A reservoir inversion method based on optimization of well-seismic characteristic parameters

Abstract

The present invention relates to a reservoir inversion method based on the optimization of well-seismic characteristic parameters. It mainly solves the problem of low reservoir prediction accuracy in the framework model of existing reservoir inversion prediction because the matching of well-seismic information is not fully considered. The method includes: determining the stratigraphic slice set of the target layer; completing the optimization of each sedimentary unit slice and calculating its well-seismic correlation coefficient; clarifying the reservoir thickness that can be reflected by seismic attributes; under its control, merging and combining the upper and lower sedimentary units of the target layer with a sedimentary unit with a higher well-seismic correlation coefficient as the center; calculating the well-seismic correlation coefficients of different combination units; optimizing the combined unit with the highest well-seismic correlation coefficient and higher than the correlation coefficient of the target sedimentary unit; repeating the above steps, optimizing multiple best matching units in the target layer and determining them as the best matching units; based on this, obtaining the framework model; completing the well-seismic combined reservoir inversion prediction in the target layer, thereby realizing the fine prediction of the river channel sand body.

Description

Technical Field

[0001] The present invention relates to the field of oil reservoir geophysics, and in particular to a reservoir inversion method based on optimization of well seismic characteristic parameters. Background Art

[0002] Older oilfields in eastern China have entered the development and interpretation phase for high or extra-high water cuts. Due to the long development time, numerous wells, numerous layers, and thin sand bodies, the accuracy requirements for oilfield reservoir description and remaining oil prediction have increased significantly. Typically, based on regional geological patterns, combined seismic data fidelity processing, and with drilled well data as constraints, well-seismic inversion methods are used to conduct detailed reservoir descriptions. This has achieved a representative anatomy of channel sand bodies in different reservoir types. These results have been effectively applied in the formulation of development plans and the detailed exploration of remaining oil potential, demonstrating the significant role of well-seismic combined reservoir inversion prediction technology in detailed reservoir description. However, the current reservoir prediction accuracy is still not in line with actual production needs, seriously restricting the efficient development of oilfields. New methods and technologies are urgently needed to improve reservoir prediction accuracy.

[0003] Stochastic inversion is a commonly used inversion method for fine reservoir prediction in oilfield development. The idea is: under the constraints of the framework model, the initial reservoir model is established by using the well point data through the variation function, and the seismic data between wells is used as a constraint to achieve the output of the prediction results. The reliability of the framework model directly affects the accuracy of reservoir prediction. There are literature reports on the use of well-seismic combined reservoir inversion prediction under the constraints of the framework model, see (1) Wang Baoli, Lin Ying, Zhang Guangzhi, et al. "Seismic random inversion method for characteristic parameters of inhomogeneous media" (Petroleum Geophysical Exploration, Issue 6, 2021), (2) Zhou Shuangshuang, Yin Xingyao, Pei Song, et al. "Application of geostatistical inversion in reservoir prediction - taking P 3 w in WSD area as an example" (Progress in Geophysics, Issue 2, 2018), (3) Xiao Zhangbo, Lei Yongchang, Yu Junqing, et al. "Monte Carlo-Markov chain stochastic inversion method constrained by seismic waveform" (Petroleum Geophysical Exploration, Issue 3, 2021). The above-mentioned document (1) obtains the characteristic parameters of heterogeneous media that accurately characterize the spatial variation characteristics of the formation based on seismic and well logging data, constructs a heterogeneous medium characteristic parameter model, and uses this as the basis for seismic random inversion; the above-mentioned document (2) determines the anisotropy of the reservoir parameters by calculating the variation function in each direction, obtains a reasonable theoretical variation function model and the range parameters in each direction, and thus selects the model that best conforms to the geological sedimentary law of the area to participate in the inversion; the above-mentioned document (3) guides the well data to perform pseudo-ordinary Kriging interpolation simulation based on the similarity characteristics between known seismic waveforms, establishes an initial model with seismic waveform indication, and thus performs multiple random simulations. It can be seen that although the above-mentioned documents all use random inversion, none of them specifically mentions the framework model construction method, that is, they mostly use oil layer group-level seismic layers as units to construct framework models. The inventor found that the framework constructed by the existing technology has a low resolution ability and can only play a constraining role on the structural trend of large sets of layers. The constraining ability on thin interlayers is relatively weak, and the predicted sand bodies are prone to "layer-crossing" phenomenon. Therefore, to meet the need for efficient application of well-seismic combined reservoir prediction in detailed reservoir description, a reservoir inversion method based on the optimization of well-seismic characteristic parameters has been developed. It should be noted that the framework model described above can be understood as a structural model. The difference between the two is that the structural model includes faults and horizons, while the framework model here only includes horizons. Summary of the Invention

[0004] This invention addresses the problem of low reservoir prediction accuracy caused by unclear framework model construction scale in existing reservoir inversion prediction techniques. By providing a reservoir inversion method based on the optimization of well-seismic characteristic parameters, this method comprehensively considers the vertical resolution of well-seismic data and establishes a framework model of appropriate scale, providing a reliable technical foundation for improving the accuracy of well-seismic combined reservoir prediction.

[0005] The present invention solves the problem by the following technical solution: the reservoir inversion method based on the optimization of well seismic characteristic parameters comprises the following steps:

[0006] S1. Identify one or two seismic marker layers near the target interval in the study area, and generate a seismic sedimentology stratigraphic slice set based on the constraints of the identified seismic marker layers;

[0007] S2 based on the seismic sedimentology stratigraphic slice set produced in step S1, extracting seismic attribute data of each slice well bypass;

[0008] S3. Extract reservoir parameters for each sedimentary unit well point within the target interval;

[0009] S4. Select the reservoir parameters of a sedimentary unit and perform intersection analysis with the well-side seismic attribute data of each slice in step S2, and calculate the well-seismic correlation coefficient using the least squares method;

[0010] S5. Select the slice with the highest correlation coefficient within the range of 15 images upward to 15 images downward near the seismic reflection time of the sedimentary unit selected in step S4 as the well-seismic correlation coefficient of the sedimentary unit selected in step S4;

[0011] S6. Repeat steps S4-5 to determine the well-seismic correlation coefficient for each remaining sedimentary unit in the target interval;

[0012] S7. Determine the reservoir thickness that can be reflected by seismic attributes;

[0013] S8. Under the control of the reservoir thickness determined in step S7, a sedimentary unit with a higher well-seismic correlation coefficient calculated in step S6 is used as the central target, and several sedimentary units above and below the central target are combined in several different ways to form multiple combined units;

[0014] S9. Based on the seismic sedimentology stratigraphic slice set produced in step S1, obtain seismic attribute data corresponding to the seismic sedimentology stratigraphic slice set, replace the sedimentary units with composite units using the method of steps S3-5; calculate the well-seismic correlation coefficient of each composite unit;

[0015] S10. Based on the well-seismic correlation coefficients of each combination unit obtained in step S9, the combination unit with the highest well-seismic correlation coefficient that is higher than the correlation coefficient of the central target in step S8 is selected as the best well-seismic matching unit; if the correlation coefficients of all combination units are lower than the correlation coefficient of the central target, the central target is the best matching unit;

[0016] S11. Repeat steps S8-10 above, select the remaining multiple best matching units in the target layer segment, and combine all the best matching units in the target layer segment to obtain a framework model;

[0017] S12. Based on the framework model obtained in S11, complete well-seismic combined reservoir inversion within the target interval.

[0018] Preferably, the reservoir parameters of each sedimentary unit in step S3 include sandstone thickness and sand-to-ground ratio.

[0019] Preferably, step S4 selects the reservoir parameters of a certain sedimentary unit and the seismic attribute data of each slice well bypass in step S2 for intersection analysis, and uses the least squares method to calculate the correlation coefficient thereof, including: the intersection analysis is to use the reservoir parameters of a certain sedimentary unit and the seismic attribute data of each slice well bypass as the dependent variable and independent variable respectively, make a scatter plot in the statistical software, form a trend line, and use the least squares method to calculate the correlation coefficient of the two variables.

[0020] Preferably, the statistical software is SPSS or Excel.

[0021] Preferably, the method of determining the reservoir thickness that can be reflected by the seismic attributes in S7 comprises the following steps:

[0022] Based on the Widess principle, the ability of seismic data to identify sand body thickness is 1 / 4 wavelength;

[0023] The reservoir thickness reflected by seismic attributes is 1 / 4 wavelength;

[0024] Obtain the dominant frequency and reservoir velocity of seismic data;

[0025] Based on the main frequency of the acquired seismic data and the reservoir velocity, the reservoir velocity divided by the main frequency of the seismic data is equal to the seismic wavelength;

[0026] 1 / 4 of the seismic wavelength is the reservoir thickness reflected by the seismic attributes.

[0027] Preferably, the step S8 performs several different combinations of 2-4 deposition units above and below the central target to form multiple combination units.

[0028] Preferably, the method of performing several different combinations of the upper and lower deposition units of the central target in S8 to form multiple combination units includes:

[0029] The central target deposition unit and the immediately preceding deposition unit are combined into a combined unit 1; the central target deposition unit and the immediately preceding two deposition units are combined into a combined unit 2;

[0030] The central target sedimentary unit and the next sedimentary unit are combined as combined unit 3;

[0031] The central target sedimentary unit and the two immediately following sedimentary units are combined as combined unit 4;

[0032] The central target deposition unit and the immediately upper and lower deposition units are combined into a combined unit 5; the central target deposition unit and the immediately upper and lower deposition units are combined into a combined unit 6;

[0033] The central target sedimentary unit and the two sedimentary units immediately above and below are combined as combined unit 7;

[0034] The central target sedimentary unit and the two immediately above and two immediately below sedimentary units are combined as combined unit 8;

[0035] The central target sedimentary unit serves as combined unit 9;

[0036] The maximum formation thickness of the combined unit must be smaller than the reservoir thickness that can be reflected by the seismic event in step S7.

[0037] Preferably, the method for completing well-seismic combined reservoir inversion in the target layer in step S12 is to use the framework model obtained in S11 in combination with existing inversion software to complete well-seismic combined reservoir inversion prediction in the target layer.

[0038] Preferably, the inversion software is seismic waveform indication inversion software SMI, Jason.

[0039] Compared with the above background technology, the present invention has the following beneficial effects:

[0040] The present invention provides a reservoir inversion method based on the optimization of well-seismic characteristic parameters, which solves the problem of low reservoir prediction accuracy caused by unclear scale of framework model construction in reservoir inversion prediction. The traditional reservoir prediction framework model is constructed with oil layer groups as units, which are used as constraints for reservoir prediction. This process can only constrain the structural trends of large layers, and its constraint ability on thin interlayers is relatively weak. Sand bodies are prone to "layer-crossing" connections, resulting in low accuracy of well-seismic combined reservoir prediction. The present invention comprehensively considers the vertical resolution of well-seismic data, establishes a framework model of suitable scale, and provides a reliable technical basis for improving the accuracy of well-seismic combined reservoir prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Attachment Figure 1 It is the original grid model diagram of the present invention;

[0042] Attachment Figure 2 This is a diagram of the present invention's current grid model;

[0043] Attachment Figure 3 is a scatter plot of the correlation coefficient and the number of merged layers according to an embodiment of the present invention;

[0044] Attachment Figure 4 This is a diagram showing the prediction effect of the traditional well-seismic combined reservoir inversion method according to an embodiment of the present invention;

[0045] Attachment Figure 5 This is a diagram showing the prediction effect of a reservoir inversion method based on the optimization of well-seismic characteristic parameters according to an embodiment of the present invention;

[0046] Attachment Figure 6 It is a schematic diagram of forming multiple combination units by performing several different combinations of two deposition units above and below a central target according to an embodiment of the present invention. DETAILED DESCRIPTION

[0047] To make the objectives, technical solutions and advantages of the present invention more clear, the embodiments of the present invention will be described in further detail below with reference to the accompanying drawings.

[0048] like Figure 1 As shown, a reservoir inversion method based on optimization of well-seismic characteristic parameters includes the following steps:

[0049] S1. Identify one or two seismic marker layers near the target interval in the study area, and generate a seismic sedimentology stratigraphic slice set based on the constraints of the identified seismic marker layers;

[0050] S2 based on the seismic sedimentology stratigraphic slice set produced in step S1, extracting seismic attribute data of each slice well bypass;

[0051] S3. Extract reservoir parameters for each sedimentary unit well point within the target layer; reservoir parameters for each sedimentary unit include sandstone thickness and sand-to-ground ratio; the sandstone thickness includes Class II sandstone thickness, Class I sandstone thickness, and effective thickness;

[0052] S4. Select the reservoir parameters of a sedimentary unit and perform intersection analysis with the well-side seismic attribute data of each slice in step S2, and calculate the well-seismic correlation coefficient using the least squares method;

[0053] The specific method of selecting the reservoir parameters of a certain sedimentary unit and performing intersection analysis with the well bypass seismic attribute data of each slice in step S2 and calculating the correlation coefficient thereof using the least squares method includes:

[0054] The intersection analysis is to use the reservoir parameters of a certain sedimentary unit and the seismic attribute data of each slice well bypass as the dependent variable and independent variable respectively, make a scatter plot in statistical software (SPSS, Excel), form a trend line, and use the least squares method to calculate the correlation coefficient of the two variables.

[0055] S5. Select the slice with the highest correlation coefficient within the range of 15 images upward to 15 images downward near the seismic reflection time of the sedimentary unit selected in step S4 as the well-seismic correlation coefficient of the sedimentary unit selected in step S4;

[0056] S6. Repeat steps S4-5 to determine the well-seismic correlation coefficient for each remaining sedimentary unit in the target interval;

[0057] S7. Determine the reservoir thickness that can be reflected by seismic attributes;

[0058] Methods for determining the reservoir thickness that can be reflected by seismic attributes include:

[0059] Based on the Widess principle, the ability of seismic data to identify sand body thickness is 1 / 4 wavelength;

[0060] The reservoir thickness reflected by seismic attributes is 1 / 4 wavelength;

[0061] Obtain the dominant frequency and reservoir velocity of seismic data;

[0062] Based on the main frequency of the acquired seismic data and the reservoir velocity, the reservoir velocity divided by the main frequency of the seismic data is equal to the seismic wavelength;

[0063] 1 / 4 of the seismic wavelength is the reservoir thickness reflected by the seismic attributes.

[0064] S8. Under the control of the reservoir thickness determined in step S7, a sedimentary unit with a higher well-seismic correlation coefficient calculated in step S6 is used as the central target, and several sedimentary units above and below the central target are combined in several different ways to form multiple combined units;

[0065] S9. Based on the seismic sedimentology stratigraphic slice set produced in step S1, obtain seismic attribute data corresponding to the seismic sedimentology stratigraphic slice set, replace the sedimentary units with composite units using the method of steps S3-5; calculate the well-seismic correlation coefficient of each composite unit;

[0066] S10. Based on the well-seismic correlation coefficients of each combination unit obtained in step S9, the combination unit with the highest well-seismic correlation coefficient that is higher than the correlation coefficient of the central target in step S8 is selected as the best well-seismic matching unit; if the correlation coefficients of all combination units are lower than the correlation coefficient of the central target, the central target is the best matching unit;

[0067] S11. Repeat steps S8-10 above, select the remaining multiple best matching units in the target layer segment, and combine all the best matching units in the target layer segment to obtain a framework model;

[0068] S12. Based on the framework model obtained in S11, complete well-seismic combined reservoir inversion within the target interval.

[0069] The method for completing well-seismic combined reservoir inversion in the target interval is to use the framework model obtained in S11 and combine it with existing inversion software to complete well-seismic combined reservoir inversion prediction in the target interval; the inversion software is seismic waveform indication inversion software SMI.

[0070] The following is the 18.5km block of Daqing Changyuan Oilfield A2 , 2310 wells, SⅡ1-SIII10 units are taken as an example to illustrate the implementation process of the method of the present invention.

[0071] Research background and experimental blocks

[0072] In recent years, the reservoir prediction results carried out by Daqing Changyuan Oilfield based on seismic sedimentology have played an important role in guiding the adjustment of development well measures, the preparation of well layout plans and the design of horizontal wells, improving the accuracy of inter-well sand body prediction drawings and demonstrating the great role of reservoir inversion prediction technology based on the combination of well and seismic data.

[0073] Currently, the well-seismic combined reservoir prediction framework is constructed based on oil-bearing formations. This has relatively weak constraints on thin interlayers, making sand bodies prone to "channeling" connections. Therefore, using the SⅡ1-SIII10 units in Block A of the Daqing Changyuan Oilfield as an example, stratigraphic slices were generated, wellside seismic attribute data and sedimentary unit reservoir parameters were extracted, and the best well-seismic matching unit was constructed. Based on this, a reservoir inversion method based on the optimization of well-seismic characteristic parameters was developed to provide technical support for the detailed description of well-seismic combined reservoirs.

[0074] Comparative Example 1

[0075] Traditional well-seismic combined reservoir prediction method:

[0076] This method mainly constructs a framework model based on oil layer groups, and then combines it with existing commercial inversion software to perform reservoir inversion prediction. Figure 1 This is the original grid model diagram; Figure 4 This is the prediction effect diagram of the traditional well-seismic combined with reservoir inversion method.

[0077] Example 1

[0078] The reservoir inversion method based on the optimization of well-seismic characteristic parameters of the present invention is used:

[0079] This method is based on the sequence stratigraphic framework to which the target layer in the study area belongs. The top geological layer of the sequence stratigraphic framework is SⅡ1 (corresponding to the seismic marker layer T1), the bottom geological layer is SIII10 (corresponding to the seismic marker layer T 1-1 ), using software with seismic interpretation module, T1, T 1-1 Seismic marker layer calibration and interpretation (or directly import existing seismic marker layer); obtain 100 stratigraphic slices by proportional subdivision (the number of single-channel seismic data sample points between two seismic marker layers is 70-90, and the number of stratigraphic slices must be equal to or slightly higher than the number of single-channel seismic data sample points); extract the wellside channel of each stratigraphic slice of each well. Amplitude data Att ij , and extract the reservoir parameters R corresponding to each sedimentary unit of each well ik(Sandstone thickness, such as: Class II sandstone thickness, Class I sandstone thickness, effective thickness, etc.), use Att ij , R ik The data were used as dependent and independent variables, and cross-analysis was performed to form a trend line. The parameters R of different sedimentary unit wells were calculated by the least squares method. ik With amplitude data Att ij The correlation degree between them is the well-seismic correlation coefficient. The slice with the highest correlation coefficient within the range of 15 upward and 15 downward images near the seismic reflection time of the K-th sedimentary unit is taken as the well-seismic correlation coefficient of the K-th sedimentary unit. According to the main seismic frequency of the target layer in the study area is 45 Hz, the reservoir velocity is 2700 m / s, and the quarter seismic wavelength is 15 m, it is clear that the reservoir thickness reflected by the seismic attributes is 15 m. Under the control of this reservoir thickness, a sedimentary unit with a higher well-seismic correlation coefficient (such as SⅡ13) is taken as the central target, and two adjacent sedimentary units SⅡ11, SⅡ12, SⅡ13, SⅡ14, and SⅡ15 within 15 m above and below the central target are merged and combined, as shown in the following figure: Figure 6 As shown, SⅡ13~SⅡ12 is combination unit 1; SⅡ13~SⅡ11 is combination unit 2; SⅡ13~SⅡ14 is combination unit 3; SⅡ13~SⅡ15 is combination unit 4; SⅡ12~SⅡ14 is combination unit 5; SⅡ11~SⅡ14 is combination unit 6; SⅡ12~SⅡ15 is combination unit 7; SⅡ11~SⅡ15 is combination unit 8; SⅡ13 is combination unit 9; and the reservoir parameters in the combination units are extracted; Figure 3 The scatter plot of the well-seismic correlation coefficient and the number of merged layers shows that when the number of merged layers reaches 3 and SⅡ12 to SⅡ14 are combined as a unit, the well-seismic correlation coefficient reaches an inflection point and is the highest. This combined unit is determined to be the best matching unit. When the well-seismic correlation coefficients of the combined units are all smaller than the well-seismic correlation coefficient of the central target, the central target is the best matching unit. Figure 2 For the grid model diagram of the present invention, repeat the above steps to determine multiple other best matching units in the target layer (see Figure 2 ), and then determine the grid model; apply the above grid model, and then combine it with existing software (such as seismic waveform indication inversion software SMI, Jason, etc.) to perform reservoir inversion prediction. Compared with the conventional inversion method (see Figure 4 ), the patented method has a higher vertical resolution for sand body prediction, and the accuracy of reservoir prediction combined with well-seismic analysis in dense well network areas is increased to 85%, and can invert the inter-well changes of river channel sand bodies (see Figure 5 ).

[0080] It can be seen from the above Example 1 and Comparative Example 1 that the traditional well-seismic combined reservoir inversion method, in which the well-seismic combined reservoir inversion prediction framework model is constructed with oil layer groups as units, can only play a constraining role on the structural trend of large layer sections, and the constraining ability on thin interlayers is relatively weak. The sand body is prone to the phenomenon of "layer-crossing" connection, resulting in low accuracy of well-seismic combined reservoir prediction.

[0081] The present invention is based on a reservoir inversion method optimized for well-seismic characteristic parameters. According to this method, the vertical resolution capability of well-seismic data is comprehensively considered, and a scale-appropriate framework model is established, which can improve the accuracy of well-seismic reservoir prediction.

[0082] The above examples specifically illustrate the entire process of the reservoir inversion method based on the optimization of well-seismic characteristic parameters of the present invention. The experimental results are accurate and reliable, and can be used for seismic reservoir prediction in the Daqing Changyuan Oilfield development. It can also be used for reservoir evaluation and development blocks with similar seismic and geological characteristics based on seismic attribute prediction. The present invention has the following characteristics:

[0083] 1) A reservoir inversion method based on the optimization of well-seismic characteristic parameters is proposed. The method mainly involves identifying one or two seismic marker layers near the target interval in the study area, and producing a set of seismic sedimentary stratigraphic slices under their constraints. The slices of each sedimentary unit in the target interval are optimized, and the well-seismic correlation coefficient is calculated. Based on the actual elastic parameters of sandstone and mudstone in the study area, the reservoir thickness that can be reflected by the seismic phase axis is determined through forward simulation. Under the control of this reservoir thickness, the sedimentary units above and below the target interval are merged and combined with a sedimentary unit with a high well-seismic correlation coefficient as the center, and the reservoir parameters of different combined units are extracted. Based on the relationship between seismic attributes and reservoir parameters, the well-seismic correlation coefficients of different combined units are quickly and quantitatively calculated. The combined merged unit with the highest well-seismic correlation coefficient that is higher than the correlation coefficient of the target sedimentary unit is selected. The above steps are repeated to optimize multiple best-matching units in the target interval and determine them as the best-matching unit. Based on this, a framework model is obtained. The framework model is used to complete the well-seismic combined reservoir inversion prediction in the target interval, thereby realizing the fine prediction of river channel sand bodies.

[0084] 2) This method was used to improve the accuracy of combined well-seismic reservoir prediction to 85% in dense well-patterned areas within the SⅡ1-SIII10 interval of Block A in the Changyuan Oilfield, Daqing. The results are accurate and reliable, providing a reliable technical foundation for the efficient application of combined well-seismic reservoir prediction technology in detailed reservoir description.

Claims

1. A reservoir inversion method based on optimization of well-seismic characteristic parameters, characterized by: The following steps are involved: S1. Identify one or two seismic marker layers near the target interval in the study area, and generate a seismic sedimentology stratigraphic slice set based on the constraints of the identified seismic marker layers; S2 based on the seismic sedimentology stratigraphic slice set produced in step S1, extracting seismic attribute data of each slice well bypass; S3. Extract reservoir parameters for each sedimentary unit well point within the target interval; S4. Select the reservoir parameters of a sedimentary unit and perform intersection analysis with the well-side seismic attribute data of each slice in step S2, and calculate the well-seismic correlation coefficient using the least squares method; S5. Select the slice with the highest correlation coefficient within the range of 15 images upward to 15 images downward near the seismic reflection time of the sedimentary unit selected in step S4 as the well-seismic correlation coefficient of the sedimentary unit selected in step S4; S6. Repeat steps S4-5 to determine the well-seismic correlation coefficient for each remaining sedimentary unit in the target interval; S7. Determine the reservoir thickness that can be reflected by seismic attributes; S8. Under the control of the reservoir thickness determined in step S7, a sedimentary unit with a higher well-seismic correlation coefficient calculated in step S6 is used as the central target, and several sedimentary units above and below the central target are combined in several different ways to form multiple combined units; S9. Based on the seismic sedimentology stratigraphic slice set produced in step S1, obtain seismic attribute data corresponding to the seismic sedimentology stratigraphic slice set, replace the sedimentary units with composite units using the method of steps S3-5; calculate the well-seismic correlation coefficient of each composite unit; S10. Based on the well-seismic correlation coefficients of each combination unit obtained in step S9, the combination unit with the highest well-seismic correlation coefficient that is higher than the correlation coefficient of the central target in step S8 is selected as the best well-seismic matching unit; if the correlation coefficients of all combination units are lower than the correlation coefficient of the central target, the central target is the best matching unit; S11. Repeat steps S8-10 above, select the remaining multiple best matching units in the target layer segment, and combine all the best matching units in the target layer segment to obtain a framework model; S12. Based on the framework model obtained in S11, complete well-seismic combined reservoir inversion within the target interval; The S8 performs several different combinations of 2-4 deposition units above and below the central target to form multiple combination units.

2. The reservoir inversion method based on well-seismic characteristic parameter optimization according to claim 1, characterized in that: In step S3, the reservoir parameters of each sedimentary unit include sandstone thickness and sand-to-ground ratio.

3. The reservoir inversion method based on well-seismic characteristic parameter optimization according to claim 1, characterized in that: Step S4 selects the reservoir parameters of a certain sedimentary unit and the seismic attribute data of each slice well bypass in step S2 for intersection analysis, and uses the least squares method to calculate the correlation coefficient thereof, including: the intersection analysis is to use the reservoir parameters of a certain sedimentary unit and the seismic attribute data of each slice well bypass as the dependent variable and the independent variable respectively, make a scatter plot in the statistical software, form a trend line, and use the least squares method to calculate the correlation coefficient of the two variables.

4. The reservoir inversion method based on well-seismic characteristic parameter optimization according to claim 3 is characterized in that: The statistical software is SPSS and Excel.

5. The reservoir inversion method based on optimization of well-seismic characteristic parameters according to claim 1, characterized in that: The method of determining the reservoir thickness that can be reflected by seismic attributes in S7 includes: Obtain the dominant frequency and reservoir velocity of seismic data; Based on the main frequency of the acquired seismic data and the reservoir velocity, the reservoir velocity divided by the main frequency of the seismic data is equal to the seismic wavelength; 1 / 4 of the seismic wavelength is the reservoir thickness reflected by the seismic attributes.

6. The reservoir inversion method based on optimization of well-seismic characteristic parameters according to claim 1, characterized in that: The method of performing several different combinations of the upper and lower deposition units of the central target in S8 to form multiple combination units includes: The central target deposition unit and the immediately preceding deposition unit are combined into a combined unit 1; the central target deposition unit and the immediately preceding two deposition units are combined into a combined unit 2; The central target sedimentary unit and the next sedimentary unit are combined as combined unit 3; The central target sedimentary unit and the two immediately following sedimentary units are combined as combined unit 4; The central target deposition unit and the immediately upper and lower deposition units are combined into a combined unit 5; the central target deposition unit and the immediately upper and lower deposition units are combined into a combined unit 6; The central target sedimentary unit and the two sedimentary units immediately above and below are combined as combined unit 7; The central target sedimentary unit and the two immediately above and two immediately below sedimentary units are combined as combined unit 8; The central target sedimentary unit serves as combined unit 9; The maximum formation thickness of the combined unit must be smaller than the reservoir thickness that can be reflected by the seismic event in step S7.

7. The reservoir inversion method based on well-seismic characteristic parameter optimization according to claim 1, characterized in that: The method for completing the well-seismic combined reservoir inversion in the target layer in step S12 is to use the framework model obtained in S11 in combination with existing inversion software to complete the well-seismic combined reservoir inversion prediction in the target layer.

8. The reservoir inversion method based on optimization of well-seismic characteristic parameters according to claim 7, characterized in that: The inversion software is seismic waveform indication inversion software SMI, Jason.

Images