CN115857047B

CN115857047B - Comprehensive prediction method for earthquake reservoir

Info

Publication number: CN115857047B
Application number: CN202211192814.0A
Authority: CN
Inventors: 张旭升; 张生郡; 赵海山; 曹连宇; 魏恒飞; 陈彦虎; 毕建军
Original assignee: Beijing Zhongheng Lihua Petroleum Technology Research Institute
Current assignee: Beijing Zhongheng Lihua Petroleum Technology Research Institute
Priority date: 2022-09-28
Filing date: 2022-09-28
Publication date: 2023-07-21
Anticipated expiration: 2042-09-28
Also published as: CN115857047A

Abstract

The invention discloses a comprehensive prediction method of an earthquake reservoir, which relates to the technical field of earthquake reservoir prediction, and comprises the steps of firstly extracting an earthquake attribute value from a target interval of the reservoir and optimizing sensitive earthquake attribute, carrying out core observation and rock phase analysis based on logging and coring data of the reservoir, and obtaining sedimentary facies types and distribution characteristics by combining logging data interpretation results and earthquake data comparison data; performing environment correction and standardization treatment on logging data, then performing petrophysical analysis, selecting sensitive parameters, and performing prestack poststack inversion on each reservoir; performing pre-stack post-stack crack prediction based on a logging result graph; performing prestack poststack gas-containing detection and seismic energy attenuation attribute extraction on the reservoir based on the sensitive seismic attribute; and finally, determining the boundary of the effective reservoir and the spreading rule of the effective reservoir according to the processing result, and performing classification evaluation according to the effective reservoir. The invention can rapidly identify the spreading rule of the effective reservoir, and improves the classification evaluation efficiency of the effective reservoir.

Description

Comprehensive prediction method for earthquake reservoir

Technical Field

The invention relates to the technical field of earthquake reservoir prediction, in particular to a comprehensive earthquake reservoir prediction method.

Background

Hydrocarbon reservoirs are areas of the formation where hydrocarbon accumulation exists underground in hydrocarbon exploration engineering. Reservoir characteristics include lithology, physical properties, oil and gas properties, and the like, which are also the primary directions for reservoir prediction. Reservoir lithology is a major feature describing reservoir mineral constituents, reflecting the reservoir properties and reservoir characteristics of the rock formations, and common parameters include reservoir rock physical structure, distribution range, reservoir top interface structure morphology, reservoir thickness, and the like. Reservoir physical properties are physical properties describing the reservoir, including physical parameter properties, physical spatial spread. Broadly, the skeletal properties, porosity, permeability, fluidics, thermal properties, conductivity, acoustic properties, radioactivity, and various sensitivities of reservoir rock are included. The narrow definition generally refers to the porosity and permeability of reservoir rock. Reservoir oil and gas properties primarily refer to the properties of fluids, fluid types, etc. within the reservoir. The oil-containing property of the reservoir is evaluated by using logging data and the like to find out the permeable layer and then evaluating the oil-containing property. The oil-gas property prediction can predict the earthquake phase and identify different layers such as an oil layer, a gas layer, a water layer, a dry layer and the like of the reservoir.

The prediction techniques that are currently traditionally applied include: attribute analysis, AVO and inversion, fracture prediction, petrophysical analysis, spectral decomposition, forward modeling, hydrocarbon detection based on absorption, attenuation, multi-wave and multi-component, well seismic, and the like. However, the existing reservoir prediction technology does not screen and analyze the seismic attribute of the reservoir, can not realize qualitative prediction of the reservoir, is easily affected by the environment in the reservoir inversion process, can not quickly identify an effective reservoir, and lacks a comprehensive seismic reservoir prediction technology.

Disclosure of Invention

The invention aims to overcome the defects of the prior art, provide a comprehensive prediction method for a seismic reservoir, and solve the problems that the prior reservoir prediction technology does not screen and analyze the seismic attribute of the reservoir, can not realize qualitative prediction of the reservoir, is easily influenced by environment in the reservoir inversion process, and can not quickly identify an effective reservoir.

The aim of the invention is realized by the following technical scheme:

a method for comprehensive prediction of a seismic reservoir, comprising the steps of:

step one: the method comprises the steps of seismic attribute analysis, namely selecting top and bottom layers of a reservoir as a fixed time window, extracting a seismic attribute value in a clastic rock deposition unit of a target interval, and optimizing a sedimentary facies sensitive seismic attribute and a gas-containing sensitive seismic attribute from the seismic attribute value;

Step two: carrying out core observation and rock phase analysis based on logging and coring data of a reservoir, and summarizing the type and distribution characteristics of the sedimentary facies of the reservoir by combining logging data interpretation results and seismic data comparison data;

step three: performing reservoir inversion, namely performing environment correction and standardization treatment on original logging data, performing petrophysical analysis on the basis of the treated logging data, selecting sensitive parameters, and performing pre-stack inversion and post-stack inversion on each reservoir by using an SMI inversion method;

step four: performing fracture prediction, namely performing FMI imaging logging on drilling of the reservoir, and after obtaining an electric imaging logging result diagram of each reservoir fracture, performing prestack fracture prediction and poststack fracture prediction on each reservoir respectively;

step five: performing gas-containing detection, namely performing prestack gas-containing detection and seismic energy attenuation attribute extraction on each gas layer group of the reservoir based on gas-containing sensitive seismic attributes;

step six: determining an effective reservoir boundary, and performing various information fusion predictions on the effective reservoir based on a pre-stack post-stack fracture prediction result, a pre-stack post-stack gas-containing detection result, a favorable phase zone characterization result and a pre-stack post-stack inversion result of the reservoir to determine the effective reservoir boundary and an effective reservoir spreading rule;

Step seven: and (3) carrying out classification evaluation on the effective reservoir, determining different types of reservoir logging classification evaluation standards according to single-well gas testing and generation data of the reservoir, and carrying out classification evaluation on the predicted effective reservoir after the fusion of various information.

Specifically, the first step specifically includes: selecting the top and bottom layer positions of the reservoir as fixed time windows to extract the seismic attribute values in the clastic rock deposition units of the target interval; according to the spreading characteristics of the reservoir on the plane, comparing and analyzing the extracted seismic attribute values, preferably identifying the sedimentary phase-sensitive seismic attribute and the gas-containing sensitive seismic attribute, taking the sedimentary phase-sensitive seismic attribute and the gas-containing sensitive seismic attribute as basic parameters, carrying out correlation analysis by combining the well reservoir characteristics, determining the seismic attribute with good correlation with the reservoir, and carrying out qualitative prediction of the favorable phase zone by utilizing the seismic attribute with good correlation with the reservoir.

Specifically, the second step specifically includes the following substeps:

s201, based on logging and coring data of a reservoir, observing and describing a rock core by a system, analyzing rock components, structures and structures, and combining logging data interpretation results and seismic data comparison data to induce sediment phase types and distribution characteristics, namely phase mark characteristics, of the reservoir;

S202, well-shock combined deposition microphase division, determining a material source and a deposition mode by taking a regional deposition phase as a direction, and guiding seismic attribute extraction by taking a well point phase as a basis; performing cluster analysis on the optimized sensitive seismic attribute from the extracted seismic attributes to obtain sensitive attribute seismic phases; and the reference well points are favorable for well-seismic combination of the reservoir, sensitive attribute seismic phases are converted into sedimentary phases, and the sedimentary phase spreading characteristics of the reservoir are determined.

Specifically, the third step specifically includes the following substeps:

s301, performing environment correction, namely performing point-by-point inspection and correction on a well logging curve which is greatly influenced by well diameter expansion and well wall irregularity in an original well logging curve of a reservoir by adopting a method of calculating an upper limit, so as to eliminate the influence of well diameter conditions;

s302, normalizing the well logging data, selecting a standard layer in the range of a reservoir operation area, performing secondary calibration on the well logging value of each well, uniformly calibrating all the well logging values by using the standard layer, and eliminating the systematic error among the well logging data;

s303, carrying out petrophysical analysis on the reservoir based on the environment corrected and standardized logging data, and optimizing and identifying a sand shale sensitive curve, physical inversion sensitive parameters and prestack gas-containing inversion sensitive parameters according to petrophysical analysis results; respectively establishing a porosity interpretation model, a permeability interpretation model and an interpretation graph plate to analyze logging data, and preferentially identifying favorable reservoir sensitive seismic attributes, effective reservoir sensitive seismic attributes, porosity sensitive curves and permeability sensitive curves;

S304, performing seismic waveform indication inversion on each reservoir by using SMI inversion software according to the sensitivity curve, the sensitivity parameter and the sensitivity seismic attribute which are preferably selected in the step S303, and obtaining a waveform indication inversion result of the reservoir.

Specifically, the fourth step specifically includes the following sub-steps:

s401, measuring each drilling crack of a reservoir by using an FMI imaging logging technology, and obtaining an electric imaging logging result diagram of each drilling crack of the reservoir;

s402, according to an electric imaging logging result diagram of the drilling fracture, pre-stack seismic information prediction is carried out on the reservoir by using an FRS fracture detection method, pre-stack anisotropy of the reservoir is calculated, and pre-stack fracture development characteristics of the reservoir are obtained;

s403, performing coherent analysis on an electric imaging logging result diagram of the drilling fracture by using a coherent fracture prediction method, and calculating a seismic coherent data volume of the reservoir to obtain the distribution characteristics of each coherent fracture of the reservoir;

s404, performing stress field numerical simulation analysis on the reservoir by taking the anticline structure as a mechanical model, taking the maximum curvature value of a point on the anticline structure surface as a criterion of the crack development degree of the point, and simultaneously indicating the possible crack trend by the minimum principal curvature direction to obtain the development degree and the spreading relation of the reservoir cracks.

Specifically, the fifth step specifically includes: carrying out pre-stack AVO analysis based on the gas-containing sensitive seismic attribute selected in the first step to obtain the distribution characteristics of a high-value region of the gas-containing probability of the reservoir pre-stack AVO; and carrying out post-stack gas-containing detection based on the gas-containing sensitive seismic attribute, extracting the attenuation gradient attribute of the high frequency band of the seismic wave, and obtaining the distribution characteristics of the high-value area of the attenuation attribute after the storage layer.

The invention has the beneficial effects that: according to the method, the attribute analysis is carried out on the seismic reservoir, the lithofacies sensitive attribute and the gas-containing sensitive attribute are preferably identified to serve as basic parameters, the correlation analysis is carried out by combining the well reservoir characteristics, the attribute with good correlation with the reservoir is determined, and the qualitative prediction of the beneficial zone is carried out by utilizing the attribute. According to the method, logging data environment correction and logging data standardization are carried out in the reservoir inversion process, influences and measurement errors of various measurement environment factors are eliminated, so that the logging curve reflects properties of stratum and pore fluid of the stratum as truly as possible, and requirements of inversion or reservoir parameter calculation on logging data precision are met; finally, by carrying out crack prediction and gas-containing detection on the reservoir, the spreading rule of the effective reservoir is rapidly identified, and the classification evaluation efficiency of the effective reservoir is improved.

Drawings

FIG. 1 is a process flow diagram of the present invention;

FIG. 2 is a schematic diagram of a well-shock bond deposition microphase partition method;

FIG. 3 is a graph of the intersection of compressional and shear wave velocities;

FIG. 4 is a preferred histogram identifying a sand shale sensitivity curve;

FIG. 5 is a preferred histogram of pre-stack inversion gas-sensitive parameters;

FIG. 6 is a schematic diagram of a five-and four-segment porosity interpretation model;

FIG. 7 is a schematic diagram of a four-segment and five-segment permeability interpretation model;

fig. 8 is a schematic diagram of a four-stage explanation plate.

Detailed Description

For a clearer understanding of technical features, objects, and effects of the present invention, a specific embodiment of the present invention will be described with reference to the accompanying drawings.

In order to more specifically explain the specific calculation method of the present invention, the following description will be made with reference to the accompanying drawings, and in order to explain the specific operability and practicality of the method, the present invention performs fine structural explanation and mapping on the target layer river group, the gravity well group and the sand temple group for the full coverage area 457km2 (data area 780km 2) of the HC block. The construction and explanation density is mainly that the objective layer must be 1 x 1CDP in the two sections of the river set and the inner small layer is 4 x 4CDP in the other objective layers. And completing 15 layers of structure diagrams, 15 layers of T0 diagrams and 18 layers of stratum thickness diagrams of each large layer and two sections of small layers. The contents of the drawings shown in fig. 2 to 8 do not affect the implementation of the scheme of the present application. A specific embodiment of the present invention will be described in detail with reference to fig. 1.

1. In this embodiment, the first seismic attribute analysis step specifically includes the following sub-steps:

the invention firstly carries out earthquake analysis on the reservoir stratum, and the aim of earthquake attribute analysis is as follows:

one is to preferentially identify facies sensitive properties. The planar deposition microphase research based on well data is based on well point deposition microphase research, and the distribution characteristics of the deposition microphase on a plane are described according to the modern deposition theory and mode prediction description method as guidance, and the spatial distribution and physical characteristics of a reservoir are subjected to interwell predictive description and drawing. The specific steps are realized according to the layers of point-line-surface, namely, the single well-continuous well-plane attribute is used for drawing the phase. Thus, the preferred identification of lithofacies sensitive properties is the need for a GS16 well Dong-Jia-He group, a gravity well group, a Shaxi temple group sedimentary rock plane facies study.

Secondly, the gas-sensitive property is preferred. Theoretical studies have shown that when the body contains fluids such as water, oil or gas, scattering of seismic waves and attenuation of seismic energy are caused as compared to a dense body. In theory, when the pores in the reservoir are relatively developed and saturated with gas, the attenuation of high frequency energy in the seismic wave is greater than the attenuation of low frequency energy. The gas-bearing development characteristics of the reservoir can be indirectly detected by extracting the attenuation gradient attribute of the high-frequency end. The method guides the delineation of the effective reservoir range of each objective interval of the Jiajia river group, the gravity flow well group and the Shaxi temple group in the east region of the GS16 well through optimizing the gas-containing sensitive attribute, the combination structure, the crack and the pre-stack post-stack inversion result.

Among these, various attributes are required for seismic analysis, and seismic attributes related to lithology and gas-bearing properties are preferable. Firstly, attribute primary selection is carried out, and gas-containing sensitive large-class attributes and rock-phase sensitive large-class attributes are primary selected from 8 large-class 83 attributes. Then, attribute optimization is performed, and specific attribute optimization is performed by testing each attribute from among the initially selected gas-containing sensitive large attributes and rock-phase sensitive large attributes.

Different software is preferred. At present, more attribute analysis software is needed, and how to select the seismic attributes of sedimentary rocks suitable for all purpose intervals of the GS16 Dongxu river group, the gravity flow well group and the Shaxi temple group is important. The seismic attribute is preferably standard, firstly, the optimized sedimentary facies and sedimentary microphase sensitive seismic attribute are matched with a single well and are matched with lithology spreading trend. And secondly, the gas-containing sensitivity attribute is in accordance with well logging interpretation results and gas testing results, and is in accordance with prestack hydrocarbon detection results and effective reservoir distribution ranges.

Firstly, selecting a reasonable time window for seismic attribute calculation, wherein if the time window is too large, the comprehensive factors are too many, and a lot of unnecessary information is contained; when the time window is too small, the data are too small, the statistical effect is not obvious, even the truncation phenomenon occurs, and the effective components are lost. The time window is selected by a fixed time window method and a sliding time window method along a layer.

The selection of the appropriate time window is important. In principle, the optimal time window of waveform clustering is between half wavelength and two wavelengths, and if the time window is too thick, the time window may contain multiple phase sequences, and the calculation result cannot be explained. The time window layer section is too thin to truly reflect waveform changes caused by phase change, the contact relation of early stratum overburden of GS16 Dongxu river group, advanced stratum denudation of Lei Koupo group and the like is complex, the stratum thickness changes of each subsection and small layer of two sections are large, if waveform clustering properties are lifted between each gas layer group and small layer, waveform characteristics of the same phase band at positions with large stratum thickness and small stratum thickness can be different, and the plane prediction capability of lithofacies is reduced. For different subsections and small layers, the time windows of waveform clustering are slightly different, and the thickness of the layer and the comprehensive judgment of the stratum contact relation are required to be integrated.

The invention selects top and bottom layer positions as fixed time windows, and takes seismic attribute values in clastic rock deposition units of all target layers of GS16 well Dongxu river groups, self-flowing well groups and Shaxi temple groups. In order to research and determine the types of the earthquakes and sedimentary facies characteristic phases of different intervals of the zone and the plane distribution rule thereof, more than 10 conventional earthquake attributes such as amplitude, frequency, similarity and the like are extracted, comparison analysis is carried out, the most sensitive attribute is determined, correlation analysis is carried out by combining well reservoir characteristics as basic parameters, the attribute with good correlation with the reservoir is determined, and qualitative prediction of the favorable zone is carried out by utilizing the attributes. "qualitative predictions" means that the seismic attributes are relative values and are quantitative predictions relative to the feature 4 seismic inversion.

Several conventional seismic attribute algorithms and implications:

1) RmS Amplitude (root-mean-square) root mean square Amplitude, which is very sensitive to Amplitude variations.

2) Average Instantaneous Frequency average instantaneous frequency, which in complex trace calculations is the rate of change of phase over time, or derivative of phase. In actual calculation, the instantaneous frequency channel is calculated first, and then the average value in the time window is calculated.

3) Average Instantaneous Phase average the instantaneous phase, which describes the vector angular variation of the real and imaginary components of the complex seismic trace. Ranging from-180 deg. to +180 deg..

4) Average Reflection Strength average reflected intensity.

5) When Energy Half-Time is used for describing the Energy decay speed, the Energy Half-Time is a very important independent attribute. Can help identify reservoirs and hydrocarbons.

6) Arc Length is used for calculating Arc Length of waveform in time window. The calculation method is expressed as the following formula (1):

the arc length can be used for distinguishing stratum conditions which are characterized by high amplitude and have high frequency and low frequency, and the stratum rich in sand or mud can be identified in the sand and mud rock interbeds;

in the formula (1), S is the arc length, a (i) is the amplitude value of the sampling point, T is the utilization rate, and N is the number of the sampling points in the time window.

7) The root mean square amplitude of the impedance, which is information extracted on the impedance body, is very sensitive to the amplitude variation of the impedance.

Further preference is given to the above determination of the most sensitive 7 properties (RmS Amplitude root mean square), average Instantaneous Frequency average instantaneous frequency, average Instantaneous Phase average instantaneous phase, average Reflection Strength average reflection intensity, energy Half-Time Energy Half-decay, arc Length, impedance root mean square Amplitude). Finally, the effective attribute of each objective interval clastic rock of the Shaxi temple group of the Jiajia river group, the artesian well group and the Shaxi temple group in the GS16 Jiandong region is preferably identified.

Identifying sedimentary phase-sensitive seismic attributes, wherein the sedimentary phase-sensitive seismic attributes of each layer section of the reservoir are specifically as follows:

1. preferably, the phase sensitive attribute of the whisker two-stage deposition

Analysis of the drilling data shows that: the two sections of GS16 Dongxu mainly take southeast and eastern sources as main sources, and the river channel changes greatly in each period. The distribution of the river sand bodies in each stage has a certain limitation, and the river sand bodies in multiple stages are longitudinally overlapped to form the wide-coverage distribution of the sand bodies. The two sections of GS16 Dongxu gradually develop from early stage to late stage of the leading edge phase of the braided river delta to the underwater diversion river channel, and the micro phase of the estuary dam is taken as the main sand body, and the whole is characterized by the positive rotation of the accumulation withdrawal, and the small layers of sand bodies are overlapped and connected.

The root mean square amplitude, maximum amplitude, average amplitude properties are preferred depending on the small single well phase and the spread characteristics in the plane. The optimized root mean square amplitude, maximum amplitude and average amplitude sensitive attribute clusters of all small layers show that the warm color high value area is the earthquake response of the dominant river channel, and the cold color area is the earthquake response of the peucedanum and the bay of diversion. The dominant river course spreads in the general north-west direction, indicating the response of southeast sources to earthquakes. The waveform clustering attribute high-value strip is the seismic response of the dominant force phase of the underwater diversion river channel and the estuary dam at the front edge of the delta.

2. The phase sensitive property of the deposition of three-six segments is preferable.

The river set of the beard is divided into six sections from bottom to top, wherein the Sichuan basin with one section, three sections and five sections of deposition periods is wholly positioned in a deposition system of a Triangzhou-lake of the Curve river, and the HC area is mainly deposited in a subphase of a shallow lake of the beach. The main sediment is mainly black mudstone and argillaceous siltstone, and lithology combination of the thin-layer coal or coal lines is a main hydrocarbon reservoir of the river group and a direct cover layer of each hydrocarbon reservoir. The Sichuan basin with two sections, four sections and six sections of main sedimentation periods is integrally positioned in a braided river delta-lake system, wherein HC areas mainly take the front edge of the braided river delta as the main underwater diversion river channel phase, and the basin can be divided into a plurality of different microphases.

Root mean square amplitude seismic attributes are preferred based on the segments Shan Jingxiang of the whisker river set and the spread characteristics in the plane. The preferred root mean square amplitude seismic attribute of each segment is similar to the depositional characteristics of each segment, the root mean square amplitude seismic attribute being the seismic response of each segment depositional phase of the set of the river.

3. The sensitive attribute of the pearl flushing section-sand one-section deposition is preferably selected

The regional research results show that: the gravity flow well group divides the river channel underwater from the front edge of the early-stage to the late-stage development delta; a shoal lake and beach dam phase; low and high energy scale beach; the whole is water inlet reverse rotation characteristic. The temple group is generally characterized by the deposition characteristic of mud-coated sand, and the front edge of the development delta is a submarine branch river channel, and the north east material source is the main source. The earthquake attribute of root mean square amplitude of the pearl punch section-sand section is optimized according to the regional research result and Shan Jingxiang division result, and the earthquake attribute of root mean square amplitude is the sensitive attribute for identifying the sediment phase of the pearl punch section-sand section.

The gas-containing sensitive seismic attributes of each layer section of the reservoir are identified as follows:

1. preferably, the gas-containing sensitive attribute of the second section is

Well logging interpretation and gas testing results of known wells indicate that: the two sections of GS16 Dongxu are generally characterized by the distribution of the upper air and the lower water, the air layer is mainly distributed in the two 2 sections and the two 3 sections, and the air-water same layer, the air-water containing layer and the water layer are mainly distributed in the two 1 sections. The gas reservoir type is mainly lithologic gas reservoirs under the construction background.

According to the spreading characteristics of two sections of single well gas testing results and logging interpretation results on a plane, the total energy, maximum energy and attenuation gradient seismic attributes are optimized, and the high-frequency attenuation characteristics of the 3 attributes are obvious. Preferably, the sensitive gas-bearing properties of each of the strata are substantially consistent with the interpretation of single well logs, the gas-bearing sensitive properties being the seismic response of the strata to gas-bearing. The post-stack attenuation high-value region is mainly distributed in the two-whisker-2 subsections, and secondly, the two-whisker-3 subsections and the two-whisker-1 subsections are worse.

2. Sensitive property of gas containing three-six sections

The results of the test of the gas from three sections to six sections show that the oil gas from four sections is better, and the gas layer, the gas difference layer and the gas-containing water layer are developed from four sections. The distribution of four sections of gas reservoirs is not completely controlled by the structure, and a uniform gas-water interface is not needed. To construct a lithologic gas reservoir in the background. The mud rock sections with three sections and five sections which are stable are hydrocarbon source rocks, the dredging fault is communicated with the underwater branch river channel sand with four sections and six sections, and a lithology gas reservoir under the construction background is formed.

According to the spreading characteristics of the three-section to six-section single well gas testing results and logging interpretation results on a plane, the gas-water distribution characteristics and the hiding mode, the total energy, maximum energy and attenuation gradient seismic attributes are optimized, and the high-frequency attenuation characteristics of the 3 attributes are obvious. Preferably, the sensitive gas-bearing properties of each segment are substantially consistent with single well log interpretation results, the gas-bearing sensitive properties being the seismic response of each segment to gas bearing. The post-stack attenuation high-value region is mainly distributed in four sections, and then six sections, three sections and five sections are worse.

3. The gas-containing sensitive attribute of the pearl flushing section-sand section is preferably selected

The pearl flushing section-sand section only has a section of air layer which is relatively developed, and the section of air layer is thin and takes the shape of a strip. The sand-segment gas reservoir is mainly controlled by sedimentary facies and lithology, and is lithology gas reservoir under the construction background.

According to the spreading characteristics of the single-well gas test results and the well logging interpretation results on the plane of the pearl flushing section-sand section, the gas-water distribution characteristics and the hiding mode, the total energy, the maximum energy and the attenuation gradient seismic attributes are optimized, the high-frequency attenuation characteristics of the 3 attributes are obvious, and the attenuation gradient attributes are sensitive. Preferably, the sensitive gas-bearing properties of each segment are substantially consistent with single well log interpretation results, the gas-bearing sensitive properties being the seismic response of each segment to gas bearing. The post-stack attenuation high value regions are distributed mainly in the sand section.

2. In this embodiment, the second deposition phase is described as a deposition microphase, and specifically includes the following sub-steps:

s201, based on logging and coring data of the reservoir, the system observes and describes the rock core, analyzes rock components, structures and structures, and combines logging data interpretation results and seismic data comparison data to induce sediment phase types and distribution characteristics, namely phase mark characteristics, of the reservoir. The phase mark characteristic acquisition process of the invention is realized by referring to the prior art, and specific documents are as follows: [1] wang Xia the Western depression of Liaohe is a sand four upper subsections sedimentary facies research [ D ]. University of geology in China (Beijing), 2017.

The specific process of the deposition phase and the deposition microphase is as follows:

1. two-stage deposition microphase description

The regional research results show that: in the second deposition period, the basin has the general trend of low east and high west, the sedimentation center is positioned in the front zone of the gantry mountain, and the research area is positioned in the stable gentle slope zone. The two-stage deposition period is multiple sources, the research area source mainly reaches the bulge in Qian in south, and delta-lake deposition is deposited under the control of the paleo-structure and paleo-weather of the area.

1.1 deposition phase signature

The facies markers refer to some markers which can reflect sedimentary facies most, and are the basis of sedimentary facies analysis and lithofacies paleogeographic research, and the facies markers comprise lithology, granularity, primary sedimentary structures, paleobiology, geochemistry, remains and the like. The phase mark features are induced by comprehensive research on a large number of core observations, descriptions, sampling analysis and rock sample analysis data and logging data in a research area.

The invention is based on logging and coring data, and the system observes and describes the rock core, analyzes rock components, structures and other items of marks, is assisted by indoor analysis and test data, and has systematic knowledge of the type and distribution characteristics of sedimentary phases of the two sections of the east whisker of the GS16 well by the rock, electric, physical and earthquake and other items of data. Based on core observation, rock phase analysis, well logging data interpretation and seismic data comparison analysis, the two sections are considered to be the front edge subphase deposition of the braided river delta of the braided river triangular continental phase, and the micro-phase of the underwater branch river, the micro-phase of the estuary dam, the micro-phase of the far sand dam and the micro-phase of the diversion bay are developed.

1) Plaited river triangular intercontinental phase

The plait river delta is a coarse chip delta formed by plait water flow entering into stable water body. Its development is controlled by the flow of seasonal water or river flow in mountainous areas. The tail end of the alluvial fan and the alluvial plain or the mountain area at the side edge of the mountain top can directly enter a stable water body after being transported for a short distance or a long distance to form a plaited river delta.

According to the deposition environment of the plait river delta, and the deposition marks of the color, composition, structure, construction, thickness and the like of the deposition rock, the phase is divided into plait plain subphases (split river channel microphase, split interwinding microphase), plait river delta front edge subphases (underwater split river channel microphase, underwater split interwinding microphase, estuary dam microphase and basket microphase) and front plait river delta subphases (front delta mud microphase).

2) Pigtail front subphase of the plaited river delta

The second section is the front edge subphase deposition of the triangular continent of the plait river, the deposition microphase mainly takes the superposition development of microphase sand bodies of the underwater diversion river channel and the estuary dam, and the partial areas are the deposition microphase of the diversion bay, the far sand dam and the like.

a, micro-phase among underwater diversion channels:

the underwater diversion river channel is deposited on the upper part of the front edge of the braided river delta. The method takes gravel-containing sandstone and coarse sandstone supported by thick layers of detritus as main lithology, and has low maturity. The sediment filled in the diversion river has positive rhythm of granularity with coarse lower part and fine upper part, the bottom is provided with a flushing surface and retained gravel and deposited mud, and the sediment is generally in a block shape, and the upward granularity is thinned.

Curve morphology: box, toothed box, bell, funnel. Curve amplitude: low GR is the dominant; contact relationship: mainly with gradual top change and low abrupt change. Representative well: HC001-27-X1 well 2425.53m-2430.41m; HC1 well 2151.46m-2161.42m; HC6 well 2153.96m-2164.2m; HC149 well 2348.64m-2355.44m; HC149 well 2341.84m-2347.28m.

The microphase sediment between the sub-phase diversion riverways at the front edge of the braided river delta mainly consists of sandstone, coarse sandstone and mudstone, and the sediment characteristic is wholly upwards granularity thinned. The micro-phase cross development among the diversion channels is common.

For example, HC3 well, two sections of 2143.23-2143.27 m fine-middle grain rock chip feldspar sandstone, branch-containing coal-shaped carbon chip delta front edge diversion river microphase. HC7 well, the second section 2188.13-2188.35 m light gray medium-grain sandstone is provided with a great amount of mud front edge flow-dividing river sand microphase. HC106 well, two sections 2199.06-2199.22 m light gray middle-coarse sandstone clamp stay gravel delta front edge diversion river microphase.

b, river mouth sand dam microphase:

the device is positioned in front of an underwater diversion river channel and continuously develops towards the center of a lake basin along the direction of the device, the granularity is mainly composed of fine powder sand-medium fine sand which are well separated, and the sediment grain sequence mainly shows reverse rhythm. Due to seasonal influences, a muddy interlayer is often accompanied, and the deposition structure mainly comprises a small-sized staggered layer and a parallel layer. In finer silty mudstones, the deformation layer or disturbance structure formed by the sliding action or biological disturbance is visible.

If HC5, the fine sandstone in light gray of 2204.94-2205.32 m is changed into the microphase of the siltstone delta front edge estuary dam to the microphase of the shallow lake sand dam.

Curve morphology: a box shape and a toothed box shape. Curve amplitude: mainly medium-low GR; contact relationship: with top gradual change, low abrupt change or bottom gradual change; representative well: HC3 well 2143.48m-2147.16m; HC149 well 2348.84m-2347.28m.

c far sand dam microphase:

the far sand dam is positioned at a far position in front of the estuary sand dam and is also called as an end sand dam. The sediment is finer than the estuary sand dam, mainly powder sand, and has a small amount of clay and fine sand, and can develop a grooved staggered layer, a wrapping layer, water flow waves and wave marks, a flushing-filling structure and the like. The structural grain layer consisting of silt and clay and the color grain layer consisting of plant carbon scraps are also characterized in that the structural grain layer is increased towards the estuary direction, and the color grain layer is reduced; the direction to the lake is opposite. Far sand dams have little fossil, and only sporadic biological scale cases can be seen, so that wormholes can be seen. On the layer sequence, below the estuary sand dam, above the front delta clay deposit, a vertical layer sequence with lower fineness and upper thickness is formed, which is an important distinction from the river deposit layer sequence

Curve morphology: finger-shaped and toothed box-shaped. Curve amplitude: mainly medium-high GR; contact relationship: top-low abrupt change or top-abrupt change bottom gradual change; representative well: HC3 well 2147.96m-2152.54m; HC149 well 2339.53m-2341.51m.

d split bay microphase:

the split bay is a relatively low-lying area between underwater branch riverways, and has weak hydrodynamic force. When the delta advances, a series of wedge-shaped muddy sediments with tips pointing to the land are formed between the branch river channels. So the tributary bay is mainly deposited with clay and contains a small amount of silt and fine sand. Sandy deposits are mostly formed by river bed flood deposition during flood seasons, often clay intercalation or thin lenticular. So the tributary bay deposit has both horizontal and lenticular layering.

Curve morphology: finger-shaped and toothed box-shaped. Curve amplitude: mainly high GR; contact relationship: top low mutation; representative well: HC001-34-X2 well 2377.64m-2391.92m; HC149 well 2361.17m-2363.33m.

1.2 phase characteristics of two-section profile

The deposition of the two sections in HC area is started under the background that the filling and the filling of the erosion surface of the lightning slope group are completed in the first section. The two sections are dark gray mudstone or coal wires with a plurality of stacked thick sandstone layers, and the thickness of the single layer of the sand body is larger. Has grain sequence characteristics of coarse grain and fine grain, bidirectional staggered layer arrangement, parallel layer arrangement, scouring structure and the like formed by developing carbon scraps. The middle part of the deposition gyratory is coarser, the rock scraps are mainly hard rock scraps, and the stratum with less gap filler content is a stratum with better reservoir physical property.

Viewed in cross section from the GS16 east parallel source and perpendicular source: the first stage of the development of the shore shallow lake phase and the second stage of the development of the plaited delta front subphase of the plaited delta phase are deposited. The deposition microphase is mainly the microphase of the underwater diversion river channel and the estuary dam, and the microphase of the local development diversion bay and the distant sand dam is deposited. The two sections of underwater diversion river sand bodies are overlapped and connected, and the main river sand bodies are distributed in a bead shape. The microphase of the shallow lake and the microphase of the diversion bay are the main part among all the small layers. The two sections of the underwater branch river sand bodies from early to late are gradually developed and are longitudinally overlapped and connected. Terminal two-phase beard ₃ SubBrucella is inferior to SubBrucella in early stage ₂ Sub-segment, second beard ₁ The subsections are more compact.

1.3 two-stage planar phase characteristics

1) Two-stage source analysis

From the previous Sichuan basin two-stage heavy mineral partition map: the Guangyuan-Mianyang-Tongnan-nan Jiang Zhuang-Zhengan-Pagjiang-Zhongxian-Wushan Zhuang characterized by a high tourmaline content; the middle and the south of the basin are characterized by high garnet content; there are both GS garnet and tourmaline regions in the southwest. Reflecting the impact of the Peng Guan muzzle rock parent zone. The heavy mineral characteristics of the research area are characterized by high tourmaline content, and the object source is mainly from the mother rock area in the southeast part.

2) Sensitive attribute preferred deposit phase transition

Based on the study of geological data such as GS16 Dongxu two-section drilling core, the single well sedimentary microphase section and the well connection Xiang Poumian are established by combining the data such as logging, logging and imaging. Preferably, cluster analysis is carried out on root mean square amplitude, maximum amplitude and average amplitude sensitive attributes of two sections of each small layer, and well shock comparison shows that the waveform cluster attribute high-value strip is the seismic response of the main force phase of the underwater diversion river channel and the estuary dam at the front edge of the delta.

As shown in fig. 2, the seismic attributes are converted into sedimentary phases by well-seismic integration, conversion method: determining an object source and a deposition mode by taking the regional deposition phase as the direction; guiding attribute extraction based on well point phases; preferably, the seismic attribute implements inter-well microphases; well-shake is combined with earthquake phase conversion sedimentary facies.

3) Two-stage small layer deposition phase spreading feature

From the analysis result of the deposition microphase plane of each small layer of the two sections, the GS16 Dongxu two sections are the front edge subphase deposition of the triangular continent of the braided river, and develop the microphase of the underwater branch river, the microphase of the estuary dam, the microphase of the far sand dam and the microphase of the diversion bay. Most areas are mainly based on the microphase of the underwater diversion river channel and the estuary dam, and only a small part of areas develop the sediment microphase of the diversion bay, the far sand dam and the like. The rock core analysis and the oil gas test data comprehensive analysis show that the micro-phases of the underwater diversion river channel and the estuary dam are sedimentary micro-phases which are favorable for the development of the reservoir, and the distribution area is the area where the reservoir develops.

3. In this embodiment, the third reservoir inversion step specifically includes the following sub-steps:

It is well known that all well logs are inevitably affected by various measurement environmental factors, such as well diameter size (especially hole expansion), drilling fluid density, mineralization, formation water mineralization, temperature, pressure, etc., and also the drilling fluid invasion zone and instrument outer diameter, gap distance, core shift or centering, etc., affect the curve measurement. If the original log, which is severely affected by these non-stratigraphic factors and of poor quality, is directly digitally processed, reliable geologic parameters and good geologic results cannot be obtained, sometimes even false conclusions as opposed to geologic, test oil and the like. Therefore, prior to inversion using the log, the original log must first be corrected for appropriate environmental effects, eliminating as much as possible the effects of various measured environmental factors, so that the log reflects as truly as possible the properties of the formation and its pore fluids. Only then can good inversion result be obtained, and good geological effect can be obtained.

The environment correction of the invention is mainly aimed at borehole effect correction by logging curves with larger effects of borehole diameter expansion and borehole wall irregularity. Wellbore effect correction models are typically examined and corrected point by point using a method that calculates an upper bound. Borehole correction is performed on the density curve ρb. The principle is as follows: assuming that the normal well diameter condition, the lower limit value of the formation density of the interpretation well section is ρmin, which is represented by the following formula (2):

ρmin＝Vsh×ρsh+(1-Vsh×ρsh)ρp (2)

In the formula (2), ρsh is the mudstone density of the explanation well section; vsh is the clay content of the stratum at the current sampling point, and can be calculated by a natural gamma logging curve and the like; ρp is a pure formation density value that accounts for the maximum porosity in the wellbore interval. When ρb < ρmin, it is considered that ρb < ρmin is measured due to the enlarged well diameter or the irregular well bore, and here, ρb=ρmin is taken as an approximation of the formation density at the sampling point. Conversely, if ρb is taken as ρmin, then it is unchanged.

For acoustic logging, let the upper limit of the formation acoustic time difference at the interpretation interval under normal wellbore conditions be Δtmax, represented by the following formula (3):

Δtmax＝Vsh×Δtsh+(1-Vsh)Δtp (3)

in the formula (3), delta tsh is a differential value of the mud sound wave time of the explanation well section; Δtp is the acoustic time difference value of the pure stratum with the greatest porosity in the interpretation well section; vsh is the clay content of the formation at the current sampling point and can be calculated from natural gamma log curves and the like. When Δt > Δtmax, then it is considered that the measured Δt is larger than Δtmax due to borehole expansion effects, at which time Δt=Δtmax is taken as the approximate acoustic time difference at that point; when Δt is Δtmax, no correction is made, and the original Δt value is still adopted.

The acoustic wave and density curves of a few wells in the GS16 area are greatly influenced by the well diameter, and the larger the well diameter is, the faster the trend of the density curve is to be smaller. Aiming at the influence of the change of the well diameter on the sound wave and density curve, the SMI processing software is utilized to perform environment correction processing on the basis of curve splicing, and the influence of the well diameter condition is eliminated. The corrected sound wave and density curves all tend to be reasonable, and can be used for later reservoir inversion.

Environmental correction is mainly used to eliminate non-systematic errors, which are eliminated by data normalization. The systematic errors mainly come from the technical performance and detection capability of the instrument itself, instrument scale deviation, instrument faults of certain stability, unreasonable operation of operators and the like. Because logging data in a work area is done by different measurement personnel at different times using different instruments, systematic errors exist between the individual logging data.

In order to meet the requirement of inversion or reservoir parameter calculation on the accuracy of logging data, the logging value of each well needs to be subjected to secondary calibration in the range of a work area, and the most common method is to uniformly calibrate all logging data by using a standard layer, namely logging data standardization.

Raw log data contains two potential measurement errors, systematic and non-systematic. The elimination of non-systematic errors is accomplished by environmental impact correction, while the elimination of systematic errors is the task of data normalization. The method adopts a mean variance method to carry out standardization treatment on the logging curve.

Aiming at the systematic errors caused by different logging ages, different instruments, different logging units and different mud, the curve must be standardized before sensitive curve analysis. And selecting compact limestone at the top of the lightning slope group as a standard layer. The wells in the work area are relatively new wells, and the systematic error of the logging curves between the wells is small and is generally 1-2 us/ft. The standardization process is realized by adopting SMI software, wherein the AC curve, the GR curve, the DEN curve and the RD curve are processed in the standardization process, and the total number of the wells is 152.

Petrophysical analysis is the basis for seismic reservoir prediction. By analyzing the relation between the rock elasticity parameter and the reservoir physical property parameter, the sensitive parameter which can better reflect lithology or reservoir physical property is found. The choice of sensitive parameters of the reservoir determines the formulation of the inversion scheme and the quantitative analysis of the inversion result.

The invention provides a two-stage petrophysical analysis process, which comprises the following steps:

1) Transverse wave prediction

And 8 real transverse wave wells are collected, and according to the relation between the longitudinal wave speed and the transverse wave speed, the longitudinal wave and transverse wave matching relation of the section of the 8 wells is slightly worse, so that the transverse wave data can be seen as a whole. As the lithology distribution of HC001-22-X1 well is complete, sand and mud rocks are distributed, and the HC001-22-X1 well is used as a well for determining the mineral skeleton modulus of a work area, as shown in FIG. 3, and the rest wells are used as check wells.

The pre-stack inversion is limited by the shear wave data, and the current acquisition adopts Xu Huaite method to predict no shear wave data well by using 8 known actual measurement shear wave data wells (HC 001-20-X1, HC001-22-X1, HC001-3-X2, HC001-8, HC1, HC104, HC107 and HC 112).

2) Identification of sandstone sensitivity curve

As shown in fig. 4, from the mudstone and sandstone log histogram analysis, it is shown that: AC. The DEN and LLD curves identify that the sand shale is bad, the sand shale curves are overlapped, the GR curves identify that the sand shale is most sensitive, and the GR of the sand shale is less than 90API.

3) Physical inversion sensitivity parameters are preferably

And (3) utilizing HC1, HC102, HC104, HC112 and HC3 well core data homing to establish a relation with the logging curve. The correlation is good between the core analysis porosity and the density, and between the core analysis porosity and the acoustic time difference, wherein the correlation between the core analysis porosity and the acoustic time difference is the best, the correlation coefficient reaches 0.92, the sample number is 237, and the porosity can be directly inverted by using the acoustic time difference.

4) Pre-stack gas-containing inversion sensitivity parameters are preferred

The gas analysis data of the transverse wave well was measured by 8 wells (HC 001-20-X1, HC001-22-X1, HC001-3-X2, HC001-8, HC1, HC104, HC107, HC 112). From four pre-stack inversion elastic parameter statistical histograms 5 of the Lap coefficient, the shear modulus, the longitudinal-transverse wave velocity ratio and the Poisson ratio, it is known that: poisson's ratio and longitudinal and transverse wave speed ratio distinguish between gas layer, gas-liquid layer and water layer. A water layer limit value, namely a longitudinal and transverse wave speed ratio limit value is 1.68; the poisson's ratio limit is 0.23.

In summary, the two-stage petrophysical analysis results are as follows:

A. two lithology sensitive parameters are identified, namely, sand shale can be identified by utilizing GR curve simulated sound waves, and the GR threshold value of the sand shale is smaller than 90API; carrying out prestack inversion by using the prestack longitudinal and transverse wave speed ratio to identify sand shale, wherein the sand shale longitudinal and transverse wave speed ratio is less than or equal to 1.76;

B. The lower limit of the porosity of the favorable reservoir is 6%, and the acoustic wave or pseudo-acoustic wave inversion porosity can identify the favorable reservoir;

C. the longitudinal and transverse wave speed ratio is less than or equal to 1.68, the Poisson ratio is less than or equal to 0.23, and the gas-containing property is identified to be most sensitive. The gas-bearing reservoir can be identified by performing a pre-stack inversion using the sensitive parameters.

In the embodiment, the rock physical analysis process of three sections to six sections is also provided, and the rock physical analysis process is specifically as follows:

1) Three-section-six-section logging interpretation

(1) Porosity interpretation model establishment

And establishing a four-section and five-section porosity interpretation model according to the collected rock core analysis data. Because of the limited data, five sections of data points are fewer, four sections of 3 wells have core data, three sections of six sections of physical property interpretation are needed according to the two plates. Five sections were used to build the porosity model with 16 data points for HC1 well, and four sections were used to build the porosity model with 110 data points for HC1, HC3, and HC107 wells, as shown in FIG. 6.

(2) Permeability interpretation model establishment

Based on the collected core analysis data, HC1 well, HC3 well, HC104, HC107, and 139 data points of 4 wells, four-section and five-section permeability interpretation models are established, as shown in FIG. 7. The power function relation between the permeability and the porosity is better than the exponential relation, so that the calculation formula of the permeability adopts the power function relation.

(3) Interpretation plate creation

Four sections of plates are required for establishing 7 well test gas data such as HC001-21-X2, HC001-34-X2, HC001-44-X3, HC001-47-X4, HC001-51-X1, HC001-51-X2, HC001-51-X3 and the like, as shown in FIG. 8, and the array induction logging series is provided with few wells.

2) Identifying favorable reservoir, effective reservoir, porosity, permeability sensitivity attribute preferences

The sensitivity curve of the favorable reservoir layer with three sections to six sections is preferably GR and DEN, when GR is less than 90API, DEN is less than 2.48g/cm3, DEN is less than 2.5g/cm3, DEN is less than 2.54g/cm3, DEN is less than 2.52g/cm3, and the sensitivity attribute of the favorable reservoir layer with three sections to six sections is identified

The effective reservoir sensitive attribute is preferably pre-stack VP/VS, and the aspect wave velocity ratio of the three sections to the six sections is respectively as follows: VP/VS v 1.52, 1.53, 1.55, 1.6.

The three-six sections are needed to identify the porosity sensitive curve, and the four-five sections DT-POR porosity model is optimized. The three and six sections have no data and are replaced by adjacent layer four and five section porosity models respectively. The permeability sensitive curve is identified from three sections to six sections, and a four-section and five-section POR-PERM permeability model is optimized. No data of the three and six sections are replaced by the four and five section permeability model.

At present, the phase control inversion method is mainly divided into two main types, namely conventional geostatistical phase control inversion; the other is phase control based on "seismic waveform indication inversion". Different inversion results can be obtained by different phase control inversion methods, and the selection of the inversion method is important.

4. In this embodiment, the step four crack prediction specifically includes the following sub-steps:

The SMI phase control inversion refers to the inversion of 'seismic waveform indication' without adding phase constraint and taking phase as precondition, but utilizing the characteristics of waveform itself to realize the purpose of phase control, and the phase rule is used for checking inversion results, not precondition of inversion, so that the true 'phase control' inversion can be achieved. SMI phase control inversion is mainly characterized in that:

1) The earthquake transverse resolution is higher;

2) No special requirement is made for well distribution;

3) Inversion modeling integrated flow.

The seismic waveform characteristic indication inversion software (SMI software) is newly developed by Henglihua petroleum technology research institute in Beijing, and the basic idea is to refer to two factors of waveform similarity and space distance when screening statistical samples, and order the samples according to the distribution distance on the basis of ensuring the consistency of the structural characteristics of the samples, so that the inversion result embodies the constraint of a sedimentary facies belt in space, and the sedimentary rule and characteristics are more met on the plane.

The basic flow of the seismic waveform indication inversion is as follows:

A. and analyzing the known well according to the characteristics of the seismic waveform, preferably establishing an initial model of the well with high correlation degree with the waveform of the channel to be discriminated, and counting the longitudinal wave impedance of the initial model as prior information.

B. And carrying out matched filtering on the initial model and the earthquake wave impedance, and calculating to obtain a likelihood function.

C. And combining the likelihood function and the prior probability under a Bayesian framework to obtain posterior probability density distribution, and sampling the posterior probability density distribution as an objective function. And continuously perturbing the model parameters to maximize the posterior probability density value, taking the solution at the moment as a feasible random implementation, and taking the average value of multiple feasible implementations as an expected value to output.

In this embodiment, a two-stage prestack, poststack inversion procedure is given:

1. two-stage post-stack inversion

The petrophysical analysis results show that: the two sections GR are required to identify that the sandstone is most sensitive, and the sandstone threshold GR is less than 90API; the lower limit of the porosity of the favorable reservoir is 6%; the gas-containing sensitivity is identified by the longitudinal and transverse wave speed ratio which is less than or equal to 1.68. The inversion process after two-stage stack is as follows:

1) Reservoir physical Properties

The reservoir of the group of the Sharphome rivers in the east GS16 region is a fracture-pore reservoir. For a fracture-pore type reservoir, the pore is the main reservoir space, and the fracture mainly plays a role in improving the seepage capability of an oil-gas layer and improving the gas production of a well drilling. Drilling has shown that reservoirs with a certain thickness are the basis for obtaining gas in this zone, and the effective configuration of the reservoir and the fracture is the key to achieving high gas production in this zone.

The GS16 Dongxu river group reservoir develops in a dense sandstone layer, the lithology of which is mainly coarse-medium grain rock chip feldspar sandstone and feldspar rock chip sandstone, and sandstone with the porosity of more than 6% can form a favorable reservoir.

2) Reservoir sensitivity parameter characterization

Thin layer mudstones of the beard river group in the east region of GS16 are developed, and the reservoir is a relatively low-speed layer in tight sandstone and generally shows the characteristic of three-low-one-high logging response of relatively low natural gamma, low density, low longitudinal wave speed and high porosity on a logging curve.

Favorable reservoir section characteristics: porosity is greater than 6%; natural gamma value is less than 90API; the longitudinal wave velocity is 4900m/s or less. After the influence of high natural gamma mudstone is removed, the longitudinal wave speed and the porosity are in a good linear negative correlation relationship, namely, the longitudinal wave speed is reduced along with the increase of the porosity.

3) Two beard ₁ Sub-segment, second beard ₂ Sub Duan Changgui tight reservoir inversion technical route

Different inversion methods are used for the two sections of unconventional tight reservoirs and the tight reservoir. Two beard ₂ Two beard ₁ A tight reservoir: removal at speed inversion>4800m/s dense sandstone and gamma inversion removing gamma>Mudstone and porosity inversion acquisition of 90API>After 6% of the beneficial reservoirs, effective reservoirs were obtained by inversion of the pre-stack VP/VS.

A. SMI speed inversion. Removing dense sandstone with the speed of >4800m/s by inversion, and reserving favorable sandstone;

B. SMI gamma pseudo-sonic inversion. On a data body with dense sandstone of >4800m/s removed, utilizing gamma pseudo-acoustic inversion to remove mudstone with gamma >90 API;

C. SMI porosity inversion. And on the mudstone data body with the elimination of gamma >90API, carrying out porosity inversion by utilizing a gamma pseudo-acoustic wave technology to obtain a favorable reservoir with the porosity of > 6%.

4) Two beard ₃ Sub-section unconventional tight reservoir inversion technical route

Two beard ₃ Unconventional tight reservoirs: fluid development reservoirs are identified through favorable reservoir prediction, and mudstone is removed through gamma pseudo-acoustic inversion and porosity inversion is obtained>4% of the beneficial reservoirs, and the effective reservoirs were obtained by pre-stack VP/VS inversion.

Two beard ₃ Sub-section must be two ₁ Sub-segment, second beard ₂ Duan Shayan is more compact, handle two ₃ Subsections as tight reservoir inversion, with beneficial reservoir porosity set to>4, specific inversion technical route:

A. SMI gamma pseudo-sonic inversion. Removing mudstone of gamma >90API by utilizing gamma pseudo-acoustic inversion;

B. SMI porosity inversion. And on the mudstone data body with the gamma of >90API removed, carrying out porosity inversion by utilizing a gamma pseudo-acoustic wave technology to obtain a favorable reservoir with the porosity of > 4%.

5) Porosity and permeability inversion technical route

The relation between the core porosity and the acoustic time difference is established, the correlation coefficient is very high, and the relation between the core porosity and the logging curve (data sources: HC1, HC102, HC104, HC112 and HC3 well core data homing) can be directly used for inverting the porosity and the permeability.

The inversion method comprises the following steps: the porosity is inverted using acoustic waves according to their correlation with porosity. The porosity body is converted into a permeability body according to the correlation of porosity and permeability.

2. Two-stage prestack inversion

The petrophysical analysis results show that: the longitudinal and transverse wave speed ratio is most sensitive to the two sections of gas contents, and the longitudinal and transverse wave speed ratio is less than 1.68. The effective reservoir inversion overall concept is: by inversion of the prestack longitudinal and transverse wave velocity ratios, an effective reservoir with a longitudinal and transverse wave velocity ratio < 1.68 is identified. Based on pre-stack inversion, the final effective reservoir plane distribution range is comprehensively determined by combining pre-stack hydrocarbon detection results, post-stack hydrocarbon detection results, pre-stack crack prediction results, post-stack favorable reservoir inversion results and the like.

In this embodiment, the pre-stack and post-stack fracture prediction of the reservoir specifically includes: carrying out pre-stack AVO analysis based on the gas-containing sensitive seismic attribute selected in the first step to obtain the distribution characteristics of a high-value region of the gas-containing probability of the reservoir pre-stack AVO; and carrying out post-stack gas-containing detection based on the gas-containing sensitive seismic attribute, extracting the attenuation gradient attribute of the high frequency band of the seismic wave, and obtaining the distribution characteristics of the high-value area of the attenuation attribute after the storage layer.

Direct acquisition of reservoir fractures includes core, microscopic observation and scanning electron microscope microscopic, ultra microscopic analysis, outcrop and imaging logging data, full-borehole microresistivity scanning imaging logging (FMI imaging logging) is the most intuitive means of fracture measurement.

The well wall FMI imaging logging can provide high-resolution well wall images, and is an effective means for researching reservoir cracks at a well shaft. The crack occurrence can be conveniently determined on the well wall image, and the crack type can be judged. The well wall image is subjected to image resolution processing, quantitative parameters such as cracks, holes and the like can be obtained, and a foundation is laid for further researching the development rule of the cracks.

On the basis of crack pickup, in order to quantitatively understand the development degree and effectiveness of the cracks, quantitative calculation of the density, length, width and porosity of the high-conductivity cracks is required. The calculation formula is expressed as the following formula (4):

ΔD＝a*A*R _xo ^b *R _m ^(1-b) (4)

in the formula (4), delta D is the width of the crack, A is the area of the abnormal conductivity caused by the crack;

R _xo formation resistivity (typically invaded zone resistivity), R _m -mud resistivity;

a. b-an instrument-dependent constant, wherein b is near zero, a, R _xo Are calculated based on the image after calibration to the shallow lateral resistivity LLS.

Prestack fracture prediction is one of the difficulties and hot spot problems in the geophysical world. Many scholars at home and abroad research the seismic attribute, and the depth and the breadth of the seismic attribute are gradually improved through the transition from the research of the anisotropic theory to the practical application of the synthetic seismic attribute.

The invention adopts an FRS crack detection method for the research of the GS16 pre-stack cracks, and the method is a seismic detection method based on longitudinal waves. When the seismic P wave is reflected when encountering a fracture stratum, the reflection is different due to the difference of the azimuth angles of the P wave and the fracture. By utilizing the characteristic of wide azimuth angle of the three-dimensional seismic data, the relative degree of crack development can be reversely deduced by extracting seismic P waves with different azimuth angles, and the method is effective for openable high-inclination cracks.

The calculation flow of crack detection:

1) And (5) extracting and superposing the azimuth gather in the CMP gather. The number of the azimuth angles added into the system is 3-6, and the system is required to be basically and uniformly distributed in the range of 0-180 degrees;

2) The seismic attribute may employ calibrated amplitude data, such as relative wave impedance data. The relative wave impedance is calculated for each azimuth superposition gather. This process is essentially a calibration of the dimension;

3) And carrying out ellipse fitting on each CDP point of the reservoir by using the time window statistical attribute values of the azimuth angles, and calculating 3 characteristic values: the length of the major axis, the length of the minor axis, and the angle between the major axis and the X axis of the ellipse. Then obtain the ellipticity (major/minor axis);

4) Judging how the included angle indicates the crack direction according to the response relation of the selected seismic attribute to the crack azimuth and the result in forward modeling, wherein the ellipticity generally indicates the crack density distribution;

5) The fracture azimuth analysis may select data of other attributes.

Seismic attenuation is related to the spatial variation of the fracture density field. The attenuation along the direction of the crack trend is slow, the attenuation along the direction perpendicular to the direction of the crack trend is fast, and the attenuation is faster as the crack density is larger. The cracks not only produce seismic attenuation, but also produce interference of seismic waves, and the instantaneous frequency is used for researching the interference phenomenon of the seismic wave field. The instantaneous frequency can be used to analyze the time-varying spectral properties of coherent waves in a fractured reservoir and the attenuation properties of the scoring seismic waves. The vertical fracture direction decays fast and strong heterogeneity appears, and the gas-containing fracture shows stronger heterogeneity than the oil-containing fracture.

Under the same porosity conditions, small cracks have a greater impact on velocity than circular holes, and crack porosities less than 0.01% in sandstone can result in a reduction in longitudinal and transverse wave velocities by more than 10%.

Thus, the direction, density and contained fluid changes of the fracture have a great effect on longitudinal and transverse wave velocities and produce strong seismic anisotropy, the impact of the fracture on the amplitude versus azimuth characteristics increases with increasing offset, and the larger offset can make the amplitude versus azimuth changes evident from the fracture, so the azimuth Amplitude Versus Offset (AVO) attribute can be used to detect the fracture. The greater the reservoir fracture density of the hydrocarbon-bearing fluid, the greater the azimuthal variation in amplitude at the same offset. It follows that the seismic anisotropy characteristics caused by the cracks are obvious, and the crack is feasible to be researched by using pre-stack seismic data.

In the single well fracture forward modeling process, the zero offset seismic synthesis record obtained by convolution model calculation is compared with the actual seismic record by forward modeling calculation to obtain the multi-azimuth pre-stack seismic synthesis record of the gas-bearing fracture reservoir. The variation of azimuth amplitude caused by the crack is quantitatively characterized by the AVO response of the normalized reflection amplitude with the incident angle, and the distribution relation of the amplitude azimuth ellipse at each incident angle and the crack is qualitatively used for explaining how to determine the direction of the crack by using the long and short axes of the amplitude azimuth ellipse.

In the pre-stack seismic anisotropy calculation process, the distribution of cracks is a main factor for controlling the productivity of a production well and the flow of underground fluid, and the pre-stack seismic anisotropy calculation process comprises the following steps:

1) Wide azimuth data split azimuth analysis

Dividing the CMP gather after wide azimuth acquisition and dynamic correction into 6 azimuth gathers according to the principle of coverage frequency balance: 20 °, 50 °, 80 °, 110 °, 140 °, 170 °.

2) Azimuthal superposition processing

(1) Pre-stack routine processing

The conventional pre-stack treatment is mainly characterized by speed selection and residual static correction, the energy clusters on the speed spectrum of the region are concentrated, the speed change trend is obvious, and the selection is easier. After three iterations of speed selection and residual static correction, the resulting superimposed profile effect is already good. The prestack conventional treatment is also carried out to cut off at first arrival and press spike energy with larger amplitude, the finally obtained CMP gather after dynamic correction can be used for crack prediction, attenuation is carried out in the treatment process, and spectral whitening is added to improve the resolution of a section.

(2) Azimuthal angle processing

The most important processing link for pre-stack crack detection is azimuthal split processing. Because the azimuth-offset distribution of the observation system is not completely uniform, the azimuth processing of the seismic data is limited to a certain extent. The original CMP gather is divided by 20 °, 50 °, 80 °, 110 °, 140 °, 170 ° according to the principle of energy balance.

The main method for predicting the post-stack cracks comprises the following steps: ants, variances, coherents, edge detection, and other analysis techniques. We performed selective analysis on these methods, respectively, and finally preferred post-stack fracture prediction results for coherent and variance volumes.

The invention uses coherent body technique to make fault interpretation and combination, which avoids randomness of interpretation result and supplements unreasonable result caused by experience of interpretation personnel. The method can be directly applied to a three-dimensional seismic data body, and the automatic tracking function is utilized to finish the fine interpretation of faults. The interpretation results thus obtained and the combination of planes greatly improve the accuracy and precision.

In principle, the calculation of the seismic coherence data volume is very simple and easy to understand. According to the track number, the dip angle and the calculation selection time window size of the given data body, calculating a correlation coefficient by using the following formula (5):

in equation (5), R is the coherence coefficient, is a function of trace time and the tilt angle of the two traces, T is time, φ is tilt angle, and T' are trace data pairs. The inclination angle is not easily given under the influence of azimuth, and the coherent data and the coherent time window of the data body are mainly determined during calculation.

The coherent data includes linear 3-way, orthogonal 5-way, and orthogonal 9-way. The larger the number of tracks involved in the coherent computation, the larger the averaging effect, the lower the resolution to faults, which is predominantly large faults. In contrast, the number of the coherent channels is small, the average effect is small, the resolution can be improved, the faults and cracks are improved, and the resolution on the cracks is particularly highlighted. The number of the coherence paths involved in the calculation is selected according to the purpose of researching the geology when the coherence of the earthquake is calculated.

The choice of coherence time window is generally determined by the period t of the reflected wave view on the seismic section, typically taking t/2 to 3t/2. When the calculated time window is smaller than t/2, because the coherence time window is small and the field of view is narrow, a complete wave crest or wave trough cannot be seen, and the zone with small calculated coherence data value may reflect noise instead of reflecting the crack existence position. At a calculated time window greater than 3t/2, multiple reflection events occur simultaneously because the time window is large, so that a band with small calculated coherence data values may reflect event continuity rather than a crack. So too large or too small a time window would reduce resolution for the crack. In practice, the appropriate time window is selected based on the geologic features of the reservoir.

When the method is applied, different coherent body calculation directions are selected, and besides the crack distribution of the main structural direction, the possible crack existence and the correlation of the crack existence in different directions are highlighted by calculation in different directions.

Of course, the coherent technology also has multiple resolvability, and the incoherent data abnormality is not necessarily a crack, and may be caused by lithology change or other geological phenomena, so in the explanation of the crack, the geological condition is comprehensively analyzed, and only the crack which is released by the seismic data with high signal to noise ratio is trusted.

The stress field research method of the invention comprises the following steps: on the basis of the main stress direction simulation, the main stress intensity simulation, the main strain intensity simulation and the structural curvature simulation, the main stress intensity and the main stress direction characteristics of each objective layer section are defined.

The invention uses the bending thin plate as a mechanical model for simulating the oil and gas reservoir structure, and uses a two-dimensional method to treat the oil layer structure, the method has convenient calculation and less manual intervention, and the boundary can be treated as a free boundary for stress field simulation without additionally considering boundary conditions. The method mainly uses a anticline structure as a model for analysis, so that the maximum curvature value of a point on a structural surface can be used as a criterion of the crack development degree of the point, and the minimum principal curvature direction is used for indicating the trend of the possible tensile cracks, so that the distribution problem of the structural cracks is changed into the principal curvature calculation problem of the structural surface.

The coordinate plane with z=0 is set as the middle plane of the sheet, and the coordinate parallel to the earth is set as X and Y coordinates according to the right hand ruleUpward is positive. The displacements along the positive X and Y directions are u _x ，u _y The displacement in the Z direction is the degree of disturbance w (x, y). The specific stress field numerical simulation process is as follows:

1) Basic equation

The deformation geometry equation in the rectangular coordinate system is the following formula (6):

From the theory of thin plates, the following formula (7) is known:

and has the following formula (8):

defining a curvature deformation component as the following formula (9):

thus, the strain component can be written as the following formula (10):

ε _x ＝-zκ _x ,ε _y ＝-zκ _y ,γ _xy ＝-2zκ _xy (10)

2) The physical constitutive relation (generalized Hooke's law) is represented by the following formula (11):

the inverse relationship is as follows (12):

in the formula (12), λ and G are the Lame constants, G is the Shear modulus (Shear modulus), E is the Young's modulus (Yong modulus), and θ is the volume strain.

Substituting the foregoing formula (12) into formula (11) gives the following formula (13 a):

thus, there is the following formula (13 b):

substituting the formation thickness t=2z into the above equation, the stress component on the formation face expressed by the curvature component is expressed as the following equation (14):

from the above, when the stratum surface is upward convex, the curvature is larger than zero, and the tensile stress is positive when the stratum surface is just corresponding to the upward convex stratum surface.

After the stress along the coordinates of the point is obtained, the principal stress and the direction thereof can be obtained, expressed by the following formula (15):

σ _max included angle alpha, sigma with X axis _min The included angle beta with the X axis is represented by the following formula (16):

thus, if a disturbance equation of the surface of the earth or the curvature of a point on the surface can be obtained, the stress field on the earth can be estimated, and the fracture generated by the stress can be calculated.

3) Calculation of formation curvature

Fitting of stratum trend surface. From the theory presented above, if the curvature component of the formation face can be found, the stress field thereon can be found. And fitting a trend function of the ground surface by adopting a trend surface fitting method, and further calculating the curvature component of the upper point. The trend surface is fitted using a least squares method.

Let the function of the undetermined coefficients of the trend surface be the following equation (17):

w(x,y)＝a ₀ +a ₁ x+a ₂ y+a ₃ x ² +a ₄ xy+a ₅ y ² (17)

from the coordinate values (x, y, z) at the layer scatter points, a least squares equation is established, expressed as the following formula (18 a) for one scatter point

And (18 b):

ε _i ＝z _i -w _i (x _i ,y _i ) (18a)

when a trend surface is fitted with n scattered points, a fitting equation set can be obtained, and the equation set is solved:

wherein the sum number indicatesI.e. sum 1..n points. Solving the linear equation set can obtain the trend surface function.

The curvature calculation of the trend surface is expressed as the following formula (19):

4) Crack parameter calculation

(a) Curvature parameter

The fitting surface coefficients a3, a4, a5 can be obtained from the above solution equation set, and the curvature at this point can be obtained from the equation (9).

(b) Strain parameters

The strain value can be obtained from the expression (10). Where z=t/2. The corresponding stress, principal stress and principal stress direction can be calculated from equations (14), (15) and (16), respectively.

(c) Stress parameter

The principal stress and the direction thereof can be obtained by solving the above equations (14), (15) and (16).

The stress field numerical simulation provided by the invention considers the thickness and lithology of reservoir rock, considers factors such as fracture distribution of reservoir under structure control, and the like, further calculates stress and strain distribution of the structure based on the elastic sheet theory on the basis of calculating curvature distribution of the structure, and then can analyze the development degree and the spreading relation of the reservoir fracture according to the stress and strain field of the structure. Wherein, construct curvature: indicating the gradient change speed of the construction surface; maximum main strain: indicating the magnitude of the deformation; tensile strain (+): related to the anisotropic density; compressive strain (-): representing compaction deformation of the formation; maximum principal stress: compressive stress (-): perpendicular to the main direction of anisotropy; tensile stress (+): parallel main directions of anisotropy; stress direction angle: the direction of maximum principal stress, combined with tensile strain, may represent the direction of anisotropic development.

5. In this embodiment, the fifth gas-containing detection step specifically includes the following substeps:

for pre-stack and post-stack gas-containing detection of reservoirs, the gas-containing detection methods commonly used are: frequency attenuation coefficient gather calculation technology, seismic energy attenuation attribute extraction analysis technology and absorption attribute calculation technology. The prestack inversion technology and the prestack AVO analysis are mature technologies at present and are widely applied. Because of strong heterogeneity and complex lithology, the gas-containing detection difficulty is great, and the detection result has multiple solutions. The invention adopts the pre-stack AVO analysis to detect the pre-stack gas-containing property. AVO analysis is performed primarily on a common-center-point gather (CMP). Within the CMP gather, a study of the relative amplitude of the traces is an analysis of the amplitude versus offset.

The effect of AVO analysis depends on the ability to predict the reflection coefficient and the ability to satisfy the Zoeppritz equation (1919) for boundary conditions of plane wave method stress and shear stress and displacement continuity at a flat interface between half spaces based on two elastic media. The Zoeppritz equation may predict the expected amplitude variation for any lithology combination.

Theoretical studies have shown that when the body contains fluids such as water, oil or gas, scattering of seismic waves and attenuation of seismic energy are caused as compared to a dense body. In theory, when the pores in the reservoir are relatively developed and saturated with gas, the attenuation of high frequency energy in the seismic wave is greater than the attenuation of low frequency energy. By extracting the attenuation gradient attribute of the high frequency band, the gas-containing development characteristic of the reservoir can be indirectly detected.

The invention optimizes the poststack sensitive seismic attribute of the identified gas interval based on a large number of attribute analyses and experiments: total energy, maximum energy, decay gradient seismic attributes, these 3 attributes are evident in high frequency decay characteristics. The preferred sensitive gas-bearing properties of each segment are substantially consistent with single well log interpretation results, with the gas-bearing sensitive properties being the seismic response of the gas bearing of the zone of interest.

The invention can obtain the distribution characteristics of the pre-stack AVO gas-containing probability high-value region and the distribution characteristics of the post-stack attenuation attribute high-value region by carrying out pre-stack and post-stack gas-containing detection on the reservoir.

In this embodiment, the gas-containing property of each gas layer group was detected from different angles by using the pre-stack AVO property analysis and the post-stack high-frequency energy attenuation property. From the detection results, although the detection results are very similar to the interpretation results of the well logging of the well completion, limitations and multiple solutions exist at different degrees, the total coincidence rate is more than 75%, and multiple information fusion is an important way for improving the gas-containing prediction accuracy. And the gas-containing boundary of each layer group is comprehensively determined according to a pre-stack crack prediction result, a post-stack crack prediction result, a pre-stack hydrocarbon detection result, a post-stack hydrocarbon detection result, a beneficial phase zone characterization result, a pre-stack inversion result, a post-stack inversion result and other results, so that the prediction precision can be further improved.

The multi-information fusion is an effective method for improving the gas-containing detection precision, and can obtain multi-information fusion response under the constraint of a known well. And from the view of multiple information fusion prediction results, the prediction precision is higher than the prediction effect of single attribute information.

6. In this embodiment, the determining the boundary of the sixth effective reservoir specifically includes the following substeps:

and (3) determining the boundary of the effective reservoir, and comprehensively predicting by adopting a G & G multiple information fusion technology. The gas-containing range determined by the fusion of various information is to take a favorable reservoir development area, a crack development area, a structure and fault development area, a Li Yan phase zone, a hydrocarbon detection favorable area and a Lei Kou slope group ancient structure high position as fusion parameters, and an effective reservoir thickness map is formed under the constraint of well point data.

The G & G is a comprehensive reservoir research software platform integrating layer sequence stratum interpretation, earthquake sediment imaging, reservoir sediment phase research and three-dimensional phase modeling, and provides a powerful lithofacies modeling tool with 3D attribute extraction, well earthquake combination layer sequence stratum interpretation and full three-dimensional visualization for users, and the resolution and the accuracy degree are over imagination. Multiple information fusion imaging is one of the functions, and the multi-attribute fusion imaging comprises the following parts:

1) Digital image fusion method based on RGB-IHS transformation;

2) Performing histogram equalization and normalization processing on the optimized and extracted three reflection deposition attributes;

3) Dividing each attribute into 16-256 equal parts according to gray scale, and designating specific attribute colors for three-channel fusion treatment;

4) The color classification after treatment is improved to 1600 ten thousand colors from 256 colors before treatment, and the transverse change characteristics of the reservoir can be reflected finely;

5) The multi-attribute fusion imaging algorithm also comprises image fusion based on wavelet transformation, image fusion based on Gaussian filtering and the like.

7. In this embodiment, the step seven is effective reservoir classification evaluation, and specifically includes the following substeps:

in this embodiment, according to single well gas testing and generating data of the reservoir, different types of reservoir logging classification evaluation standards are determined, and three types of classification evaluation are performed on the predicted effective reservoir after multiple information fusion. The specific classification criteria for an effective reservoir are shown in table 1 below:

TABLE 1 effective reservoir logging classification criteria table

Reservoir class	Sound wave time difference (mu s/ft)	Lithology Density (g/cm) ³ )
			Class I reservoirs	≥69	≤2.47
Class II reservoirs	66～69	2.47～2.52
			Class III reservoirs	63～66	2.52～2.57

The foregoing has shown and described the basic principles and main features of the present invention and the advantages of the present invention. It will be understood by those skilled in the art that the present invention is not limited to the embodiments described above, and that the above embodiments and descriptions are merely illustrative of the principles of the present invention, and various changes and modifications may be made without departing from the spirit and scope of the invention, which is defined in the appended claims. The scope of the invention is defined by the appended claims and equivalents thereof.

Claims

1. The comprehensive prediction method for the seismic reservoir is characterized by comprising the following steps of:

step seven: the method comprises the steps of (1) classifying and evaluating an effective reservoir, determining different types of reservoir logging classifying and evaluating standards according to single well gas testing and generating data of the reservoir, and classifying and evaluating the predicted effective reservoir after various information fusion;

The third step specifically comprises the following substeps:

s304, performing seismic waveform indication inversion on each reservoir by using SMI inversion software according to the sensitivity curve, the sensitivity parameter and the sensitivity seismic attribute which are preferably selected in the step S303, and obtaining a waveform indication inversion result of the reservoir;

The fourth step further includes: and (3) taking the anticline structure as a mechanical model, carrying out numerical simulation analysis on the stress field of the reservoir, taking the maximum curvature value of a point on the anticline structure surface as a criterion of the crack development degree of the point, and simultaneously indicating the possible crack trend with the minimum principal curvature direction to obtain the development degree and the spreading relation of the reservoir cracks.

2. The method for comprehensive prediction of a seismic reservoir according to claim 1, wherein the first step specifically comprises:

selecting the top and bottom layer positions of the reservoir as fixed time windows to extract the seismic attribute values in the clastic rock deposition units of the target interval; according to the spreading characteristics of the reservoir on the plane, comparing and analyzing the extracted seismic attribute values, preferably identifying the sedimentary phase-sensitive seismic attribute and the gas-containing sensitive seismic attribute, taking the sedimentary phase-sensitive seismic attribute and the gas-containing sensitive seismic attribute as basic parameters, carrying out correlation analysis by combining the well reservoir characteristics, determining the seismic attribute with good correlation with the reservoir, and carrying out qualitative prediction of the favorable phase zone by utilizing the seismic attribute with good correlation with the reservoir.

3. The method for comprehensive prediction of a seismic reservoir according to claim 1, wherein the second step specifically comprises the following sub-steps:

4. The method for comprehensive prediction of a seismic reservoir according to claim 1, wherein the fourth step specifically comprises the following sub-steps:

5. The method for comprehensive prediction of a seismic reservoir according to claim 1, wherein the fifth step specifically comprises: carrying out pre-stack AVO analysis based on the gas-containing sensitive seismic attribute selected in the first step to obtain the distribution characteristics of a high-value region of the gas-containing probability of the reservoir pre-stack AVO; and carrying out post-stack gas-containing detection based on the gas-containing sensitive seismic attribute, extracting the attenuation gradient attribute of the high frequency band of the seismic wave, and obtaining the distribution characteristics of the high-value area of the attenuation attribute after the storage layer.