Shallow AVO response feature correction method based on forward modeling

Abstract

The invention relates to the technical field of exploration, in particular to a shallow AVO response feature correction method based on forward modeling. The method comprises the following steps: acquiring logging data, a shallow penetration profile and seismic data; constructing an initial geologic model according to the logging data, the shallow penetration profile and the seismic data to obtain the initial geologic model; performing high-resolution elastic parameter analysis according to the initial geologic model to obtain a fine elastic model; performing multi-angle combined wave field simulation according to the fine elastic model to obtain a multi-angle combined wave field; performing analog data preprocessing on the multi-angle combined wave field, and performing multi-angle reflection wave field extraction to obtain multi-angle reflection wave field data; and performing gas reservoir identification on the multi-angle reflection wave field data to obtain a gas reservoir identification preliminary result. According to the method, the problem that a conventional AVO inversion method is poor in shallow applicability is solved, and the accuracy and reliability of shallow AVO inversion are improved.

Description

Technical Field

[0001] The invention relates to the field of exploration technology, and in particular to a shallow AVO response characteristic correction method based on forward simulation. Background Art

[0002] The purpose of shallow AVO (amplitude variation with offset) analysis is to use the variation of seismic wave amplitude with offset to infer the lithology and fluid properties of underground media. Early AVO technology was mainly used in deep exploration, and AVO inversion was performed based on simplified forms of the Zoeppritz equation (such as the Shuey equation). These simplified forms usually assume that the strata are horizontal, uniform, isotropic, and have a small angle of incidence. As exploration targets shift to more complex shallow geological bodies, such as shallow gas reservoirs, groundwater, etc., these simplified assumptions are no longer applicable, resulting in low accuracy of inversion results. Existing shallow AVO response feature correction methods based on forward simulation have poor applicability in shallow applications: The problem of poor shallow applicability is reflected in: 1. Shallow seismic data has low signal-to-noise ratio and limited resolution: Shallow seismic wave propagation path is short and is seriously affected by near-surface and multiple waves, resulting in low signal-to-noise ratio and poor resolution of seismic data, making it difficult to accurately extract AVO information, thus affecting the inversion accuracy.

[0003] 2. The shallow geological structure is complex and the lateral changes are drastic: Shallow geological bodies are usually small in scale and change rapidly in the lateral direction. Compared with deep layers, the assumptions of stratigraphic uniformity and horizontality are more difficult to hold, resulting in the failure of the AVO inversion method based on simplified assumptions.

[0004] 3. Significant dispersion effect: The high-frequency components of shallow seismic waves decay quickly, resulting in significant dispersion effect, which affects the accuracy of AVO response and thus the reliability of inversion results.

[0005] 4. Drastic changes in near-surface velocity: Drastic changes in shallow near-surface velocity have a complex impact on the seismic wave propagation path and AVO response, which are difficult to estimate and correct accurately, thereby reducing the inversion accuracy. Summary of the invention

[0006] Based on this, it is necessary to provide a shallow AVO response characteristic correction method based on forward simulation to solve at least one of the above technical problems.

[0007] To achieve the above purpose, a shallow AVO response characteristic correction method based on forward simulation includes the following steps: Step S1: acquiring well logging data, shallow penetration profiles and seismic data; constructing an initial geological model based on the well logging data, shallow penetration profiles and seismic data to obtain an initial geological model; performing high-resolution elastic parameter analysis based on the initial geological model to obtain a refined elastic model; Step S2: performing multi-angle joint wave field simulation according to the refined elastic model to obtain a multi-angle joint wave field; performing simulation data preprocessing on the multi-angle joint wave field, and performing multi-angle reflection wave field extraction to obtain multi-angle reflection wave field data; Step S3: performing gas reservoir identification on the multi-angle reflection wave field data to obtain preliminary results of gas reservoir identification; performing groundwater layer identification on the multi-angle reflection wave field data to obtain preliminary results of groundwater layer identification; performing local reflection wave feature extraction based on the preliminary results of gas reservoir identification and preliminary results of groundwater layer identification to obtain local reflection anomaly data; Step S4: performing dispersion correction on the local reflection anomaly data to obtain dispersion-corrected data; performing multi-parameter AVO inversion based on the dispersion-corrected data and the local reflection anomaly data to obtain inverted local AVO attribute data; performing AVO attribute fine correction on the inverted AVO attribute data using multi-angle reflection wave field data to obtain corrected AVO attribute data; Step S5: Acquire actual seismic data; perform inversion model error analysis on the corrected AVO attribute data based on the actual seismic data to obtain model data comparison results and error analysis reports; optimize and verify the model based on the model data comparison results and error analysis reports to obtain optimized geological and AVO models.

[0008] The present invention integrates well logging, shallow penetration profile and seismic data, constructs an initial geological model and performs high-resolution elastic parameter analysis, provides an accurate geological and rock physical basis for subsequent AVO simulation and inversion, and effectively improves the accuracy of shallow AVO inversion. Based on the fine elastic model, multi-angle joint wave field simulation and preprocessing are performed to obtain multi-angle reflection wave field data, which can simulate the propagation law of real seismic waves and provide reliable multi-angle information for subsequent AVO analysis. Using multi-angle reflection wave field data to identify gas reservoirs and groundwater layers and extract local reflection anomaly data can highlight the AVO anomaly characteristics of the target area and provide more accurate data input for subsequent fine AVO inversion. Dispersion correction and multi-parameter AVO inversion are performed on local reflection anomaly data, and multi-angle reflection wave field data is used to perform fine correction of AVO attributes, which can effectively improve the accuracy and resolution of AVO inversion and obtain more reliable AVO attributes. By comparing and error analyzing with actual seismic data, and optimizing and verifying the model, the accuracy and reliability of the final AVO model can be ensured, providing a more effective means for shallow oil and gas exploration and groundwater resource evaluation. Therefore, the present invention provides a shallow AVO response characteristic correction method based on forward simulation, which effectively solves the problem of poor applicability of conventional AVO inversion methods in shallow layers, improves the accuracy and reliability of shallow AVO inversion, and provides a more effective means for shallow oil and gas exploration and groundwater resource evaluation.

[0009] Preferably, step S1 comprises the following steps: Step S11: acquiring well logging data, shallow penetration profiles and seismic data; performing data preprocessing on the well logging data, shallow penetration profiles and seismic data, and performing data quality control to obtain preprocessed multi-source data; Step S12: performing stratigraphic division according to the pre-processed multi-source data, and performing stratigraphic comparison and correction processing to obtain layered data and stratigraphic information; Step S13: constructing an initial geological model using the layered data and the stratigraphic information to obtain an initial geological model; Step S14: constructing a high-resolution elastic parameter model according to the initial geological model to obtain an elastic parameter model; constraining and correcting the elastic parameter model according to the initial geological model to obtain a multi-scale elastic parameter model; Step S15: finely characterize the local anomalies of the multi-scale elastic parameter model to obtain an elastic parameter model containing local anomalies; perform model smoothing and consistency processing on the elastic parameter model containing local anomalies to obtain a fine elastic model.

[0010] The present invention can improve the signal-to-noise ratio and consistency of data, eliminate errors and outliers in the data, and provide a reliable data basis for subsequent geological model construction and elastic parameter analysis by preprocessing and quality controlling the well logging data, shallow penetration profiles, and seismic data. According to the preprocessed multi-source data, stratigraphic division, stratigraphic comparison, and correction processing are performed to establish an accurate stratigraphic framework, clarify the spatial distribution and contact relationship of different strata, and provide accurate stratigraphic information and layered data for subsequent geological model construction. By constructing an initial geological model using layered data and stratigraphic information, a preliminary understanding of the geological structure and lithology distribution of the study area can be established, providing a basic model for subsequent elastic parameter analysis and AVO inversion. According to the initial geological model, a high-resolution elastic parameter model is constructed, and constraints and corrections are performed in combination with the initial geological model, so that a more accurate and refined elastic parameter model can be obtained, which reflects the elastic differences of different strata and lithology, and provides more realistic model parameters for subsequent multi-angle wave field simulation. Fine characterization of local anomalies and model smoothing and consistency processing of the multi-scale elastic parameter model can highlight the elastic characteristics of local anomalies, while ensuring the overall smoothness and geological rationality of the model, and improving the accuracy and reliability of AVO inversion.

[0011] Preferably, step S2 comprises the following steps: Step S21: setting forward modeling parameters according to the spatial scale and target accuracy of the refined elastic model to obtain a forward modeling parameter table; Step S22: setting the earthquake source position for the refined elastic model and performing earthquake source wavelet excitation to obtain an earthquake source excitation model; Step S23: performing multi-angle joint wave field simulation according to the earthquake source excitation model and the forward parameter table to obtain a multi-angle joint wave field; Step S24: setting the detection point position for the fine elastic model, and recording the time series data of the multi-angle joint wave field at the detection point to obtain multi-angle detection point data; Step S25: converting the multi-angle detection point data into a seismic data format to obtain multi-angle detection point format data; Step S26: performing simulation data preprocessing on the multi-angle joint wave field according to the multi-angle detection point format data to obtain multi-angle reflection wave field data.

[0012] The present invention can ensure the accuracy and efficiency of forward simulation by setting forward modeling parameters according to the spatial scale and target accuracy of the fine elastic model, and make the simulation results more consistent with the actual geological conditions, so as to provide reliable simulation data for subsequent AVO analysis. By setting the source position and exciting the source wavelet in the fine elastic model, the propagation process of seismic waves in the underground medium can be simulated, and the source excitation conditions can be provided for the subsequent multi-angle joint wavefield simulation. According to the source excitation model and the forward parameter table, the multi-angle joint wavefield simulation can be performed to obtain seismic wavefield data with different incident angles, and provide multi-angle information for the subsequent AVO analysis. By setting the detection point and recording the time series data of the multi-angle joint wavefield at the detection point, the simulated seismic record can be obtained, which provides the data basis for the subsequent data processing and AVO analysis. Converting the multi-angle detection point data into the standard seismic data format can facilitate the subsequent use of seismic data processing software for processing and analysis, and improve the data processing efficiency. Performing simulation data preprocessing on the multi-angle detection point format data can remove the noise and interference in the simulation data, improve the signal-to-noise ratio and resolution of the data, and provide high-quality data for the subsequent AVO analysis.

[0013] Preferably, step S3 comprises the following steps: Step S31: performing multi-angle data fusion on the multi-angle reflection wave field data to obtain fused multi-angle data; performing adaptive filtering on the fused multi-angle data to obtain filtered multi-angle data; Step S32: performing local signal amplification on the filtered multi-angle data to obtain locally amplified data; Step S33: extracting gas reservoir identification attributes and groundwater layer identification attributes from the locally enlarged data to obtain gas reservoir identification attribute data and groundwater layer identification attribute data; Step S34: performing gas reservoir identification on the gas reservoir identification attribute data to obtain a preliminary result of gas reservoir identification; Step S35: performing groundwater layer identification on the groundwater layer identification attribute data to obtain a preliminary result of groundwater layer identification; Step S36: spatially locate and evaluate the properties of the abnormal area based on the preliminary results of gas reservoir identification and groundwater layer identification to obtain local abnormality identification results; Step S37: Extract the reflection wave features of the local abnormal area from the locally amplified data according to the local abnormality recognition result to obtain local reflection abnormality data.

[0014] The present invention can improve the signal-to-noise ratio and resolution of data, suppress noise interference, highlight effective signals, and provide a high-quality data basis for subsequent local signal amplification and attribute extraction through multi-angle data fusion and adaptive filtering. Local signal amplification of the filtered multi-angle data can enhance the signal strength of the target area, highlight local abnormal information, and facilitate subsequent attribute extraction and identification. Extracting gas reservoir identification attributes and groundwater layer identification attributes can convert seismic data into more intuitive and easier to interpret attribute parameters, providing a basis for subsequent gas reservoir and groundwater layer identification. Gas reservoir identification can be performed on gas reservoir identification attribute data to preliminarily determine whether there is a gas reservoir in the study area, and to delineate the potential gas reservoir range, providing a target area for subsequent fine AVO inversion. Groundwater layer identification can be performed on groundwater layer identification attribute data to preliminarily determine whether there is a groundwater layer in the study area, and to delineate the potential groundwater layer range, which is helpful for the exploration and evaluation of groundwater resources. According to the preliminary results of gas reservoir identification and groundwater layer identification, the spatial positioning and property evaluation of the abnormal area can be performed, and the spatial position, scale and property of the abnormal body can be more accurately determined, providing guidance for the subsequent reflection wave feature extraction. By extracting the reflection wave characteristics of the local abnormal area based on the local anomaly identification results, more detailed AVO information can be obtained, providing high-quality data input for subsequent dispersion correction and AVO inversion.

[0015] Preferably, step S34 includes the following steps: Step S341: performing AVO gradient and intercept calculation on the gas reservoir identification attribute data to obtain AVO gradient and intercept data; performing gradient intercept intersection analysis based on the AVO gradient and intercept data to obtain AVO attribute analysis results; Step S342: performing instantaneous attribute calculation on the locally amplified data according to the gas reservoir identification attribute data to obtain instantaneous amplitude and instantaneous frequency data; performing spatial distribution characteristic analysis on the instantaneous amplitude and instantaneous frequency data to obtain instantaneous attribute analysis results; Step S343: performing multi-attribute intersection analysis on the AVO attribute analysis result and the instantaneous attribute analysis result to obtain a multi-attribute intersection analysis result; Step S344: Calculate the probability of each data point belonging to a gas reservoir based on the multi-attribute intersection analysis result to obtain a multi-attribute comprehensive analysis result; Step S345: Perform preliminary judgment on gas reservoir identification based on the multi-attribute comprehensive analysis results and the preset probability threshold to obtain preliminary gas reservoir identification results.

[0016] The present invention can effectively identify AVO anomalies related to gas reservoirs, such as low-impedance gas reservoirs or high-impedance gas reservoirs, by calculating AVO gradients and intercepts and performing gradient-intercept intersection analysis, thereby providing important attribute information for gas reservoir identification. By calculating instantaneous attributes (instantaneous amplitude and instantaneous frequency) and analyzing their spatial distribution characteristics, amplitude and frequency anomalies related to gas reservoirs, such as low-frequency and low-amplitude anomalies, can be identified, further enhancing the reliability of gas reservoir identification. By performing multi-attribute intersection analysis on the AVO attribute analysis results and the instantaneous attribute analysis results, a variety of attribute information can be comprehensively utilized to improve the accuracy of gas reservoir identification and reduce the multi-solution of single attribute interpretation. By calculating the probability that each data point belongs to a gas reservoir, the gas reservoir identification results can be converted from qualitative analysis to quantitative analysis, providing more refined gas reservoir identification results. Preliminary judgment of gas reservoir identification based on the multi-attribute comprehensive analysis results and the preset probability threshold can effectively distinguish gas reservoirs from non-gas reservoirs, and delineate the potential gas reservoir range, providing a target area for subsequent fine AVO inversion.

[0017] Preferably, step S35 includes the following steps: Step S351: Calculate the Q value of the target area for the groundwater layer identification attribute data to obtain the Q value of the target area; use the preset Q value threshold to identify the spatial distribution anomaly of the Q value of the target area to obtain the Q value analysis result; Step S352: Perform velocity analysis on the groundwater layer identification attribute data, and calculate the ratio of the longitudinal wave velocity to the transverse wave velocity to obtain Vp / Vs ratio data; use a preset Vp / Vs ratio threshold to perform spatial distribution anomaly identification on the Vp / Vs ratio data to obtain a Vp / Vs ratio analysis result, where Vp is the longitudinal wave velocity and Vs is the transverse wave velocity; Step S353: performing frequency attenuation analysis on the groundwater layer identification attribute data to obtain a frequency attenuation analysis result; Step S354: performing a multi-attribute superposition analysis according to the Q value analysis result, the Vp / Vs ratio analysis result, and the frequency attenuation analysis result to obtain a multi-attribute superposition analysis result; Step S355: Generate preliminary results of groundwater layer identification based on the multi-attribute superposition analysis results to obtain preliminary results of groundwater layer identification.

[0018] The present invention can identify low Q value anomalies caused by groundwater layers by calculating the Q value of the target area and using a preset Q value threshold to identify spatial distribution anomalies, thereby providing important attribute information for groundwater layer identification. By calculating the Vp / Vs ratio data through velocity analysis and using a preset Vp / Vs ratio threshold to identify spatial distribution anomalies, the high Vp / Vs ratio anomalies caused by groundwater layers can be identified, further enhancing the reliability of groundwater layer identification. Frequency attenuation analysis is performed on the groundwater layer identification attribute data to identify high-frequency attenuation anomalies caused by groundwater layers, providing another important attribute information for groundwater layer identification. The Q value analysis results, the Vp / Vs ratio analysis results, and the frequency attenuation analysis results are subjected to multi-attribute superposition analysis, which can comprehensively utilize a variety of attribute information, improve the accuracy of groundwater layer identification, and reduce the multi-solution of a single attribute interpretation. The preliminary results of groundwater layer identification are generated based on the multi-attribute superposition analysis results, which can effectively distinguish groundwater layers from non-groundwater layers, and delineate the potential groundwater layer range, providing a target area for subsequent research.

[0019] Preferably, step S4 comprises the following steps: Step S41: performing preliminary AVO attribute extraction on the local reflection anomaly data to obtain preliminary AVO attribute data; Step S42: performing multiple wave suppression processing on the local reflection anomaly data to obtain data after multiple wave suppression; Step S43: performing dispersion correction on the data after suppressing the multiple waves to obtain dispersion-corrected data; Step S44: performing multi-parameter AVO inversion according to the preliminary AVO attribute data, the dispersion-corrected data, and the local reflection anomaly data to obtain the inverted local AVO attribute data; Step S45: performing AVO attribute fine correction on the inverted AVO attribute data using the multi-angle reflection wave field data to obtain corrected AVO attribute data.

[0020] The present invention can obtain preliminary AVO information of the target area by performing preliminary extraction of AVO attributes on local reflection anomaly data, and provide initial constraints for subsequent multi-parameter AVO inversion. Performing multiple wave suppression processing on local reflection anomaly data can eliminate the interference of multiple waves on AVO analysis and improve the accuracy of AVO inversion. Performing dispersion correction on the data after suppressing multiple waves can compensate for the energy attenuation and phase distortion generated by seismic waves in the process of underground medium propagation, and improve the resolution and stability of AVO inversion. Performing multi-parameter AVO inversion based on preliminary AVO attribute data, data after dispersion correction and local reflection anomaly data can obtain more accurate elastic parameter (P-wave velocity, S-wave velocity and density) information, and convert it into more refined AVO attributes, thereby improving the accuracy and reliability of AVO inversion. Performing fine AVO attribute correction on the inverted AVO attribute data using multi-angle reflection wave field data can further improve the accuracy and resolution of AVO attributes, making it more consistent with the characteristics of actual seismic data, and providing a more reliable basis for subsequent geological interpretation.

[0021] Preferably, step S44 includes the following steps: Step S441: Selecting a local reflection anomaly area according to the dispersion-corrected data and the local reflection anomaly data to obtain local reflection anomaly selection data; Step S442: extracting local AVO response from the local reflection anomaly selection data according to the preliminary AVO attribute data to obtain local AVO response data; Step S443: performing multi-parameter initialization according to the local AVO response data to obtain a multi-parameter initial model; Step S444: constructing a local AVO inversion model according to the multi-parameter initial model to obtain a local AVO inversion model; Step S445: Perform multi-parameter joint inversion on the local AVO inversion model to obtain inverted local AVO attribute data.

[0022] The present invention can focus the AVO inversion on the target area by selecting the local reflection anomaly area, improve the inversion efficiency and accuracy, and avoid the interference of background noise. By extracting the local AVO response from the local reflection anomaly selection data, purer AVO information can be obtained, providing a more reliable data basis for multi-parameter initialization. Multi-parameter initialization is performed according to the local AVO response data, and a reasonable initial model can be provided for the AVO inversion, improving the convergence speed and stability of the inversion. By constructing a local AVO inversion model according to the multi-parameter initial model, an AVO inversion framework of the target area can be established, providing a basis for multi-parameter joint inversion. By performing multi-parameter joint inversion on the local AVO inversion model, multiple elastic parameters, such as longitudinal wave velocity, shear wave velocity and density, can be inverted simultaneously, improving the accuracy and stability of the inversion result, and reducing multi-solution.

[0023] Preferably, step S445 is specifically as follows: The local AVO response residual is calculated according to the local AVO inversion model and the local AVO response data to obtain the AVO residual data; Perform model parameter perturbation on the multi-parameter initial model to obtain a parameter perturbation model; perform local forward simulation on the parameter perturbation model to obtain local seismic data; Based on the local seismic data and the multi-parameter initial model, the difference calculation is performed to obtain the model difference; based on the model difference and the parameter perturbation model, the sensitivity calculation is performed to obtain the sensitivity matrix; The sensitivity matrix and AVO residual data are used to calculate the parameter update direction to obtain the parameter update direction data; Perform multi-parameter update on the local AVO inversion model according to the parameter update direction data and the preset step size to obtain a multi-parameter update model; The local AVO attribute data is generated according to the multi-parameter updating model to obtain the inverted local AVO attribute data.

[0024] The present invention can quantify the difference between simulated data and observed data by calculating the local AVO response residual, and provide an objective function for subsequent inversion parameter updating. By perturbing the multi-parameter initial model, the model parameter space can be explored to find a better combination of model parameters. By performing local forward simulation on the parameter perturbation model, the influence of the perturbation parameters on the seismic response can be obtained, which is used to calculate the sensitivity matrix. By calculating the model difference, the influence of the parameter perturbation on the seismic response can be quantified. By calculating the sensitivity matrix, the connection between the model parameters and the seismic response can be established, providing a basis for calculating the parameter update direction. By calculating the parameter update direction, the direction of model parameter adjustment can be determined to reduce the difference between the simulated data and the observed data. By updating the local AVO inversion model, the model parameters can be gradually optimized to make the simulated data closer to the observed data. By generating the local AVO attribute data after inversion, more accurate AVO attributes can be obtained, providing a more reliable basis for subsequent geological interpretation.

[0025] Preferably, step S5 comprises the following steps: Step S51: acquiring actual seismic data; performing actual seismic data preprocessing on the actual seismic data to obtain preprocessed actual seismic data; Step S52: extracting AVO attributes from the pre-processed actual seismic data to obtain actual AVO attribute data; Step S53: performing model data comparison and error analysis on the corrected AVO attribute data and the actual AVO attribute data to obtain a model data comparison result and an error analysis report; Step S54: adjusting and optimizing the model parameters of the fine elastic model according to the model data comparison result and the error analysis report to obtain an optimized geological model; Step S55: Perform model verification on the optimized geological model to obtain the optimized geological and AVO models.

[0026] The present invention can provide high-quality actual data for subsequent AVO attribute extraction and model verification by acquiring actual seismic data and performing preprocessing. AVO attribute extraction is performed on the actual seismic data after preprocessing to obtain actual AVO information for comparison and error analysis with simulation results. Comparison and error analysis of the corrected AVO attribute data and the actual AVO attribute data can evaluate the accuracy and reliability of the inversion model and provide guidance for model optimization. Parameter adjustment and optimization of the fine elastic model based on the model data comparison results and the error analysis report can improve the model parameters and enhance the accuracy and prediction ability of the model. Model verification is performed on the optimized geological model to finally confirm the validity and reliability of the optimized model and provide a reliable model basis for subsequent geological interpretation and oil and gas exploration.

[0027] The present invention integrates well logging, shallow penetration profile and seismic data, constructs an initial geological model and performs high-resolution elastic parameter analysis, provides an accurate geological and rock physical basis for subsequent AVO simulation and inversion, and effectively improves the accuracy of shallow AVO inversion. Based on the fine elastic model, multi-angle joint wave field simulation and preprocessing are performed to obtain multi-angle reflection wave field data, which can simulate the propagation law of real seismic waves and provide reliable multi-angle information for subsequent AVO analysis. Using multi-angle reflection wave field data to identify gas reservoirs and groundwater layers and extract local reflection anomaly data can highlight the AVO anomaly characteristics of the target area and provide more accurate data input for subsequent fine AVO inversion. Dispersion correction and multi-parameter AVO inversion are performed on local reflection anomaly data, and multi-angle reflection wave field data is used to perform fine correction of AVO attributes, which can effectively improve the accuracy and resolution of AVO inversion and obtain more reliable AVO attributes. By comparing and error analyzing with actual seismic data, and optimizing and verifying the model, the accuracy and reliability of the final AVO model can be ensured, providing a more effective means for shallow oil and gas exploration and groundwater resource evaluation. Therefore, the present invention provides a shallow AVO response characteristic correction method based on forward simulation, which effectively solves the problem of poor applicability of conventional AVO inversion methods in shallow layers, improves the accuracy and reliability of shallow AVO inversion, and provides a more effective means for shallow oil and gas exploration and groundwater resource evaluation. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 A schematic diagram of the steps of a method for correcting shallow AVO response characteristics based on forward simulation; Figure 2 for Figure 1 Detailed implementation steps of step S3 in FIG. Figure 3 for Figure 1 Detailed implementation steps of step S4 in FIG.

[0029] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings in conjunction with the embodiments. DETAILED DESCRIPTION

[0030] The technical method of the present invention is described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by technicians in this field without creative work are within the scope of protection of the present invention.

[0031] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0032] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0033] To achieve this, please refer to Figures 1 to 3 A shallow AVO response characteristic correction method based on forward modeling includes the following steps: Step S1: acquiring well logging data, shallow penetration profiles and seismic data; constructing an initial geological model based on the well logging data, shallow penetration profiles and seismic data to obtain an initial geological model; performing high-resolution elastic parameter analysis based on the initial geological model to obtain a refined elastic model; Step S2: performing multi-angle joint wave field simulation according to the refined elastic model to obtain a multi-angle joint wave field; performing simulation data preprocessing on the multi-angle joint wave field, and performing multi-angle reflection wave field extraction to obtain multi-angle reflection wave field data; Step S3: performing gas reservoir identification on the multi-angle reflection wave field data to obtain preliminary results of gas reservoir identification; performing groundwater layer identification on the multi-angle reflection wave field data to obtain preliminary results of groundwater layer identification; performing local reflection wave feature extraction based on the preliminary results of gas reservoir identification and preliminary results of groundwater layer identification to obtain local reflection anomaly data; Step S4: performing dispersion correction on the local reflection anomaly data to obtain dispersion-corrected data; performing multi-parameter AVO inversion based on the dispersion-corrected data and the local reflection anomaly data to obtain inverted local AVO attribute data; performing AVO attribute fine correction on the inverted AVO attribute data using multi-angle reflection wave field data to obtain corrected AVO attribute data; Step S5: Acquire actual seismic data; perform inversion model error analysis on the corrected AVO attribute data based on the actual seismic data to obtain model data comparison results and error analysis reports; optimize and verify the model based on the model data comparison results and error analysis reports to obtain optimized geological and AVO models.

[0034] In the embodiment of the present invention, reference Figure 1 The above is a schematic flow chart of the steps of the shallow AVO response characteristic correction method based on forward modeling of the present invention. In this example, the shallow AVO response characteristic correction method based on forward modeling includes the following steps: Step S1: acquiring well logging data, shallow penetration profiles and seismic data; constructing an initial geological model based on the well logging data, shallow penetration profiles and seismic data to obtain an initial geological model; performing high-resolution elastic parameter analysis based on the initial geological model to obtain a refined elastic model; In the embodiment of the present invention, well logging data (density, acoustic wave time difference, gamma ray, etc.), shallow penetration profile and seismic data are first obtained, and preprocessed and quality controlled. Then, stratigraphic division, stratigraphic comparison and correction are performed based on the preprocessed multi-source data to establish a unified layered framework. Using the layered data and stratigraphic information, an initial geological model is constructed in the modeling software, including a structural model, a lithology model and a velocity model. Finally, based on the initial geological model, a high-resolution elastic parameter model is constructed using rock physics software, and constraints and corrections are performed to finally obtain a fine elastic model.

[0035] Step S2: performing multi-angle joint wave field simulation according to the refined elastic model to obtain a multi-angle joint wave field; performing simulation data preprocessing on the multi-angle joint wave field, and performing multi-angle reflection wave field extraction to obtain multi-angle reflection wave field data; In the embodiment of the present invention, the forward modeling parameters are set according to the scale and target accuracy of the fine elastic model. The source position and wavelet are set in the model to perform multi-angle joint wave field simulation. The detection point is set in the model, and the time series data of the multi-angle joint wave field at the detection point is recorded. The detection point data is converted into a seismic data format (such as SEG-Y), and the simulation data is preprocessed (such as removing direct waves, surface waves, noise suppression and amplitude compensation) to obtain multi-angle reflection wave field data.

[0036] Step S3: performing gas reservoir identification on the multi-angle reflection wave field data to obtain preliminary results of gas reservoir identification; performing groundwater layer identification on the multi-angle reflection wave field data to obtain preliminary results of groundwater layer identification; performing local reflection wave feature extraction based on the preliminary results of gas reservoir identification and preliminary results of groundwater layer identification to obtain local reflection anomaly data; In the embodiment of the present invention, multi-angle data fusion and adaptive filtering are performed on multi-angle reflection wave field data. Local signal amplification is performed on the filtered data to highlight the target area. Gas reservoir identification attributes (instantaneous amplitude, instantaneous frequency, AVO gradient and AVO intercept) and groundwater layer identification attributes (Vp / Vs, Q value and frequency attenuation) are extracted. Gas reservoir identification and groundwater layer identification are performed separately to obtain preliminary results. Spatial positioning and property evaluation of abnormal areas are performed based on the preliminary results, and finally the reflection wave characteristics of the local abnormal area are extracted to obtain local reflection abnormal data.

[0037] Step S4: performing dispersion correction on the local reflection anomaly data to obtain dispersion-corrected data; performing multi-parameter AVO inversion based on the dispersion-corrected data and the local reflection anomaly data to obtain inverted local AVO attribute data; performing AVO attribute fine correction on the inverted AVO attribute data using multi-angle reflection wave field data to obtain corrected AVO attribute data; In the embodiment of the present invention, the AVO attributes are initially extracted from the local reflection anomaly data, and multiple wave suppression and dispersion correction are performed. Multi-parameter AVO inversion is performed based on the preliminary AVO attributes, the dispersion-corrected data, and the local reflection anomaly data to obtain the inverted local AVO attribute data. The inverted AVO attribute data is finely corrected using the multi-angle reflection wave field data to obtain the final corrected AVO attribute data.

[0038] Step S5: Acquire actual seismic data; perform inversion model error analysis on the corrected AVO attribute data according to the actual seismic data to obtain model data comparison results and error analysis reports; perform model optimization and verification according to the model data comparison results and error analysis reports to obtain optimized geological and AVO models; In the embodiment of the present invention, actual seismic data is obtained and preprocessed. AVO attributes are extracted from the preprocessed actual seismic data. The corrected AVO attribute data is compared with the actual AVO attribute data, and error analysis is performed to generate model data comparison results and error analysis reports. According to the error analysis results, the parameters of the fine elastic model are adjusted and optimized to obtain an optimized geological model. Finally, the optimized geological model is verified by forward simulation to obtain an optimized geological and AVO model.

[0039] Preferably, step S1 comprises the following steps: Step S11: acquiring well logging data, shallow penetration profiles and seismic data; performing data preprocessing on the well logging data, shallow penetration profiles and seismic data, and performing data quality control to obtain preprocessed multi-source data; Step S12: performing stratigraphic division according to the pre-processed multi-source data, and performing stratigraphic comparison and correction processing to obtain layered data and stratigraphic information; Step S13: constructing an initial geological model using the layered data and the stratigraphic information to obtain an initial geological model; Step S14: constructing a high-resolution elastic parameter model according to the initial geological model to obtain an elastic parameter model; constraining and correcting the elastic parameter model according to the initial geological model to obtain a multi-scale elastic parameter model; Step S15: finely characterize the local anomalies of the multi-scale elastic parameter model to obtain an elastic parameter model containing local anomalies; perform model smoothing and consistency processing on the elastic parameter model containing local anomalies to obtain a fine elastic model.

[0040] In an embodiment of the present invention, first, density, acoustic time difference, gamma ray and other information are extracted from the logging data, and depth correction and environmental correction are performed to remove outliers and noise data. Then, the shallow penetration profile data is converted into a depth domain, and depth matching is performed with the logging data, and the logging information is used to interpret the lithology and velocity of the shallow penetration profile data. Finally, the seismic data is subjected to preprocessing operations such as denoising, static correction, and dynamic correction to improve the signal-to-noise ratio and resolution of the seismic data. The processed logging data, shallow penetration profile data and seismic data are formatted and converted into coordinate systems to generate preprocessed multi-source data. Data quality assessment is performed on the preprocessed multi-source data, including data integrity, consistency and accuracy assessment, to ensure that the data quality meets the subsequent modeling requirements. For example, the logging data is preprocessed and interpreted using Petrel software, the seismic data is preprocessed and depth converted using GeoDepth software, and multi-source data integration and quality control are performed using Techlog software.

[0041] Based on the well logging data in the preprocessed multi-source data, combined with lithology, resistivity, porosity and other information, Stratimagic and other software are used to perform fine stratigraphic division and identify stratigraphic interfaces of different lithologies. The stratigraphic information interpreted by the shallow penetration profile is compared with the logging stratification results, and the stratigraphic information in the shallow penetration profile is corrected to the logging depth reference plane using the depth matching relationship. Seismic horizons are picked up using seismic interpretation software (such as Kingdom), and the picked up seismic horizons are compared with the logging stratification and shallow penetration profile stratification results. Through the horizon correction processing, the consistency of the stratification results of the three is ensured, and finally the stratification data and stratigraphic information are obtained. For example, the strata are divided into sandstone, mudstone, shale, etc. using the logging curve characteristics, and these stratification information are associated with the reflection characteristics in the shallow penetration profile and seismic data to establish a unified stratification framework.

[0042] Using the layered data and stratigraphic information obtained in step S12, an initial geological model is established in modeling software such as Petrel. First, a structural model is established based on the stratigraphic information to define the spatial distribution and geometric morphology of the strata. Then, based on the layered data, different lithologies are assigned to corresponding spatial positions to construct an initial lithology model. Finally, combined with logging data and shallow penetration profile data, the initial model is assigned velocity and density to improve the initial geological model. For example, a fault model and a fold model are constructed based on the interpreted stratigraphic interface, and lithologies such as sandstone and mudstone are assigned to the model, and the velocity is calculated based on the density and acoustic time difference information in the logging data, and the velocity information is assigned to the model.

[0043] Based on the initial geological model, Jason and other rock physics software are used to construct a high-resolution elastic parameter model. First, according to the lithology distribution in the initial geological model, a rock physics model is established to determine the elastic parameters corresponding to different lithologies, such as P-wave velocity, S-wave velocity and density. Then, the rock physics model is calibrated and corrected using logging data to improve the accuracy of the model. Finally, the rock physics model is applied to the initial geological model to generate a high-resolution elastic parameter model. The elastic parameter model is constrained and corrected using the stratigraphic stratification information and lithology distribution in the initial geological model. For example, the elastic parameters are smoothed in different stratigraphic and lithology units to ensure the lateral continuity and geological rationality of the model, and a multi-scale elastic parameter model is obtained.

[0044] According to the preliminary results of gas reservoir identification and groundwater layer identification obtained in step S1, the multi-scale elastic parameter model is finely characterized for local anomalies. For example, in the gas reservoir area, the P-wave velocity and density are reduced and the Poisson's ratio is increased; in the groundwater layer area, the P-wave velocity and density are increased and the Poisson's ratio is reduced. The local grid encryption technology is used to finely grid the abnormal area to improve the resolution of the local model. The elastic parameter model containing local anomalies is smoothed and processed for consistency. The model is smoothed using methods such as Gaussian filtering to eliminate numerical noise and discontinuities in the model. At the same time, the model is checked for consistency to ensure that the physical parameters of the model are geologically reasonable and consistent with the initial geological model, and finally a fine elastic model is obtained.

[0045] Preferably, step S2 comprises the following steps: Step S21: setting forward modeling parameters according to the spatial scale and target accuracy of the refined elastic model to obtain a forward modeling parameter table; Step S22: setting the earthquake source position for the refined elastic model and performing earthquake source wavelet excitation to obtain an earthquake source excitation model; Step S23: performing multi-angle joint wave field simulation according to the earthquake source excitation model and the forward parameter table to obtain a multi-angle joint wave field; Step S24: setting the detection point position for the fine elastic model, and recording the time series data of the multi-angle joint wave field at the detection point to obtain multi-angle detection point data; Step S25: converting the multi-angle detection point data into a seismic data format to obtain multi-angle detection point format data; Step S26: performing simulation data preprocessing on the multi-angle joint wave field according to the multi-angle detection point format data to obtain multi-angle reflection wave field data.

[0046] In the embodiment of the present invention, the spatial grid size of the forward simulation is determined according to the spatial scale of the fine elastic model. For example, when the model scale is 1000m 1000m, when the target accuracy is 1m, set the grid size to 1m 1m. According to the target accuracy and simulation frequency range, determine the time sampling interval of the forward simulation. For example, when the maximum simulation frequency is 50Hz, set the time sampling interval to 0.001s. According to the research objectives and the geological conditions of the simulation area, determine the frequency band range of the forward simulation. For example, when the research target is a shallow gas reservoir, set the frequency band range to 10-50Hz. Record parameters such as spatial grid size, time sampling interval, and frequency band range in the forward parameter table. In addition, other parameters such as simulation duration and boundary conditions need to be set according to the specific situation.

[0047] According to the actual seismic acquisition parameters, the source position is set on the surface of the fine elastic model. For example, the source points are set in an equally spaced grid with a spacing of 10 meters. The Ricker wavelet is selected as the source wavelet, and the main frequency of the Ricker wavelet is adjusted according to the frequency band range set in the forward parameter table. For example, the main frequency is set to 30Hz. The source wavelet is combined with the source position information to generate a source excitation model. The source excitation model contains the spatial position, excitation time and source wavelet information of each source point.

[0048] Based on the source excitation model and the forward parameter table, the finite difference method is used to perform multi-angle joint wave field simulation. The acoustic wave equation is numerically solved using the staggered grid finite difference algorithm to simulate the propagation process of seismic waves in the fine elastic model. During the simulation, according to the source excitation model, seismic waves are excited at each source point, and the snapshot of the seismic wave field at each time step is recorded. In order to simulate incident waves at different angles, the relative relationship between the source position and the detector position is changed, and multiple forward simulations are performed to obtain joint wave fields at different angles. For example, wave fields with incident angles of 0°, 15°, 30°, and 45° are simulated.

[0049] According to the actual seismic acquisition parameters, the detection point position is set on the surface of the fine elastic model. For example, the detectors are set in an equidistant grid with a spacing of 10 meters. The time series data of the multi-angle joint wave field at each detection point is extracted to generate multi-angle detection point data. The data of each detection point contains the reflection wave field information of the incident waves at different angles.

[0050] Convert multi-angle receiver point data into seismic data format, such as SEG-Y format. During the conversion process, the data of each receiver point needs to be organized according to the shot gather channel number, and necessary header file information needs to be added, such as sampling rate, channel spacing, shot point coordinates, etc. The generated multi-angle receiver point format data meets the input requirements of seismic data processing software.

[0051] The multi-angle detection point format data is pre-processed by simulation data, for example, interference waves such as direct waves and surface waves are removed. Bandpass filtering is applied to remove noise and improve the signal-to-noise ratio, and the filter parameters are determined according to the frequency band range set in the forward modeling parameter table. Amplitude compensation is performed on the data to eliminate the spherical diffusion effect during seismic wave propagation. After pre-processing, multi-angle reflection wave field data is obtained for subsequent gas reservoir identification and AVO analysis.

[0052] Preferably, step S3 comprises the following steps: Step S31: performing multi-angle data fusion on the multi-angle reflection wave field data to obtain fused multi-angle data; performing adaptive filtering on the fused multi-angle data to obtain filtered multi-angle data; Step S32: performing local signal amplification on the filtered multi-angle data to obtain locally amplified data; Step S33: extracting gas reservoir identification attributes and groundwater layer identification attributes from the locally enlarged data to obtain gas reservoir identification attribute data and groundwater layer identification attribute data; Step S34: performing gas reservoir identification on the gas reservoir identification attribute data to obtain a preliminary result of gas reservoir identification; Step S35: performing groundwater layer identification on the groundwater layer identification attribute data to obtain a preliminary result of groundwater layer identification; Step S36: spatially locate and evaluate the properties of the abnormal area based on the preliminary results of gas reservoir identification and groundwater layer identification to obtain local abnormality identification results; Step S37: Extract the reflection wave features of the local abnormal area from the locally amplified data according to the local abnormality recognition result to obtain local reflection abnormality data.

[0053] As an example of the present invention, refer to Figure 2 As shown, in this example, step S3 includes: Step S31: performing multi-angle data fusion on the multi-angle reflection wave field data to obtain fused multi-angle data; performing adaptive filtering on the fused multi-angle data to obtain filtered multi-angle data; In an embodiment of the present invention, the multi-angle reflection wave field data obtained in step S2, such as the reflection wave field data of 0°, 15°, 30° and 45°, are fused using a multi-angle data fusion method based on weighted superposition. Different weight coefficients are set according to the signal-to-noise ratio and resolution of data at different angles. For example, angle data with a high signal-to-noise ratio is given a larger weight, and angle data with a low signal-to-noise ratio is given a smaller weight. The weighted multi-angle data are superimposed to obtain fused multi-angle data. The fused multi-angle data is denoised using an adaptive filtering method. The noise characteristics in the data are analyzed, and the parameters of the filter are adaptively adjusted. For example, the cutoff frequency and filtering strength of the filter are adjusted according to the local noise level, so as to effectively suppress the noise while retaining the effective signal to obtain the filtered multi-angle data.

[0054] Step S32: performing local signal amplification on the filtered multi-angle data to obtain locally amplified data; In an embodiment of the present invention, local signal amplification is performed on the filtered multi-angle data obtained in step S31. First, the target area to be amplified is determined based on known geological information or preliminary exploration results. Then, the amplification method based on time-varying gain is applied to the multi-angle data in the target area. According to the change in the strength of the signal in the target area, the gain coefficient is dynamically adjusted to amplify the weak signal, suppress the strong signal, highlight the local abnormal information, and obtain the locally amplified data. For example, at the suspected gas reservoir location, the data with an amplitude value lower than a certain threshold is amplified, and the amplification factor is dynamically adjusted according to the amplitude value.

[0055] Step S33: extracting gas reservoir identification attributes and groundwater layer identification attributes from the locally enlarged data to obtain gas reservoir identification attribute data and groundwater layer identification attribute data; In an embodiment of the present invention, the locally amplified data is subjected to extraction of gas reservoir identification attributes and groundwater layer identification attributes. For gas reservoir identification, attributes such as instantaneous amplitude, instantaneous frequency, AVO gradient, and AVO intercept are extracted. For groundwater layer identification, the longitudinal and transverse wave velocity ratio (Vp / Vs), the quality factor Q value, and the frequency attenuation attribute are extracted. When calculating the instantaneous attributes, the complex seismic trace analysis method is used to calculate the instantaneous amplitude, instantaneous frequency, instantaneous phase, etc. of each sampling point. When calculating the AVO attributes, the locally amplified data is divided according to different angles, and then the amplitude analysis is performed on the data at each angle to extract the AVO gradient and AVO intercept. When calculating the Vp / Vs, Q value, and frequency attenuation attributes, the methods based on velocity analysis, spectral ratio method, and attenuation gradient are used respectively. The extracted attribute data are stored as gas reservoir identification attribute data and groundwater layer identification attribute data respectively.

[0056] Step S34: performing gas reservoir identification on the gas reservoir identification attribute data to obtain a preliminary result of gas reservoir identification; In the embodiment of the present invention, gas reservoir identification is performed on gas reservoir identification attribute data. First, based on the intersection relationship between the AVO gradient and the intercept, it is determined whether there is an AVO anomaly. For example, when the AVO gradient is negative and the AVO intercept is positive, it indicates that a gas reservoir may exist. Then, the spatial distribution characteristics of the instantaneous amplitude and the instantaneous frequency are analyzed to identify abnormal areas related to the gas reservoir. For example, gas reservoirs usually exhibit low instantaneous amplitude and low instantaneous frequency. Finally, combined with the analysis results of the AVO attributes and the instantaneous attributes, a preliminary judgment of gas reservoir identification is performed to obtain a preliminary result of gas reservoir identification.

[0057] Step S35: performing groundwater layer identification on the groundwater layer identification attribute data to obtain a preliminary result of groundwater layer identification; In the embodiment of the present invention, groundwater layer identification is performed on groundwater layer identification attribute data. First, the spatial distribution characteristics of the Q value are analyzed. Low Q value anomalies may indicate the existence of groundwater layers. Then, the spatial distribution characteristics of the Vp / Vs ratio are analyzed. High Vp / Vs ratio anomalies may also indicate the existence of groundwater layers. Finally, combined with the analysis results of the Q value, Vp / Vs ratio and frequency attenuation attribute, a preliminary judgment of groundwater layer identification is performed to obtain a preliminary result of groundwater layer identification.

[0058] Step S36: spatially locate and evaluate the properties of the abnormal area based on the preliminary results of gas reservoir identification and groundwater layer identification to obtain local abnormality identification results; In the embodiment of the present invention, the abnormal area is spatially located based on the preliminary results of gas reservoir identification and the preliminary results of groundwater layer identification. For example, the positions of the identified gas reservoir and groundwater layer are marked on the seismic profile. The properties of the abnormal area are evaluated in combination with geological data and well logging information. For example, the lithology, porosity, permeability and other parameters of the abnormal area are analyzed to determine the scale and reserves of the gas reservoir or groundwater layer. The spatial position and property evaluation results of the abnormal area are integrated into the local abnormality identification results.

[0059] Step S37: extracting the reflection wave characteristics of the local abnormal area from the locally amplified data according to the local abnormality recognition result to obtain local reflection abnormality data; In the embodiment of the present invention, the reflection wave characteristics of the abnormal area in the locally amplified data are extracted according to the local anomaly identification result. For example, information such as the amplitude, frequency, phase, AVO gradient and AVO intercept of the abnormal area is extracted. The extracted reflection wave characteristic data is stored as local reflection anomaly data for subsequent dispersion correction and AVO inversion.

[0060] Preferably, step S34 includes the following steps: Step S341: performing AVO gradient and intercept calculation on the gas reservoir identification attribute data to obtain AVO gradient and intercept data; performing gradient intercept intersection analysis based on the AVO gradient and intercept data to obtain AVO attribute analysis results; Step S342: performing instantaneous attribute calculation on the locally amplified data according to the gas reservoir identification attribute data to obtain instantaneous amplitude and instantaneous frequency data; performing spatial distribution characteristic analysis on the instantaneous amplitude and instantaneous frequency data to obtain instantaneous attribute analysis results; Step S343: performing multi-attribute intersection analysis on the AVO attribute analysis result and the instantaneous attribute analysis result to obtain a multi-attribute intersection analysis result; Step S344: Calculate the probability of each data point belonging to a gas reservoir based on the multi-attribute intersection analysis result to obtain a multi-attribute comprehensive analysis result; Step S345: Perform preliminary judgment on gas reservoir identification based on the multi-attribute comprehensive analysis results and the preset probability threshold to obtain preliminary gas reservoir identification results.

[0061] In an embodiment of the present invention, AVO gradient and intercept calculations are performed on multi-angle reflection wave field data in gas reservoir identification attribute data. First, the reflection wave field data at different angles (for example, 0°, 15°, 30°, and 45°) are extracted respectively. Then, the amplitude analysis of the seismic trace set is performed on the data at each angle. The linear relationship between the reflection coefficient and the sine square of the incident angle is fitted using the least squares method to obtain the AVO gradient and intercept data. The calculated AVO gradient and intercept data are stored as a gradient data volume and an intercept data volume, respectively. The gradient data volume and the intercept data volume are subjected to intersection analysis, for example, data points are plotted on a gradient-intercept intersection diagram, the distribution pattern of the data points is analyzed, and it is determined whether there is an AVO anomaly related to the gas reservoir. The analysis results are recorded in the AVO attribute analysis results.

[0062] Calculate the instantaneous attributes of the locally amplified data. Use Hilbert transform to calculate the complex seismic trace of the locally amplified data. Then, calculate the instantaneous amplitude and instantaneous frequency of each sampling point based on the real and imaginary parts of the complex seismic trace to obtain the instantaneous amplitude and instantaneous frequency data. Perform spatial distribution characteristic analysis on the calculated instantaneous amplitude and instantaneous frequency data. For example, draw a spatial distribution diagram of the instantaneous amplitude and instantaneous frequency, and analyze its variation patterns in the plane and section. Gas reservoirs usually show low instantaneous amplitude and low instantaneous frequency anomalies. Record the analysis results in the instantaneous attribute analysis results.

[0063] The AVO attribute analysis results obtained in step S341 and the instantaneous attribute analysis results obtained in step S342 are subjected to multi-attribute intersection analysis. For example, the AVO gradient-intercept intersection diagram is superimposed and displayed with the spatial distribution diagram of the instantaneous amplitude and instantaneous frequency to analyze the correlation between different attributes. Alternatively, the values ​​of different attributes are cross-plotted, for example, a cross plot of the instantaneous amplitude and the AVO gradient is plotted to analyze the distribution pattern of the data points. The results of the multi-attribute intersection analysis are recorded in the multi-attribute intersection analysis results.

[0064] According to the results of multi-attribute intersection analysis, the probability of each data point belonging to a gas reservoir is calculated. For example, a Bayesian classifier is used to calculate the probability of each data point belonging to a gas reservoir based on the combined characteristics of AVO attributes and instantaneous attributes. Alternatively, a machine learning method such as a support vector machine is used to train a classification model based on the results of multi-attribute intersection analysis, and then the trained model is used to predict the probability of each data point belonging to a gas reservoir. The calculated probability value is stored as a multi-attribute comprehensive analysis result.

[0065] A preliminary judgment on gas reservoir identification is made based on the results of multi-attribute comprehensive analysis and the preset probability threshold. For example, the probability threshold is set to 0.5. When the probability that a data point belongs to a gas reservoir is greater than or equal to 0.5, the data point is judged to belong to a gas reservoir. The judgment result is recorded in the preliminary result of gas reservoir identification for subsequent spatial positioning and property evaluation of abnormal areas.

[0066] Preferably, step S35 includes the following steps: Step S351: Calculate the Q value of the target area for the groundwater layer identification attribute data to obtain the Q value of the target area; use the preset Q value threshold to identify the spatial distribution anomaly of the Q value of the target area to obtain the Q value analysis result; Step S352: Perform velocity analysis on the groundwater layer identification attribute data, and calculate the ratio of the longitudinal wave velocity to the transverse wave velocity to obtain Vp / Vs ratio data; use a preset Vp / Vs ratio threshold to perform spatial distribution anomaly identification on the Vp / Vs ratio data to obtain a Vp / Vs ratio analysis result, where Vp is the longitudinal wave velocity and Vs is the transverse wave velocity; Step S353: performing frequency attenuation analysis on the groundwater layer identification attribute data to obtain a frequency attenuation analysis result; Step S354: performing a multi-attribute superposition analysis according to the Q value analysis result, the Vp / Vs ratio analysis result, and the frequency attenuation analysis result to obtain a multi-attribute superposition analysis result; Step S355: Generate preliminary results of groundwater layer identification based on the multi-attribute superposition analysis results to obtain preliminary results of groundwater layer identification.

[0067] In an embodiment of the present invention, the Q value of the target area is calculated for the groundwater layer identification attribute data. The Q value calculation method based on the spectral ratio method is used to calculate the Q value of the target area. First, the groundwater layer identification attribute data is subjected to spectral analysis to obtain the amplitude spectrum of different frequency components. Then, two amplitude spectra of different frequencies are selected, their ratio is calculated, and the Q value is calculated based on the frequency difference and the spectral ratio. The calculated Q value is assigned to the corresponding spatial position to obtain the Q value of the target area. Using a preset Q value threshold, such as 30, the spatial distribution anomaly of the Q value of the target area is identified. The area with a Q value lower than the threshold is marked as a potential groundwater layer abnormal area, and the analysis result is recorded in the Q value analysis result.

[0068] Velocity analysis is performed on the groundwater layer identification attribute data. The semblance-based velocity analysis method is used to calculate the longitudinal wave velocity and transverse wave velocity of the target area. First, a time window is selected for the groundwater layer identification attribute data, and then a velocity scan is performed within the time window to find the velocity with the largest semblance value as the longitudinal wave velocity or transverse wave velocity at that location. The calculated longitudinal wave velocity and transverse wave velocity are stored as a longitudinal wave velocity data body and a transverse wave velocity data body, respectively. The longitudinal and transverse wave velocity ratio (Vp / Vs) is calculated using the longitudinal wave velocity data body and the transverse wave velocity data body. The Vp / Vs ratio data is compared with a preset Vp / Vs ratio threshold, such as 1.8. Areas with Vp / Vs ratios higher than the threshold are marked as potential groundwater layer anomaly areas, and the analysis results are recorded in the Vp / Vs ratio analysis results.

[0069] Perform frequency attenuation analysis on groundwater layer identification attribute data. First, perform spectrum decomposition on groundwater layer identification attribute data to obtain seismic data of different frequencies. Then, analyze the amplitude attenuation law of seismic data of different frequencies. For example, calculate the attenuation gradient of the amplitude of seismic data of different frequencies with propagation distance or time. Groundwater layers usually show the characteristic of faster attenuation of high-frequency components. Record the frequency attenuation analysis results in the frequency attenuation analysis results.

[0070] Perform multi-attribute superposition analysis on the Q value analysis results, Vp / Vs ratio analysis results, and frequency attenuation analysis results. Superimpose the spatial distribution maps of different attributes, for example, superimpose low Q value abnormal areas, high Vp / Vs ratio abnormal areas, and high-frequency attenuation abnormal areas on the seismic profile or plane map. Through superposition analysis, identify areas that simultaneously meet the abnormal characteristics of multiple attributes to improve the reliability of groundwater layer identification. Record the multi-attribute superposition analysis results in the multi-attribute superposition analysis results.

[0071] Based on the results of multi-attribute superposition analysis, the preliminary results of groundwater layer identification are generated. Based on the superposition relationship of different attributes, it is judged whether there is a groundwater layer. For example, when a certain area meets the low Q value, high Vp / Vs ratio and high-frequency attenuation anomaly at the same time, the area is judged to be a potential groundwater layer. The judgment results are recorded in the preliminary results of groundwater layer identification for subsequent spatial positioning and property evaluation of abnormal areas.

[0072] Preferably, step S4 comprises the following steps: Step S41: performing preliminary AVO attribute extraction on the local reflection anomaly data to obtain preliminary AVO attribute data; Step S42: performing multiple wave suppression processing on the local reflection anomaly data to obtain data after multiple wave suppression; Step S43: performing dispersion correction on the data after suppressing the multiple waves to obtain dispersion-corrected data; Step S44: performing multi-parameter AVO inversion according to the preliminary AVO attribute data, the dispersion-corrected data, and the local reflection anomaly data to obtain the inverted local AVO attribute data; Step S45: performing AVO attribute fine correction on the inverted AVO attribute data using the multi-angle reflection wave field data to obtain corrected AVO attribute data.

[0073] As an example of the present invention, refer to Figure 3 As shown, in this example, step S4 includes: Step S41: performing preliminary AVO attribute extraction on the local reflection anomaly data to obtain preliminary AVO attribute data; In an embodiment of the present invention, the AVO attributes of the local reflection anomaly data are initially extracted. The local reflection anomaly data are divided according to the incident wave fields at different angles (e.g., 0°, 15°, 30°, and 45°). The data at each angle are respectively subjected to amplitude analysis to extract the peak amplitude or root mean square amplitude of the seismic trace. The amplitude data at different angles are fitted with the corresponding incident angles, for example, the relationship between the reflection coefficient and the sine square of the incident angle is fitted using the least squares method. The slope and intercept obtained by fitting are the preliminary AVO gradient and AVO intercept attributes. The calculated AVO gradient and AVO intercept attributes are stored as preliminary AVO attribute data.

[0074] Step S42: performing multiple wave suppression processing on the local reflection anomaly data to obtain data after multiple wave suppression; In an embodiment of the present invention, multiple wave suppression processing is performed on local reflection anomaly data. A multiple wave suppression method based on predictive deconvolution is adopted. First, velocity analysis is performed on the local reflection anomaly data to obtain velocity information of the underground medium. Then, a prediction operator is established based on the velocity information to predict the arrival time and waveform of the multiple waves. The predicted multiple waves are subtracted from the original data to obtain data after suppressing the multiple waves. In order to avoid excessive suppression, it is necessary to adjust the parameters of the prediction operator according to the actual data situation, such as the prediction step size and the filter length.

[0075] Step S43: performing dispersion correction on the data after suppressing the multiple waves to obtain dispersion-corrected data; In an embodiment of the present invention, dispersion correction is performed on the data after suppressing multiple waves. A dispersion correction method based on inverse Q filtering is adopted. First, the Q value of the study area is estimated based on well logging data or other geological data. Then, the attenuation coefficient of the seismic wave is calculated based on the Q value and the propagation time of the seismic wave. The data after suppressing multiple waves is compensated by using an inverse Q filter to restore the high-frequency components of the seismic wave, thereby correcting the dispersion effect. The corrected data is stored as dispersion-corrected data.

[0076] Step S44: performing multi-parameter AVO inversion according to the preliminary AVO attribute data, the dispersion-corrected data, and the local reflection anomaly data to obtain the inverted local AVO attribute data; In an embodiment of the present invention, multi-parameter AVO inversion is performed based on preliminary AVO attribute data, data after correction for dispersion, and local reflection anomaly data. A multi-parameter AVO inversion method based on Bayesian theory is adopted. First, an initial elastic parameter model is established based on preliminary AVO attribute data, including longitudinal wave velocity, shear wave velocity, and density. Then, an AVO forward model is constructed using the data after correction for dispersion and local reflection anomaly data. Elastic parameters such as longitudinal wave velocity, shear wave velocity, and density are inverted by an iterative optimization algorithm, such as a conjugate gradient method or a simulated annealing algorithm. The elastic parameters obtained by inversion are converted into AVO attributes, such as gradients and intercepts, to obtain local AVO attribute data after inversion.

[0077] Step S45: performing AVO attribute fine correction on the inverted AVO attribute data using the multi-angle reflection wave field data to obtain corrected AVO attribute data; In an embodiment of the present invention, the inverted AVO attribute data is finely corrected using multi-angle reflection wave field data. The multi-angle reflection wave field data is compared and analyzed with the inverted AVO attribute data. For example, the difference or correlation coefficient between the two is calculated. According to the comparative analysis results, the inverted AVO attribute data is finely corrected. For example, the weighted average method is used to merge the AVO information in the multi-angle reflection wave field data into the inverted AVO attribute data. Alternatively, a model-based update method is used to correct the inverted elastic parameter model according to the multi-angle reflection wave field data, and then the AVO attributes are recalculated. The corrected AVO attribute data is stored as the corrected AVO attribute data.

[0078] Preferably, step S44 includes the following steps: Step S441: Selecting a local reflection anomaly area according to the dispersion-corrected data and the local reflection anomaly data to obtain local reflection anomaly selection data; Step S442: extracting local AVO response from the local reflection anomaly selection data according to the preliminary AVO attribute data to obtain local AVO response data; Step S443: performing multi-parameter initialization according to the local AVO response data to obtain a multi-parameter initial model; Step S444: constructing a local AVO inversion model according to the multi-parameter initial model to obtain a local AVO inversion model; Step S445: Perform multi-parameter joint inversion on the local AVO inversion model to obtain inverted local AVO attribute data.

[0079] In an embodiment of the present invention, the local reflection anomaly region is selected based on the dispersion-corrected data and the local reflection anomaly data. First, in the local reflection anomaly data, the region with strong amplitude energy and obvious AVO anomaly is identified. For example, by setting an amplitude threshold or a gradient threshold, the data points that meet the conditions are screened out. Then, the reliability of the abnormal region is further confirmed in combination with the dispersion-corrected data. For example, it is checked whether the dispersion-corrected data has obvious reflection features in the abnormal region. The data of the abnormal region finally determined is extracted as the local reflection anomaly selection data.

[0080] The local AVO response is extracted from the local reflection anomaly selection data according to the preliminary AVO attribute data. The local reflection anomaly selection data is divided according to the incident wave fields at different angles (such as 0°, 15°, 30° and 45°). For the data at each angle, its amplitude information is extracted, such as the peak amplitude or the root mean square amplitude. The amplitude information at different angles is combined with the corresponding incident angle to form a data pair to construct the local AVO response data. The local AVO response data reflects the AVO characteristics of the local reflection anomaly area.

[0081] Perform multi-parameter initialization based on local AVO response data. Use logging data or existing geological data to obtain background elastic parameter information of the study area, such as P-wave velocity, S-wave velocity, and density. Use this background information as the parameters of the initial model. Fine-tune the parameters of the initial model based on the local AVO response data. For example, adjust the ratio of P-wave velocity to S-wave velocity of the initial model based on the gradient and intercept information of the AVO response. Use the adjusted parameters as the multi-parameter initial model.

[0082] The local AVO inversion model is constructed based on the multi-parameter initial model. The multi-parameter initial model is used as the initial input for the local AVO inversion. The spatial range of the data is selected according to the local reflection anomaly, and the size and location of the inversion area are determined. The inversion parameters are set, such as the inversion algorithm, the number of iterations, and the convergence condition. This information is used together with the multi-parameter initial model to construct the local AVO inversion model.

[0083] Perform multi-parameter joint inversion on the local AVO inversion model. Use an inversion algorithm based on iterative optimization, such as the conjugate gradient method or the Gauss-Newton method. In each iteration, use the local AVO inversion model to perform forward simulation and calculate the residual between the simulated data and the local reflection anomaly selection data. Update the elastic parameters in the local AVO inversion model based on the residual information. Iterate until the residual meets the preset convergence condition. Convert the elastic parameters obtained by the final inversion into AVO attributes, such as gradient and intercept, to obtain the inverted local AVO attribute data.

[0084] Preferably, step S445 is specifically as follows: The local AVO response residual is calculated according to the local AVO inversion model and the local AVO response data to obtain the AVO residual data; Perform model parameter perturbation on the multi-parameter initial model to obtain a parameter perturbation model; perform local forward simulation on the parameter perturbation model to obtain local seismic data; Based on the local seismic data and the multi-parameter initial model, the difference calculation is performed to obtain the model difference; based on the model difference and the parameter perturbation model, the sensitivity calculation is performed to obtain the sensitivity matrix; The sensitivity matrix and AVO residual data are used to calculate the parameter update direction to obtain the parameter update direction data; Perform multi-parameter update on the local AVO inversion model according to the parameter update direction data and the preset step size to obtain a multi-parameter update model; The local AVO attribute data is generated according to the multi-parameter updating model to obtain the inverted local AVO attribute data.

[0085] In the embodiment of the present invention, the local AVO inversion model is used for forward simulation to obtain simulated local AVO response data. The simulated local AVO response data and the actual observed local AVO response data are difference calculated to obtain AVO residual data. The AVO residual data reflects the difference between the inversion model and the actual geological conditions.

[0086] The elastic parameters in the multi-parameter initial model, such as P-wave velocity, S-wave velocity and density, are slightly perturbed. The perturbation method can be random perturbation or gradient perturbation. The perturbation amplitude needs to be set according to the value range of the model parameters and the inversion accuracy. The perturbed model parameters are constructed as a parameter perturbation model.

[0087] The parameter perturbation model is used for local forward modeling. The forward modeling method can use the finite difference method or the ray tracing method. The simulated seismic data includes the influence of the perturbation parameters on the propagation of seismic waves.

[0088] The local seismic data simulated by the parameter perturbation model and the local seismic data simulated by the multi-parameter initial model are calculated to obtain the model difference, which reflects the influence of the model parameter perturbation on the propagation of seismic waves.

[0089] The model difference and the corresponding parameter disturbance are calculated to obtain the sensitivity matrix. The sensitivity matrix reflects the influence of different model parameters on seismic wave propagation. The calculation method can be the difference method or the adjoint state method.

[0090] Using the sensitivity matrix and the AVO residual data, the parameter update direction is calculated. For example, the steepest descent method or the conjugate gradient method is used. The parameter update direction indicates how to adjust the model parameters to reduce the AVO residual.

[0091] The elastic parameters in the local AVO inversion model are updated according to the calculated parameter update direction and the preset step size. The step size needs to be adjusted according to the convergence speed and stability of the inversion. The updated parameters are constructed as a multi-parameter update model.

[0092] The elastic parameters in the multi-parameter update model are converted into AVO attributes, such as gradient and intercept. The conversion method can adopt the Zoeppritz equation or its approximate formula. The converted AVO attributes are used as the local AVO attribute data after inversion.

[0093] Preferably, step S5 comprises the following steps: Step S51: acquiring actual seismic data; performing actual seismic data preprocessing on the actual seismic data to obtain preprocessed actual seismic data; Step S52: extracting AVO attributes from the pre-processed actual seismic data to obtain actual AVO attribute data; Step S53: performing model data comparison and error analysis on the corrected AVO attribute data and the actual AVO attribute data to obtain a model data comparison result and an error analysis report; Step S54: adjusting and optimizing the model parameters of the fine elastic model according to the model data comparison result and the error analysis report to obtain an optimized geological model; Step S55: Perform model verification on the optimized geological model to obtain the optimized geological and AVO models.

[0094] In an embodiment of the present invention, actual seismic data is obtained from a seismic data acquisition center. The actual seismic data is in SEG-Y format and contains seismic wave reflection information of the study area. The actual seismic data is preprocessed, including processing steps such as noise removal, static correction, dynamic correction and offset. The denoising process uses the fx domain denoising method to suppress interference such as random noise and surface waves. The static correction uses the surface consistency static correction method to eliminate the influence of surface elevation and near-surface velocity changes. The dynamic correction uses velocity analysis and NMO correction methods to eliminate the influence of underground structures. The offset processing uses prestack time offset or prestack depth offset method to image the seismic data to the correct spatial position. The preprocessed seismic data is stored as preprocessed actual seismic data.

[0095] AVO attributes are extracted from the actual preprocessed seismic data. The actual preprocessed seismic data is divided according to the incident wave fields at different angles (such as 0°, 15°, 30° and 45°). Amplitude analysis is performed on the data at each angle to extract the peak amplitude or root mean square amplitude of the seismic trace. The amplitude data at different angles are fitted with the corresponding incident angles. For example, the relationship between the reflection coefficient and the sine square of the incident angle is fitted using the least squares method. The slope and intercept obtained by fitting are the actual AVO gradient and AVO intercept attributes, which are stored as actual AVO attribute data.

[0096] Perform model data comparison and error analysis on the corrected AVO attribute data and actual AVO attribute data. Match the corrected AVO attribute data with the actual AVO attribute data in spatial position. Calculate the difference or relative error between the two to obtain the model data comparison result. For example, calculate the difference between the corrected AVO gradient and the actual AVO gradient, and the difference between the corrected AVO intercept and the actual AVO intercept. Perform statistical analysis on the error, for example, calculate the mean, variance and standard deviation of the error, draw a histogram and spatial distribution diagram of the error, and generate an error analysis report.

[0097] Adjust and optimize the model parameters of the refined elastic model according to the model data comparison results and error analysis report. If the model data comparison results show that the error is large, the refined elastic model needs to be adjusted according to the error analysis report. For example, if the AVO gradient error is large, the P-wave velocity ratio in the refined elastic model needs to be adjusted. If the AVO intercept error is large, the density or impedance in the refined elastic model needs to be adjusted. The adjustment method can be manual adjustment or automatic inversion. Update the adjusted model parameters to the refined elastic model to obtain an optimized geological model.

[0098] Perform model verification on the optimized geological model. Perform forward simulation using the optimized geological model to generate simulated AVO attribute data. Compare and analyze the simulated AVO attribute data with the actual AVO attribute data to evaluate the accuracy and reliability of the optimized model. If the simulation results are consistent with the actual data, the optimized geological model is considered to be valid. The optimized geological model and the corresponding AVO attribute data are stored as the optimized geological and AVO model.

[0099] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0100] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A shallow AVO response characteristic correction method based on forward simulation, characterized in that: The following steps are involved: Step S1: acquiring well logging data, shallow penetration profiles and seismic data; constructing an initial geological model based on the well logging data, shallow penetration profiles and seismic data to obtain an initial geological model; performing high-resolution elastic parameter analysis based on the initial geological model to obtain a refined elastic model; Step S2: performing multi-angle joint wave field simulation according to the refined elastic model to obtain a multi-angle joint wave field; performing simulation data preprocessing on the multi-angle joint wave field, and performing multi-angle reflection wave field extraction to obtain multi-angle reflection wave field data; Step S3: Perform gas reservoir identification on the multi-angle reflection wave field data to obtain preliminary gas reservoir identification results; The groundwater layer is identified by multi-angle reflection wave field data, and the preliminary results of groundwater layer identification are obtained; Based on the preliminary results of gas reservoir identification and groundwater layer identification, local reflection wave characteristics are extracted to obtain local reflection anomaly data; Step S4: performing dispersion correction on the local reflection anomaly data to obtain dispersion-corrected data; performing multi-parameter AVO inversion based on the dispersion-corrected data and the local reflection anomaly data to obtain inverted local AVO attribute data; performing AVO attribute fine correction on the inverted AVO attribute data using multi-angle reflection wave field data to obtain corrected AVO attribute data; Step S5: obtaining actual earthquake data; The inversion model error analysis is performed on the corrected AVO attribute data according to the actual seismic data to obtain the model data comparison results and error analysis report; the model is optimized and verified according to the model data comparison results and error analysis report to obtain the optimized geological and AVO models.

2. The shallow AVO response characteristic correction method based on forward modeling according to claim 1 is characterized in that: Step S1 includes the following steps: Step S11: acquiring well logging data, shallow penetration profiles and seismic data; performing data preprocessing on the well logging data, shallow penetration profiles and seismic data, and performing data quality control to obtain preprocessed multi-source data; Step S12: performing stratigraphic division according to the pre-processed multi-source data, and performing stratigraphic comparison and correction processing to obtain layered data and stratigraphic information; Step S13: constructing an initial geological model using the layered data and the stratigraphic information to obtain an initial geological model; Step S14: constructing a high-resolution elastic parameter model according to the initial geological model to obtain an elastic parameter model; constraining and correcting the elastic parameter model according to the initial geological model to obtain a multi-scale elastic parameter model; Step S15: finely characterize the local anomalies of the multi-scale elastic parameter model to obtain an elastic parameter model containing local anomalies; perform model smoothing and consistency processing on the elastic parameter model containing local anomalies to obtain a fine elastic model.

3. The shallow AVO response characteristic correction method based on forward modeling according to claim 1 is characterized in that: Step S2 includes the following steps: Step S21: setting forward modeling parameters according to the spatial scale and target accuracy of the refined elastic model to obtain a forward modeling parameter table; Step S22: setting the earthquake source position for the refined elastic model and performing earthquake source wavelet excitation to obtain an earthquake source excitation model; Step S23: performing multi-angle joint wave field simulation according to the earthquake source excitation model and the forward parameter table to obtain a multi-angle joint wave field; Step S24: setting the detection point position for the fine elastic model, and recording the time series data of the multi-angle joint wave field at the detection point to obtain multi-angle detection point data; Step S25: converting the multi-angle detection point data into a seismic data format to obtain multi-angle detection point format data; Step S26: performing simulation data preprocessing on the multi-angle joint wave field according to the multi-angle detection point format data to obtain multi-angle reflection wave field data.

4. The shallow AVO response characteristic correction method based on forward modeling according to claim 1 is characterized in that: Step S3 includes the following steps: Step S31: performing multi-angle data fusion on the multi-angle reflection wave field data to obtain fused multi-angle data; performing adaptive filtering on the fused multi-angle data to obtain filtered multi-angle data; Step S32: performing local signal amplification on the filtered multi-angle data to obtain locally amplified data; Step S33: extracting gas reservoir identification attributes and groundwater layer identification attributes from the locally enlarged data to obtain gas reservoir identification attribute data and groundwater layer identification attribute data; Step S34: performing gas reservoir identification on the gas reservoir identification attribute data to obtain a preliminary result of gas reservoir identification; Step S35: performing groundwater layer identification on the groundwater layer identification attribute data to obtain a preliminary result of groundwater layer identification; Step S36: spatially locate and evaluate the properties of the abnormal area based on the preliminary results of gas reservoir identification and groundwater layer identification to obtain local abnormality identification results; Step S37: Extract the reflection wave features of the local abnormal area from the locally amplified data according to the local abnormality recognition result to obtain local reflection abnormality data.

5. The shallow AVO response characteristic correction method based on forward simulation according to claim 4 is characterized in that: Step S34 includes the following steps: Step S341: performing AVO gradient and intercept calculation on the gas reservoir identification attribute data to obtain AVO gradient and intercept data; performing gradient intercept intersection analysis based on the AVO gradient and intercept data to obtain AVO attribute analysis results; Step S342: performing instantaneous attribute calculation on the locally amplified data according to the gas reservoir identification attribute data to obtain instantaneous amplitude and instantaneous frequency data; performing spatial distribution characteristic analysis on the instantaneous amplitude and instantaneous frequency data to obtain instantaneous attribute analysis results; Step S343: performing multi-attribute intersection analysis on the AVO attribute analysis result and the instantaneous attribute analysis result to obtain a multi-attribute intersection analysis result; Step S344: Calculate the probability of each data point belonging to a gas reservoir based on the multi-attribute intersection analysis result to obtain a multi-attribute comprehensive analysis result; Step S345: Perform preliminary judgment on gas reservoir identification based on the multi-attribute comprehensive analysis results and the preset probability threshold to obtain preliminary gas reservoir identification results.

6. The shallow AVO response characteristic correction method based on forward modeling according to claim 4 is characterized in that: Step S35 includes the following steps: Step S351: Calculate the Q value of the target area for the groundwater layer identification attribute data to obtain the Q value of the target area; use the preset Q value threshold to identify the spatial distribution anomaly of the Q value of the target area to obtain the Q value analysis result; Step S352: Perform velocity analysis on the groundwater layer identification attribute data, and calculate the ratio of the longitudinal wave velocity to the transverse wave velocity to obtain Vp / Vs ratio data; use a preset Vp / Vs ratio threshold to perform spatial distribution anomaly identification on the Vp / Vs ratio data to obtain a Vp / Vs ratio analysis result, where Vp is the longitudinal wave velocity and Vs is the transverse wave velocity; Step S353: performing frequency attenuation analysis on the groundwater layer identification attribute data to obtain a frequency attenuation analysis result; Step S354: performing a multi-attribute superposition analysis according to the Q value analysis result, the Vp / Vs ratio analysis result, and the frequency attenuation analysis result to obtain a multi-attribute superposition analysis result; Step S355: Generate preliminary results of groundwater layer identification based on the multi-attribute superposition analysis results to obtain preliminary results of groundwater layer identification.

7. The shallow AVO response characteristic correction method based on forward simulation according to claim 1 is characterized in that: Step S4 includes the following steps: Step S41: performing preliminary AVO attribute extraction on the local reflection anomaly data to obtain preliminary AVO attribute data; Step S42: performing multiple wave suppression processing on the local reflection anomaly data to obtain data after multiple wave suppression; Step S43: performing dispersion correction on the data after suppressing the multiple waves to obtain dispersion-corrected data; Step S44: performing multi-parameter AVO inversion according to the preliminary AVO attribute data, the dispersion-corrected data, and the local reflection anomaly data to obtain the inverted local AVO attribute data; Step S45: performing AVO attribute fine correction on the inverted AVO attribute data using the multi-angle reflection wave field data to obtain corrected AVO attribute data.

8. The shallow AVO response characteristic correction method based on forward simulation according to claim 7 is characterized in that: Step S44 includes the following steps: Step S441: Selecting a local reflection anomaly region according to the dispersion-corrected data and the local reflection anomaly data to obtain local reflection anomaly selection data; Step S442: extracting local AVO response from the local reflection anomaly selection data according to the preliminary AVO attribute data to obtain local AVO response data; Step S443: performing multi-parameter initialization according to the local AVO response data to obtain a multi-parameter initial model; Step S444: constructing a local AVO inversion model according to the multi-parameter initial model to obtain a local AVO inversion model; Step S445: Perform multi-parameter joint inversion on the local AVO inversion model to obtain inverted local AVO attribute data.

9. The shallow AVO response characteristic correction method based on forward modeling according to claim 8 is characterized in that: Step S445 is specifically as follows: The local AVO response residual is calculated according to the local AVO inversion model and the local AVO response data to obtain the AVO residual data; Perform model parameter perturbation on the multi-parameter initial model to obtain a parameter perturbation model; perform local forward simulation on the parameter perturbation model to obtain local seismic data; Perform difference calculation based on local seismic data and multi-parameter initial model to obtain model difference; perform sensitivity calculation based on model difference and parameter perturbation model to obtain sensitivity matrix; The sensitivity matrix and AVO residual data are used to calculate the parameter update direction to obtain the parameter update direction data; Perform multi-parameter update on the local AVO inversion model according to the parameter update direction data and the preset step size to obtain a multi-parameter update model; The local AVO attribute data is generated according to the multi-parameter updating model to obtain the inverted local AVO attribute data.

10. The shallow AVO response characteristic correction method based on forward simulation according to claim 1 is characterized in that: Step S5 includes the following steps: Step S51: acquiring actual seismic data; performing actual seismic data preprocessing on the actual seismic data to obtain preprocessed actual seismic data; Step S52: extracting AVO attributes from the pre-processed actual seismic data to obtain actual AVO attribute data; Step S53: performing model data comparison and error analysis on the corrected AVO attribute data and the actual AVO attribute data to obtain a model data comparison result and an error analysis report; Step S54: adjusting and optimizing the model parameters of the fine elastic model according to the model data comparison result and the error analysis report to obtain an optimized geological model; Step S55: Perform model verification on the optimized geological model to obtain the optimized geological and AVO models.

Images