as a group of antibodies as a whole. Therefore, the reservoir parameter a (the product of the average porosity of the reservoir and the reservoir thickness) is introduced. This parameter includes the factors that affect the change of drift integral data - reservoir thickness and reservoir physical properties (porosity). Through actual drilling statistics, it can also be found that this reservoir parameter is directly related to the drift integral amplitude. However, due to the complex carbonate reservoir, the reservoir has undergone multiple transformations and does not follow the same law. However, after introducing geomorphic control factors, the reservoir can be identified as multiple laws. The paleo-geomorphic highland reservoir conforms to one amplitude function, and the paleo-geomorphic slope reservoir conforms to another amplitude function. Calculating the two areas separately can achieve a quantitative description of the reservoir in the study area. This method can be extended to other complex reservoir areas and, under the constraints of known reservoir control factors, can be used to analyze complex reservoirs that do not conform to a unified geophysical law.

[0108] Step 6: Calculate the average porosity based on reservoir parameters and development thickness to achieve quantitative prediction of beach-facies carbonate reservoirs.

[0109] The average porosity is calculated as follows:

[0110] Por=a 研究区 / H 发

[0111] Where Por is the average porosity of the reservoir.

[0112] In summary, the present invention is equivalent to treating the complex beach-phase carbonate reservoir as a thicker impedance body, and identifying the complex reservoir as a whole through relative impedance changes. As long as there is an impedance difference between the impedance body and the surrounding rock, macroscopic identification can be performed through changes in the waveform of the trace integral data. In addition, in the prediction method of the present invention, the trace integral data is simple, fast and easy to obtain, and has the advantages of waveform indication and phase-controlled reservoir prediction. It can fully integrate geological information into reservoir prediction on the basis of making full use of seismic data to achieve accurate prediction of the reservoir, thereby effectively solving the technical problem that the existing complex beach-phase carbonate reservoir cannot achieve accurate reservoir prediction.

[0113] Example 2

[0114] In this embodiment, a research area in which the target layer is a set of limestone-based strata was selected to verify the method described in Example 1. Figure 2 This is a schematic diagram of the top boundary of the target layer flattened by the conventional seismic well-connected profile in the study area. It can be seen from the figure that the target layer in the study area forms a strong reflection with the overlying mud shale formation, the bottom of the target layer is the mud limestone formation, and it forms a crest reflection with the underlying limestone formation. The reservoir is developed inside the high-speed limestone formation above the target layer. There is no obvious correlation between the internal reflection of the target layer and the reservoir development, so it is difficult to identify the reservoir response in this study area through conventional seismic.

[0115] Step 1: Describe the pre-depositional paleo-geomorphology based on the channel integral data of the study area.

[0116] Figure 3 It is a schematic diagram of the top boundary of the target layer flattened by the track integral data and the well profile. Since the track integral data is very sensitive to the change of lithology, the lithology interface can be accurately identified through the track integral data.

[0117] Figure 4 This is a depiction of the pre-depositional paleo-geomorphology. From the map, we can see that mudstone strata with a time thickness greater than 30 ms are paleo-geomorphological depressions, those with a time thickness between 25 and 30 ms are slopes, and those with a time thickness less than 25 ms are paleo-geomorphological highlands.

[0118] Step 2: Based on the trace integral data of the study area, establish the reservoir seismic response model through fine calibration of well seismic data. Figure 3 It can be seen that the low-resistivity reservoir shows a biaxial feature in the trace integral data, the high-resistivity reservoir shows a strong complex wave feature in the trace integral data, and the non-resistivity reservoir shows a uniaxial strong wave peak reflection feature.

[0119] Step 3: According to the reservoir seismic response mode, waveform classification is performed on the favorable phase belts for reservoir development. The change of waveform can distinguish the type of reservoir. The paleo-geomorphological highlands and slopes are comprehensively considered as favorable phase belts for reservoir development (mudstone formation time thickness is less than 30ms), and the reservoir development area is obtained. Figure 5 This is a plan view of the qualitatively identified reservoir development area, in which double-peak reflection is the low-resistivity reservoir development area, strong complex wave reflection is the high-resistivity reservoir development area, and uniaxial strong peak reflection is the absence of reservoir, specifically including underdeveloped reservoir area and undeveloped reservoir area.

[0120] Step 4: Calculate the development thickness of low-resistivity reservoir development area and high-resistivity reservoir development area respectively according to statistical rules.

[0121] Among them, the reservoir thickness in the study area is controlled by the degree of beach development. The sedimentary thickness increases when the reservoir develops. However, since the top interface of the target layer has been eroded, the total thickness of the high-speed formation cannot reflect the change in sedimentary thickness. Since the reservoir development section is located at the bottom of the high-speed layer, it is not affected by erosion. Regardless of whether it is a high-resistance or low-resistance reservoir, there is an impedance difference with the overlying surrounding rock as a whole.

[0122] Therefore, the maximum amplitude position of the high-speed formation in the overlying strata in the reservoir development area is tracked on the integrated data (the upper axis in the dual-axis and complex wave reflection) as the reference layer, and the time thickness between the reference layer and the top interface of the mud layer is obtained. Figure 6 It is a plane diagram of the time thickness between the reference layer and the top interface of the mud formation. Among them, the reference layer can be regarded as the stratigraphic center of the high-speed surrounding rock overlying the Maokou Formation reservoir. Its thickness relationship with the bottom boundary of the high-speed stratum, that is, the top boundary of the mud formation, can approximately reflect the changes in the deposition of the reservoir section. However, due to the differences in internal impedance of different types of reservoirs, the response on the track integral also changes, resulting in changes in the position relationship represented by the reference layer. Therefore, the thickness relationship needs to be calculated by reservoir type. According to the waveform classification results, the two types of reservoirs are statistically analyzed and the rules are obtained:

[0123] The calculation formula for the thickness of low-resistivity reservoir development area is:

[0124] H 发 =f 低阻 (h)

[0125] The calculation formula for the thickness of the high-resistance reservoir development zone is:

[0126] H 发 =f 高阻 (h)

[0127] Where h is the time thickness, H 发 For development thickness.

[0128] Furthermore, this embodiment selects multiple wells to calculate the development thickness and verify the error, as shown in Table 1 below:

[0129] Table 1 Thickness calculation and error statistics of the invention

[0130]

[0131] Due to the limited drilling, 3 statistical wells and 1 verification well were used for low-impedance reservoirs, and the error between the calculated results and the logging interpretation of reservoir thickness was less than 1.1m. 4 statistical wells and 1 verification well were used for high-impedance reservoirs, and the error between the calculated results and the logging interpretation of thickness was less than 1.2m. It can be seen that this method can accurately calculate the development thickness of low-resistance reservoir development areas and high-resistance reservoir development areas.

[0132] Step 5: Describe the reservoir development in the study area through reservoir parameters. Figure 7 It is the plane diagram of the maximum peak amplitude attribute (amplitude x) inside the channel integral. Through statistics of the drilled wells, it is found that the higher the reservoir development degree, the greater the maximum peak amplitude value (lower axis of the dual axis of the channel integral data, complex wave amplitude) 30-40ms below the top boundary of the target layer (such as Figure 7 As shown in the figure, the wells located in the paleo-geomorphological highland (the underlying mudstone layer has a time thickness of less than 25ms), the reservoirs in this area are located in the high part of the sedimentary geomorphology, the beach body is highly developed, the degree of cloudification is high, and the contribution of the porous reservoir is greater. The reservoir parameters and amplitude values ​​conform to the functional relationship:

[0133] When the underlying mud layer thickness of the reservoir is less than the set value,

[0134] a 研究区 =g 高地 (x)

[0135] For wells located in the slope area around the highlands (where the thickness of the mudstone formation is greater than 25ms and less than 30ms), karstification contributes more to the formation of the reservoir, and its reservoir parameters and amplitude values ​​conform to another functional relationship:

[0136] When the underlying mud layer of the reservoir has a time thickness greater than the set value,

[0137] a 研究区 =g 斜坡 (x)

[0138] In the formula, a 研究区 are the reservoir parameters of the low-resistivity reservoir development area and the high-resistivity reservoir development area to be obtained in the study area, and x is the maximum peak amplitude value 30 to 40 ms below the top boundary of the target layer.

[0139] Furthermore, this embodiment selects multiple wells to perform reservoir parameter calculation and error verification, as shown in Table 2 below:

[0140] Table 2 Coefficient calculation results and error statistics

[0141]

[0142] Due to the limited drilling, there are 3 statistical regular wells and 3 verification wells in the highland location, and the error of the calculated parameters between the calculation results and the logging interpretation results is less than 7.9; there are 3 statistical wells in the slope location, and no verification wells, and the error of the calculated parameters between the calculation results and the logging interpretation calculation parameters is less than 0.42. It can be seen that this method can accurately calculate the reservoir parameters of the low-resistance reservoir development area and the high-resistance reservoir development area.

[0143] Step 6: The reservoir parameters and development thickness have been obtained, and the average porosity can be obtained according to the calculation. The details are shown in Table 3 below:

[0144] Table 3 Porosity calculation results and error statistics

[0145]

[0146] It can be seen from the above table that the carbonate reservoir in the study area is located more than 5,000 meters underground. The error of seismic prediction of reservoir thickness is less than 1.2m, and the error between the calculated average porosity of the reservoir in the study area and the average porosity interpreted by well logging is less than 0.33%. The accuracy fully meets the requirements of quantitative prediction of seismic reservoirs. As the number of drilling increases, the understanding of the rules becomes clearer, and the prediction results will be further improved.

[0147] The above description is only a specific implementation mode of the present invention. Any feature disclosed in this specification, unless otherwise stated, can be replaced by other alternative features that are equivalent or have similar purposes; all the disclosed features, or all the steps in the methods or processes, except for mutually exclusive features and / or steps, can be combined in any way.

Claims

1. A quantitative prediction method for beach-facies carbonate reservoirs based on statistics, characterized in that: The following steps are involved: Step 1: Describe the pre-depositional paleo-geomorphology based on the track integral data of the study area, and identify favorable phase zones for reservoir development based on the pre-depositional paleo-geomorphology; Step 2: Establish reservoir seismic response model through fine calibration of well seismic data in the study area; Step 3: Classify the waveform of favorable phase zones for reservoir development according to the reservoir seismic response mode, and qualitatively identify the low-resistivity reservoir development zone and the high-resistivity reservoir development zone in the reservoir development zone; Step 4: Calculate the development thickness of the low-resistance reservoir development area and the high-resistance reservoir development area respectively according to statistical rules; Step 5: Obtain the reservoir parameters of the low-resistance reservoir development area and the high-resistance reservoir development area respectively according to statistical laws; Step 6: Calculate the average porosity based on reservoir parameters and development thickness to achieve quantitative prediction of beach-facies carbonate reservoirs.

2. The method for quantitative prediction of beach-facies carbonate reservoirs based on statistics according to claim 1, characterized in that: In step 1, the specific method for depicting the paleo-geomorphology map before deposition is as follows: first, perform trace integral calculation based on the conventional seismic data of the study area to obtain trace integral data; then, identify the interface between the upper high-speed stratum and the lower mudstone stratum of the target layer on the trace integral data, and depict the thickness of the mudstone stratum to obtain the paleo-geomorphology map before deposition of the study area.

3. The method for quantitative prediction of beach-facies carbonate reservoirs based on statistics according to claim 2, characterized in that: In step 1, conventional seismic data refers to prestack time migration seismic data.

4. The method for quantitative prediction of beach-facies carbonate reservoirs based on statistics according to claim 1, characterized in that: In step 2, the reservoir seismic response mode established is: the low-resistance reservoir shows a double-peak reflection response mode in the trace integral data, the high-resistance reservoir shows a strong complex wave reflection response mode in the trace integral data, and the non-reservoir shows a uniaxial strong peak reflection response mode.

5. The method for quantitative prediction of beach-facies carbonate reservoirs based on statistics according to claim 1, characterized in that: In step 3, the waveform classification refers to: using the channel integral data, taking the top and bottom positions of the target layer as the upper and lower interfaces of the time window respectively, using the unsupervised waveform clustering analysis method based on the adaptive neural network, dividing the waveforms with similar seismic data in the favorable phase belt of reservoir development into one category, and extracting the typical waveform of each category, comparing the typical waveform with the corresponding response waveform in different reservoir response modes, identifying the reservoir type corresponding to each typical waveform, and finally identifying the high-resistance reservoir development area, low-resistance reservoir development area and no reservoir area.

6. A statistically based quantitative prediction method for beach-facies carbonate reservoirs according to any one of claims 1 to 5, characterized in that: In step 4, the method for calculating the development thickness is: S1: Track the maximum amplitude position of the high-speed formation in the overlying strata in the reservoir development area on the trace integral data as the reference layer, and calculate the time thickness between the reference layer and the top interface of the mud layer; S2: The actual reservoir thickness of all wells drilled in the study area is analyzed with the time thickness, and the linear relationship between the actual reservoir thickness and time thickness of different types of reservoirs is statistically obtained; S3: The calculation formula of development thickness is obtained based on the statistical linear relationship; S4: According to the formula obtained in S3, the development thickness of the low-resistivity reservoir development area and the high-resistivity reservoir development area are calculated respectively.

7. The method for quantitative prediction of beach-facies carbonate reservoirs based on statistics according to claim 6, characterized in that: In step S3, the calculation formula for the thickness of the low-resistivity reservoir development zone is: H 发 =f 低阻 (h) The calculation formula for the thickness of the high-resistance reservoir development zone is: H 发 =f 高阻 (h) Where h is the time thickness, H 发 For development thickness.

8. The method for quantitative prediction of beach-facies carbonate reservoirs based on statistics according to claim 7, characterized in that: In step 5, the method for obtaining reservoir parameters is: S11: Using the trace integral data and seismic attribute analysis, the maximum amplitude value within a 10ms window upward from the top of the mudstone formation is extracted; S22: Calculating actual reservoir parameters of the well according to the actual reservoir thickness and average porosity of the well; S33: Perform intersection analysis on the actual reservoir parameters of the wells that have been drilled and the extracted maximum amplitude value, statistically obtain the linear relationship between the actual reservoir parameters and the maximum amplitude value, and then obtain the calculation formula of the reservoir parameters of the study area based on the statistical linear relationship; S44: According to the formula obtained in S33, the reservoir parameters of the low-resistance reservoir development area and the high-resistance reservoir development area are respectively calculated.

9. The method for quantitative prediction of beach-facies carbonate reservoirs based on statistics according to claim 8, characterized in that: In step S33, the linear relationship between the actual reservoir parameters of the drilled well and the extracted maximum amplitude value is: When the underlying mud layer thickness of the reservoir is less than the set value, a 实 =g 高地 (x); When the underlying mud layer of the reservoir has a time thickness greater than the set value, a 实 =g 斜坡 (x); According to the statistical linear relationship, the calculation formula of reservoir parameters in the study area is: When the underlying mud layer thickness of the reservoir is less than the set value, a 研究区 =g 高地 (x) When the underlying mud layer of the reservoir has a time thickness greater than the set value, a 研究区 =g 斜坡 (x) In the formula, a 实 is the actual reservoir parameter of the well, a 研究区 are the reservoir parameters of the low-resistivity reservoir development area and the high-resistivity reservoir development area to be obtained in the study area, and x is the maximum peak amplitude value below the top boundary of the target layer.

10. A quantitative prediction method for beach-facies carbonate reservoirs based on statistics according to claim 9, characterized in that: In step 6, the average porosity is calculated as: For=a 研究区 / H 发 Where Por is the average porosity of the reservoir.

Images