Method and device for predicting reservoir thickness error, storage medium and electronic equipment

Abstract

The invention discloses a method and device for predicting a reservoir thickness error, a storage medium and electronic equipment, and relates to an earthquake prediction technology, and the method comprises the following steps: generating a theoretical model earthquake data volume, and carrying out the relation deduction through a model; constructing an actual data volume, intersecting the actual data volume with the theoretical model seismic data volume, and performing analysis and correction processing to obtain a data source I for predicting the reservoir thickness; a second data source for predicting the reservoir thickness is obtained; and generating a predicted reservoir thickness map. Compared with the prior art, the method has the advantages that (1) a simple and practical technology for predicting the reservoir by using the seismic attributes is provided; (2) the workload is greatly reduced, and the requirement on computer equipment is lower; (3) the requirements on technical and knowledge backgrounds of personnel are low; and (4) the problem that modeling is needed for inversion prediction of the reservoir is solved, particularly, modeling of a complex geological structure area is difficult, reservoir prediction of the complex geological structure area becomes easy and fast, and the precision of seismic amplitude attribute prediction of the reservoir is accurately improved.

Description

Technical Field

[0001] The present invention relates to earthquake prediction technology, and in particular to a method, a device, a storage medium and an electronic device for predicting reservoir thickness error. Background Art

[0002] Seismic attribute reservoir prediction techniques can be roughly divided into three categories: seismic attribute analysis technology, seismic inversion technology, and AVO technology. At the same time, the prediction content mainly includes reservoir lithology, reservoir physical properties, reservoir morphology, and reservoir oil and gas content. In the 1970s, with the emergence of three-dimensional seismic exploration technology, recursive inversion and trace integrals and other technical methods for predicting reservoirs gradually became popular; in the 1990s, the detection of faults and geological anomalies based on coherence body analysis technology was effective, which led to a boom in the application of three-dimensional attribute body technology; and in recent years, the organic combination of geological statistical simulation technology and seismic inversion technology has also made the inversion results more consistent with the actual reservoir characteristics.

[0003] From the above seismic attribute research process, it is found that there are many parameters such as oil and gas content, porosity, lithology, water saturation, etc. predicted by seismic attributes, but there are few studies on quantitatively predicting reservoir thickness, especially thin reservoirs, by using seismic attributes. At present, seismic attributes are generally used to qualitatively predict reservoir thickness, with low accuracy.

[0004] In response to the above problems, there are some improvement schemes in the prior art, such as the thin interbed reservoir prediction method based on seismic multi-attribute deep learning proposed in application number CN201910572785.2, which improves the prediction accuracy by training the machine learning model. However, this method still relies on a large amount of actual data, and does not provide an effective solution to the mismatch between theoretical models and actual data.

[0005] The existing technology still has the following problems: (1) General seismic attributes are mostly used for qualitative prediction of reservoir distribution, which results in large prediction errors, especially for thin reservoirs; (2) Inversion prediction of reservoirs requires modeling, which is difficult in areas with complex geological structures; (3) Inversion prediction of reservoirs requires a lot of work and high requirements for computer equipment; (4) It requires high technical and knowledge background of the actual staff. Summary of the invention

[0006] In order to solve the above technical problems, the present invention provides a technical solution: a method for predicting reservoir thickness error, comprising the following steps:

[0007] S1. Generate a theoretical model seismic data body and deduce the relationship through the model; specifically, establish a model through the drilling data of the work area, establish a model through the statistical velocity data obtained from the previous velocity research results of the drilling work area, and then use the model and wavelet to perform convolution operation to obtain a model seismic data body, interpret the corresponding target layer, extract the amplitude data of the model seismic data body, and then obtain function a through intersection analysis of the relationship between the model formation thickness and the model seismic data body amplitude, and use function a to predict the model formation thickness, and finally define the difference between the model thickness and the thickness predicted by the model seismic data body amplitude as the prediction error, and analyze the relationship between the model thickness and the prediction error thickness through intersection analysis to obtain the error function b;

[0008] S2. Construct an actual data body, intersect, analyze and correct it with the theoretical model seismic data body, and obtain the data source for predicting reservoir thickness. Specifically, through the seismic data collected and processed in the actual work area, use the drilling data calibration work to obtain the wave group position of the seismic data collected and processed corresponding to the research target layer, interpret the layer of the seismic data collected and processed, extract the amplitude data of the seismic data collected and processed, and then use the interpretation layer to take the ±300 millisecond time window up and down to extract the actual sub-waves of the seismic data collected and processed, and then intersect and analyze the amplitude data of the two seismic data bodies obtained above, and use the obtained relationship formula to correct the amplitude data of the seismic data collected and processed;

[0009] S3. Obtain the second data source for predicting reservoir thickness; specifically, for predicting target layer thickness, when there is a lot of drilling data, the drilling thickness and the amplitude data of the corrected acquired and processed seismic data are intersected and analyzed to obtain a relationship function between the two, and the formation thickness is predicted using the relationship function, and then the error between the drilling thickness and the formation thickness predicted by the corrected acquired and processed seismic data amplitude is calculated, and the drilling thickness and the formation thickness predicted by the corrected acquired and processed seismic data amplitude are intersected and analyzed to obtain the actual data error function c; if there is little drilling data or even no drilling, the relationship function a can be used to predict the target layer thickness;

[0010] S4. Generate a predicted reservoir thickness map, specifically, correct the predicted target layer thickness. When there is little drilling data or even no drilling, the error function b is used to correct the amplitude-predicted formation thickness, that is, the formation thickness predicted by the amplitude of the corrected acquired and processed seismic data; when there is a lot of drilling data, the error function c is used to correct the amplitude-predicted formation thickness, that is, the formation thickness predicted by the amplitude of the corrected acquired and processed seismic data.

[0011] Optionally, S1 further includes the following specific steps:

[0012] S1.1. First, establish the actual drilling formation model of the work area, then extract wavelets from the interpretation results of the actual target layer of the work area, and then screen the theoretical wavelets matching the work area. The actual drilling formation model data of the work area extracted from the above three data sources, the wavelet data of the interpretation results of the actual target layer of the work area, and the theoretical wavelet data matching the work area are synthesized into a theoretical model seismic data body; the wavelet data can be directly obtained through the various theoretical wavelets provided by the interpretation software, and the actual wavelet adopts the interpretation layer data along the actual seismic data. The interpretation layer data takes the upper and lower ±300 millisecond time windows and then is calculated by the software;

[0013] S1.2, extracting the amplitude attributes of the seismic data volume of the theoretical model in S1.1;

[0014] S1.3., regress the amplitude attribute of the seismic data volume of the theoretical model in S1.2, and the relationship formula between reservoir thickness and amplitude attribute;

[0015] Optionally, S2 further includes the following specific steps:

[0016] S2.1, import the amplitude attributes of the theoretical model seismic data volume in S1.2;

[0017] S2.2. Extract the amplitude attributes of the target layer along the layer from the actual target layer interpretation results in the work area. The amplitude attributes of the target layer along the layer are the actual data body;

[0018] S2.3, the amplitude attributes of the seismic data volume of the theoretical model in S1.2 and the amplitude attributes of the target layer along the layer are analyzed for the intersection relationship of the amplitude attributes;

[0019] S2.4. According to the intersection analysis structure of S2.2, the amplitude attributes along the layers of the actual seismic data volume are corrected, and the corrected amplitude attributes are used as the first data source for predicting reservoir thickness.

[0020] Optionally, S3 also includes the following specific steps:

[0021] S3.1. Obtaining amplitude attribute prediction reservoir thickness data;

[0022] S3.2, performing error analysis between the predicted reservoir thickness data obtained in S3.1 and the actual reservoir thickness data to generate an error function, which is used as the second data source for predicting reservoir thickness;

[0023] Optionally, S4 further includes the following specific steps:

[0024] S4.1. Correct the predicted reservoir thickness according to data source 1 and data source 2 to obtain accurate reservoir thickness;

[0025] S4.2. Generate a corrected predicted reservoir thickness map.

[0026] Optionally, the amplitude attributes of the theoretical model seismic data volume extracted in S1.1 are used to select data points near the tuning thickness. The selection criteria are: data less than 5 meters or greater than 5 meters of the tuning thickness, and various formula regressions are generated for the selected data points.

[0027] Optionally, the predicted reservoir thickness obtained by adjusting and correcting the actual seismic attribute value minus the error amount is used to obtain the accurate reservoir thickness, where the error amount is the error between the actual thickness obtained and the predicted thickness.

[0028] Optionally, the means for predicting reservoir thickness error comprises:

[0029] Data preprocessing and theoretical model building module, which cleans, organizes and formats the input drilling data and seismic data to ensure data consistency and accuracy. Based on the preprocessed data, a reservoir theoretical model is built, including stratigraphic structure, lithology distribution and seismic response characteristics, and the theoretical relationship between reservoir thickness and seismic amplitude is deduced;

[0030] Actual data extraction and correction module, which extracts amplitude attributes from seismic data. The attributes are used for subsequent reservoir thickness prediction. The actual amplitude attributes are compared with the amplitude attributes in the theoretical model, and necessary corrections are made to eliminate data deviations and improve prediction accuracy.

[0031] Reservoir thickness prediction module, which uses the corrected amplitude attributes and the relationship between reservoir thickness and amplitude in the theoretical model to predict reservoir thickness and output preliminary reservoir thickness prediction results, providing a basis for subsequent error analysis;

[0032] The error analysis and correction module compares the preliminary prediction results with the actual reservoir thickness data, calculates the prediction error, analyzes the source and distribution of the error, evaluates the accuracy and reliability of the prediction model, and iteratively corrects the prediction model based on the error analysis results to improve the prediction accuracy;

[0033] The reservoir thickness optimization and mapping module further optimizes the reservoir thickness based on the prediction results after error correction to obtain a more accurate reservoir thickness distribution, and converts the optimized reservoir thickness data into an intuitive mapping representation to facilitate geological interpretation and reservoir evaluation.

[0034] Optionally, a computer-readable storage medium is provided, characterized in that the storage medium stores a computer program, and when the computer program is executed by a processor, the method of any one of claims 1 to 5 is implemented.

[0035] Optionally, an electronic device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the method of any one of claims 1 to 5 is implemented when the processor executes the program.

[0036] Compared with the prior art, the advantages of the present invention are: (1) providing a simple and practical technology for predicting reservoirs using seismic attributes; (2) greatly reducing the workload and requiring lower computer equipment; (3) requiring lower technical and knowledge background of personnel; and (4) overcoming the problem of modeling required for inversion prediction of reservoirs, especially the difficulty in modeling in areas with complex geological structures, making reservoir prediction in complex geological areas easy, fast and accurate, and improving the accuracy of reservoir prediction by seismic amplitude attributes. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 It is a schematic diagram of the process of the present invention.

[0038] Figure 2 It is a schematic diagram of the relationship between amplitude and thickness of the generated seismic data volume.

[0039] Figure 3 It is a function fitting diagram of the relationship between thickness and amplitude.

[0040] Figure 4 It is the correction formula and error diagram generated by the theoretical model.

[0041] Figure 5 It is a schematic diagram of the error between the actual drilling thickness and the thickness predicted by the amplitude attribute.

[0042] Figure 6 It is a schematic diagram of the corrected sandstone predicted thickness. DETAILED DESCRIPTION

[0043] The present invention is further described in detail below in conjunction with the accompanying drawings.

[0044] Embodiment 1

[0045] like Figure 1 As shown in the figure, this scheme proposes a method for predicting reservoir thickness error, aiming to improve the accuracy and reliability of reservoir thickness prediction. Specifically:

[0046] like Figure 1 As shown, the theoretical model seismic data volume is generated: by establishing the actual drilling formation model of the work area, extracting sub-waves, and screening the theoretical sub-waves that match the work area, the theoretical model seismic data volume is synthesized. Then the amplitude attributes of the theoretical model seismic data volume are extracted, and the relationship is deduced. In steps, the actual drilling forward model of the work area is established, and then the model seismic data consistent with the main frequency of the actual work area is generated. The amplitude and thickness relationship of the generated seismic data volume are intersected, analyzed and generated.

[0047] like Figure 2As shown in the figure, the actual data volume is constructed: the amplitude attributes of the target layer along the layer are extracted from the actual target layer interpretation results in the work area as the actual data volume. The amplitude attributes of the theoretical model seismic data volume are intersected and analyzed with the amplitude attributes of the actual data volume, and the amplitude attributes of the actual seismic data volume along the layer are corrected. In terms of steps, the amplitude attributes near the tuning thickness are first fitted with the data function of the reservoir thickness, and the amplitude predicted thickness is then fitted with the model thickness error analysis and error function.

[0048] like Figure 3 As shown in the figure, the data sources for predicting reservoir thickness are: on the one hand, the corrected amplitude attributes are used to predict reservoir thickness and obtain preliminary prediction results. On the other hand, the prediction results are compared with the actual reservoir thickness data, the error is calculated, and the error function is generated. In terms of steps, the amplitude attributes of the interpreted layers in the actual work area are first extracted, then the relationship between the amplitude attributes of the theoretical synthetic records is analyzed, and then the actual seismic amplitude attribute values ​​are corrected. Finally, the actual seismic attribute values ​​after adjustment and correction are used to predict reservoir thickness.

[0049] like Figure 3 As shown in Figure 1, a predicted reservoir thickness map is generated: the predicted reservoir thickness is corrected according to the error function to obtain the accurate reservoir thickness, and a corrected predicted reservoir thickness map is generated. The reservoir thickness correction (i.e. Figure 3 The error amount obtained in ).

[0050] Embodiment 2

[0051] like Figures 2 to 5 As shown, the device for predicting reservoir thickness error includes: (1) a data preprocessing and theoretical model building module, which cleans, organizes and formats the input drilling data and seismic data to ensure the consistency and accuracy of the data, and builds a reservoir theoretical model based on the preprocessed data, including the stratigraphic structure, lithology distribution and seismic response characteristics, and deduces the theoretical relationship between reservoir thickness and seismic amplitude;

[0052] (2) Actual data extraction and correction module, which extracts amplitude attributes from seismic data. The attributes are used for subsequent reservoir thickness prediction. The actual amplitude attributes are compared with the amplitude attributes in the theoretical model and necessary corrections are made to eliminate data deviations and improve prediction accuracy.

[0053] (3) Reservoir thickness prediction module: This module uses the corrected amplitude attributes and the relationship between reservoir thickness and amplitude in the theoretical model to predict reservoir thickness and output preliminary reservoir thickness prediction results, providing a basis for subsequent error analysis;

[0054] (4) Error analysis and correction module, which compares the preliminary prediction results with the actual reservoir thickness data, calculates the prediction error, analyzes the source and distribution of the error, evaluates the accuracy and reliability of the prediction model, and iteratively corrects the prediction model based on the error analysis results to improve the prediction accuracy;

[0055] (5) The reservoir thickness optimization and mapping module further optimizes the reservoir thickness based on the prediction results after error correction to obtain a more accurate reservoir thickness distribution. The optimized reservoir thickness data is converted into an intuitive mapping representation to facilitate geological interpretation and reservoir evaluation.

[0056] The working process of the device is as follows:

[0057] (1) Data preprocessing and theoretical model construction: Drilling data and seismic data are collected, cleaned, organized, and formatted to ensure data consistency and accuracy. Based on the preprocessed data, a reservoir theoretical model is constructed, including stratigraphic structure, lithology distribution, and seismic response characteristics. The relationship between reservoir thickness and seismic amplitude is established through relationship deduction using the model.

[0058] (2) Actual data extraction and correction: Extract amplitude attributes from seismic data as key data for predicting reservoir thickness. Intersect and analyze the extracted amplitude attributes with the amplitude attributes in the theoretical model to find out the differences and make corrections to eliminate data deviation.

[0059] (3) Reservoir thickness prediction and error analysis: Using the corrected amplitude attributes and the relationship between reservoir thickness and amplitude in the theoretical model, a preliminary prediction of reservoir thickness is made. The preliminary prediction results are compared with the actual reservoir thickness data, the prediction error is calculated, and an error function is generated.

[0060] (4) Correction and optimization of predicted reservoir thickness: The predicted reservoir thickness is corrected according to the error function to obtain a more accurate reservoir thickness. The corrected reservoir thickness is further optimized to improve the accuracy and reliability of the prediction results.

[0061] (5) Generate predicted reservoir thickness map: Convert the optimized reservoir thickness data into an intuitive map to facilitate geological interpretation and reservoir evaluation.

[0062] It can be seen that this scheme can use seismic attributes to quantitatively predict reservoir thickness and distribution; the method is simple, efficient and accurate, simpler and more practical than complex inversion prediction of reservoirs, and overcomes the problem of inversion prediction of reservoirs requiring modeling (difficult modeling in complex geological structure areas). This technology only requires an interpretation horizon of the target layer and does not require modeling. The workload is greatly reduced, and the requirements for computer equipment are low; the requirements for personnel technology, knowledge background, training, etc. are low.

[0063] Embodiment 3

[0064] This embodiment provides the specific steps of the method for predicting reservoir thickness error:

[0065] 1. Step S1: Generate theoretical model seismic data volume and perform relationship deduction

[0066] S1.1 Build model and synthesize seismic data volume

[0067] The formation model is established using the actual drilling formation data of the work area. Wavelets are extracted from the actual target layer interpretation results of the work area, and the theoretical wavelets that match the work area are screened. The theoretical model seismic data body is synthesized, including the actual drilling formation model data of the work area, the wavelet data of the actual target layer interpretation results, and the theoretical wavelet data. The wavelet data is obtained through the theoretical wavelet provided by the interpretation software, and the actual wavelet is obtained by using the interpretation layer data along the actual seismic data, taking the upper and lower ±300 millisecond time windows, and calculated by the software.

[0068] S1.2 Extracting amplitude attributes

[0069] Extract amplitude attributes from synthetic theoretical model seismic data volumes.

[0070] S1.3 Regression relationship formula

[0071] The amplitude attribute of the seismic data volume of the theoretical model and the reservoir thickness are regressed to obtain the relationship function a.

[0072] The relationship function a uses linear regression:

[0073] H pred =a·A+b H pred is the predicted reservoir thickness, A is the amplitude attribute, and a and b are the regression coefficients.

[0074] Select data points near the tuning thickness (less than 5 meters or greater than 5 meters of the tuning thickness), and generate various formula regressions for the selected data points to improve the accuracy of the formula.

[0075] Then predict the error and error function b, and use the relationship function a to predict the model formation thickness. Calculate the difference between the model thickness and the predicted thickness to get the prediction error. Intersect and analyze the relationship between the model thickness and the predicted error thickness to get the error function b. The error function b is based on the relationship between the model thickness and the predicted error. The specific formula is:

[0076] E b =f(H model ,H pred -H model )

[0077] Among them, E b is the output of the error function b, Hmodel is the model thickness, H pred is the predicted thickness.

[0078] Step S2: Constructing the actual data volume and conducting intersection analysis

[0079] S2.1 Import theoretical model amplitude attributes

[0080] Import the amplitude attributes of the theoretical model seismic data volume obtained in S1.2.

[0081] S2.2 Extracting the amplitude attributes of the actual data volume

[0082] The amplitude attributes of the target layer along the layer, i.e. the actual data body, are extracted from the interpretation results of the actual target layer in the work area.

[0083] S2.3 Intersection analysis

[0084] The amplitude attributes of the theoretical model are combined with the amplitude attributes of the actual data volume for analysis.

[0085] S2.4 Correction of amplitude properties

[0086] According to the results of the intersection analysis, the amplitude attributes along the layers of the actual seismic data volume are corrected to obtain the corrected amplitude attributes, which serve as the first data source for predicting reservoir thickness.

[0087] 3. Step S3: Obtaining Data Source 2 for Predicting Reservoir Thickness

[0088] S3.1 Obtaining amplitude attributes to predict reservoir thickness data

[0089] The reservoir thickness is predicted using the corrected amplitude attribute and the relationship function a (or a function adjusted according to actual conditions).

[0090] S3.2 Error analysis and error function c

[0091] Compare the predicted reservoir thickness with the actual reservoir thickness and calculate the error. Intersect and analyze the error between the drilling thickness and the formation thickness predicted by the corrected acquisition and processing seismic data amplitude to obtain the actual data error function c. The error function c is based on the relationship between the drilling thickness and the prediction error. The specific formula is:

[0092] E c =g(H drill ,H pred -H drill )

[0093] Among them, E c is the output of the error function c, H drill is the drilling thickness.

[0094] Step S4: Generate a predicted reservoir thickness map

[0095] S4.1 Correction of predicted reservoir thickness

[0096] When drilling data is scarce, the error function b is used to correct the formation thickness predicted by amplitude.

[0097] When there are a lot of drilling data, the error function c is used to correct the formation thickness predicted by the amplitude.

[0098] S4.2 Generate a graph

[0099] Convert the corrected reservoir thickness data into an intuitive graphical representation.

[0100] The above shows and describes the basic principles and main features of the present invention and the advantages of the invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A method for predicting reservoir thickness error, characterized in that The following steps are involved: S1. Generate a theoretical model seismic data body and deduce the relationship through the model; specifically, establish a model through the drilling data of the work area, establish a model through the statistical velocity data obtained from the previous velocity research results of the drilling work area, and then use the model and wavelet to perform convolution operation to obtain a model seismic data body, interpret the corresponding target layer, extract the amplitude data of the model seismic data body, and then obtain function a through intersection analysis of the relationship between the model formation thickness and the model seismic data body amplitude, and use function a to predict the model formation thickness, and finally define the difference between the model thickness and the thickness predicted by the model seismic data body amplitude as the prediction error, and analyze the relationship between the model thickness and the prediction error thickness through intersection analysis to obtain the error function b; S2. Construct an actual data body, intersect, analyze and correct it with the theoretical model seismic data body, and obtain the data source for predicting reservoir thickness. Specifically, through the seismic data collected and processed in the actual work area, use the drilling data calibration work to obtain the wave group position of the seismic data collected and processed corresponding to the research target layer, interpret the layer of the seismic data collected and processed, extract the amplitude data of the seismic data collected and processed, and then use the interpretation layer to take the ±300 millisecond time window up and down to extract the actual sub-waves of the seismic data collected and processed, and then intersect and analyze the amplitude data of the two seismic data bodies obtained above, and use the obtained relationship formula to correct the amplitude data of the seismic data collected and processed; S3. Obtain the second data source for predicting reservoir thickness; specifically, for predicting target layer thickness, when there is a lot of drilling data, the drilling thickness and the amplitude data of the corrected acquired and processed seismic data are intersected and analyzed to obtain a relationship function between the two, and the formation thickness is predicted using the relationship function, and then the error between the drilling thickness and the formation thickness predicted by the corrected acquired and processed seismic data amplitude is calculated, and the drilling thickness and the formation thickness predicted by the corrected acquired and processed seismic data amplitude are intersected and analyzed to obtain the actual data error function c; if there is little drilling data or even no drilling, the relationship function a can be used to predict the target layer thickness; S4. Generate a predicted reservoir thickness map, specifically, correct the predicted target layer thickness. When there is little drilling data or even no drilling, the error function b is used to correct the amplitude-predicted formation thickness, that is, the formation thickness predicted by the amplitude of the corrected acquired and processed seismic data; when there is a lot of drilling data, the error function c is used to correct the amplitude-predicted formation thickness, that is, the formation thickness predicted by the amplitude of the corrected acquired and processed seismic data.

2. The method for predicting reservoir thickness error according to claim 1, characterized in that The S1 further comprises the following specific steps: S1.

1. First, establish the actual drilling formation model of the work area, then extract wavelets from the interpretation results of the actual target layer of the work area, and then screen the theoretical wavelets matching the work area. The actual drilling formation model data of the work area extracted from the above three data sources, the wavelet data of the interpretation results of the actual target layer of the work area, and the theoretical wavelet data matching the work area are synthesized into a theoretical model seismic data body; the wavelet data can be directly obtained through the various theoretical wavelets provided by the interpretation software, and the actual wavelet adopts the interpretation layer data along the actual seismic data. The interpretation layer data takes the upper and lower ±300 millisecond time windows and then is calculated by the software; S1.2, extracting the amplitude attributes of the seismic data volume of the theoretical model in S1.1; S1.3, regress the amplitude attribute of the seismic data volume of the theoretical model in S1.2, and the relationship formula between reservoir thickness and amplitude attribute.

3. The method for predicting reservoir thickness error according to claim 1, characterized in that The S2 further comprises the following specific steps: S2.1, import the amplitude attributes of the theoretical model seismic data volume in S1.2; S2.

2. Extract the amplitude attributes of the target layer along the layer from the actual target layer interpretation results in the work area. The amplitude attributes of the target layer along the layer are the actual data body; S2.3, conducting an amplitude attribute intersection analysis on the amplitude attribute relationship between the amplitude attributes of the theoretical model seismic data volume in S1.2 and the amplitude attributes of the target layer along the layer; S2.

4. According to the intersection analysis structure of S2.2, the amplitude attributes along the layers of the actual seismic data volume are corrected, and the corrected amplitude attributes are used as the first data source for predicting reservoir thickness.

4. The method for predicting reservoir thickness error according to claim 1, characterized in that The S3 also includes the following specific steps: S3.

1. Obtaining amplitude attribute prediction reservoir thickness data; S3.

2. Perform error analysis on the predicted reservoir thickness data obtained in S3.1 and the actual reservoir thickness data to generate an error function. The error function serves as the second data source for predicting reservoir thickness.

5. The method for predicting reservoir thickness error according to claim 1, characterized in that The S4 further comprises the following specific steps: S4.

1. Correct the predicted reservoir thickness according to data source 1 and data source 2 to obtain accurate reservoir thickness; S4.

2. Generate a corrected predicted reservoir thickness map.

6. The method for predicting reservoir thickness error according to claim 1, characterized in that: The amplitude attributes of the theoretical model seismic data body extracted in S1.1 are used to select data points near the tuning thickness. The selection criteria are: data less than 5 meters or greater than 5 meters of the tuning thickness, and various formula regressions are generated for the selected data points.

7. The method for predicting reservoir thickness error according to claim 4, characterized in that: The predicted reservoir thickness obtained by adjusting and correcting the actual seismic attribute value minus the error amount is used to obtain the accurate reservoir thickness, where the error amount is the error between the actual thickness obtained and the predicted thickness.

8. A device for predicting reservoir thickness error, characterized in that include: Data preprocessing and theoretical model building module, which cleans, organizes and formats the input drilling data and seismic data to ensure data consistency and accuracy. Based on the preprocessed data, a reservoir theoretical model is built, including stratigraphic structure, lithology distribution and seismic response characteristics, and the theoretical relationship between reservoir thickness and seismic amplitude is deduced; Actual data extraction and correction module, which extracts amplitude attributes from seismic data. The attributes are used for subsequent reservoir thickness prediction. The actual amplitude attributes are compared with the amplitude attributes in the theoretical model, and necessary corrections are made to eliminate data deviations and improve prediction accuracy. Reservoir thickness prediction module, which uses the corrected amplitude attributes and the relationship between reservoir thickness and amplitude in the theoretical model to predict reservoir thickness and output preliminary reservoir thickness prediction results, providing a basis for subsequent error analysis; The error analysis and correction module compares the preliminary prediction results with the actual reservoir thickness data, calculates the prediction error, analyzes the source and distribution of the error, evaluates the accuracy and reliability of the prediction model, and iteratively corrects the prediction model based on the error analysis results to improve the prediction accuracy; The reservoir thickness optimization and mapping module further optimizes the reservoir thickness based on the prediction results after error correction to obtain a more accurate reservoir thickness distribution, and converts the optimized reservoir thickness data into an intuitive mapping representation to facilitate geological interpretation and reservoir evaluation.

9. A computer-readable storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the method according to any one of claims 1 to 5 is implemented.

Images