CN108957532B

CN108957532B - Reservoir stratum prediction method and device

Info

Publication number: CN108957532B
Application number: CN201810598725.3A
Authority: CN
Inventors: 李勇根; 李红兵; 董世泰; 徐右平; 马晓宇; 潘豪杰
Original assignee: China Petroleum and Natural Gas Co Ltd
Current assignee: China Petroleum and Natural Gas Co Ltd
Priority date: 2018-06-12
Filing date: 2018-06-12
Publication date: 2020-01-07
Anticipated expiration: 2038-06-12
Also published as: CN108957532A

Abstract

The invention provides a reservoir prediction method and a device, and the method comprises the following steps: performing phase control modeling based on the seismic data to obtain an initial impedance model of a reservoir to be predicted; synthesizing the initial impedance model to obtain a seismic record corresponding to the initial impedance model; and judging whether the reservoir prediction result of the reservoir to be predicted can be output according to the initial impedance model or not according to the seismic record corresponding to the initial impedance model, and outputting the reservoir prediction result according to the initial impedance model under the condition that the judgment result is yes. The invention can improve the reservoir prediction resolution.

Description

Reservoir stratum prediction method and device

Technical Field

The invention relates to the technical field of oil and gas exploration, in particular to a reservoir prediction method and device.

Background

Along with the complication of oil and gas exploration and development objects, the oil and gas exploration and development gradually turns from constructive oil and gas reservoir development to complex hidden lithology and unconventional oil and gas reservoir development, and meanwhile, the resolution and precision of reservoir prediction by utilizing seismic data are required to be higher and higher. Three prominent difficulties mainly exist in the current characterization of thin reservoirs: (1) the thickness of the single layer of the reservoir is thin, and the single layers are longitudinally overlapped; (2) the reservoir stratum has fast transverse change and strong uncertainty among wells; (3) low porosity and low permeability, various pore types and complex oil-water relationship. Therefore, how to improve the prediction resolution and accuracy of the thin reservoir by using the seismic inversion technology is one of the hot and difficult problems in exploration geophysical research.

Over the past few decades, seismic inversion techniques have been developed and are widely used in the field of oil and gas exploration. Lindseth proposed a recursion inversion method in 1979, which is established on the basis of the post-stack one-dimensional wave impedance inversion of a convolution model, and rewrites wave impedance information into a form of direct summation of a reflection coefficient sequence of an underground stratum; an inversion method based on a model is proposed by Cook and Schneider in 1983, and the method comprises the steps of firstly establishing an initial impedance model containing low frequency, medium frequency and high frequency by using logging and seismic interpretation results, then manufacturing a synthetic seismic section by using a forward algorithm, comparing a synthetic record with an actual seismic section, repeatedly correcting the initial model until the error between the synthetic record and the actual seismic section is minimum, and finally correcting the impedance model to be an inversion result; the method comprises the steps of firstly obtaining a transverse variation function on the basis of constraint sparse pulse inversion, then calculating probability density functions and vertical variation functions of different lithologic wave impedances, finally taking seismic data as hard constraints, generating inter-well wave impedance and corresponding synthetic seismic traces through Markov chain Monte Carlo simulation, and repeatedly and iteratively calculating by using a nonlinear optimization solution method until the synthetic seismic traces are well matched with original seismic data to obtain a final inversion body.

The above three existing typical inversion methods each have advantages and disadvantages: the recursive inversion method has high calculation speed, but the resolution is low due to the limitation of seismic frequency bands, and thin layers cannot be predicted; the model inversion method and the geostatistical inversion method have higher resolution, but the low frequency and the high frequency in the inversion are from logging information, the difference between the two methods is mainly that the modeling methods are different, and different modeling algorithms and modeling parameters easily cause the multi-solution of the inversion result, and the two methods have certain advantages for identifying the thin sand bodies, but are only suitable for areas with stable deposition, more wells and more uniform distribution.

Disclosure of Invention

The invention provides a reservoir prediction method and a reservoir prediction device, which are used for improving the resolution of reservoir prediction.

The embodiment of the invention provides a reservoir prediction method, which comprises the following steps: performing phase control modeling based on the seismic data to obtain an initial impedance model of a reservoir to be predicted; synthesizing the initial impedance model to obtain a seismic record corresponding to the initial impedance model; and judging whether the reservoir prediction result of the reservoir to be predicted can be output according to the initial impedance model or not according to the seismic record corresponding to the initial impedance model, and outputting the reservoir prediction result according to the initial impedance model under the condition that the judgment result is yes.

The embodiment of the invention also provides a reservoir prediction device, which comprises: a phase control modeling unit to: performing phase control modeling based on the seismic data to obtain an initial impedance model of a reservoir to be predicted; a seismic record generation unit to: synthesizing the initial impedance model to obtain a seismic record corresponding to the initial impedance model; a prediction result determination unit configured to: and judging whether the reservoir prediction result of the reservoir to be predicted can be output according to the initial impedance model or not according to the seismic record corresponding to the initial impedance model, and outputting the reservoir prediction result according to the initial impedance model under the condition that the judgment result is yes.

Embodiments of the present invention further provide a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the steps of the method described in the above embodiments.

The embodiment of the present invention further provides a computer device, which includes a memory, a processor, and a computer program stored in the memory and capable of running on the processor, and when the processor executes the computer program, the steps of the method described in the above embodiment are implemented.

According to the reservoir prediction method, the reservoir prediction device, the computer readable storage medium and the computer equipment, the phase-controlled modeling is performed based on the seismic data to obtain the initial impedance model of the reservoir to be predicted, and the simple extrapolation and interpolation modeling is performed under the seismic horizon constraint by using the logging data. In the phase-controlled modeling process, the internal relation between the geological deposition background and the seismic reflection can be fully considered, so that the problem that the logging data is excessively depended when the logging data is used for carrying out extrapolation and interpolation modeling under the constraint of the seismic horizon can be solved, the phase-controlled modeling has higher modeling precision, and the high-precision impedance model can improve the resolution ratio of reservoir prediction.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts. In the drawings:

fig. 1 is a schematic flow diagram of a prior art reservoir prediction method.

Fig. 2 is a schematic flow chart of a reservoir prediction method according to an embodiment of the invention.

FIG. 3 is a schematic flow chart of a method for performing phase-controlled modeling based on seismic data to obtain an initial impedance model of a reservoir to be predicted according to an embodiment of the present invention.

FIG. 4 is a schematic flow chart of a method for obtaining seismic facies of a reservoir to be predicted by facies prediction based on well-logging facies and using a plurality of seismic attributes in one embodiment of the invention.

Fig. 5 is a schematic flow chart of a reservoir prediction method according to another embodiment of the present invention.

FIG. 6 is a schematic flow chart of a method for performing high resolution processing on initial seismic data based on reservoir characteristics to obtain seismic data according to an embodiment of the invention.

Fig. 7 is a schematic flow chart of a method for determining whether a reservoir prediction result of a reservoir to be predicted can be output according to an initial impedance model in an embodiment of the present invention.

Fig. 8 is a schematic flow chart of a reservoir prediction method according to yet another embodiment of the present invention.

Fig. 9 is a schematic flow chart of a reservoir prediction method according to still another embodiment of the present invention.

Fig. 10 is a schematic flow chart of a reservoir prediction method according to an embodiment of the invention.

Fig. 11 is a graph comparing prior reservoir predictions for blind well a with reservoir predictions of the method of the present invention.

FIG. 12 is a graph comparing prior art reservoir predictions for a posterior B-well and reservoir predictions for the method of the present invention.

Fig. 13 is a schematic structural diagram of a reservoir prediction apparatus according to an embodiment of the present invention.

FIG. 14 is a schematic diagram of the structure of a phased modeling unit in an embodiment of the invention.

FIG. 15 is a schematic diagram of the structure of a seismic lithofacies generation module in an embodiment of the present invention.

Fig. 16 is a schematic structural diagram of a reservoir prediction apparatus according to another embodiment of the present invention.

Fig. 17 is a schematic structural diagram of a high resolution processing unit according to an embodiment of the invention.

Fig. 18 is a schematic structural diagram of a prediction result determining unit according to an embodiment of the present invention.

Fig. 19 is a schematic structural diagram of a reservoir prediction apparatus according to still another embodiment of the present invention.

Fig. 20 is a schematic structural diagram of a reservoir prediction apparatus according to still another embodiment of the present invention.

Detailed Description

In order to make the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the embodiments of the present invention are further described in detail below with reference to the accompanying drawings. The exemplary embodiments and descriptions of the present invention are provided to explain the present invention, but not to limit the present invention.

Fig. 1 is a schematic flow diagram of a prior art reservoir prediction method. With reference to fig. 1, in the prior art, a synthetic record is obtained by interpolating or extrapolating logging data under horizon control, and whether an error meets a requirement is judged by comparing the synthetic record with an actual seismic record, and by analyzing the principles and application effects of the prior art, it is known that the high resolution of the seismic inversion method in the prior art mainly comes from logging information, and when reservoir prediction is performed, most of high-frequency parts of seismic inversion are obtained by interpolating and extrapolating high-frequency components of the logging data or by geostatistical random simulation, the dependence on an extrapolation model is strong, and the seismic data themselves have a small effect on high-resolution constraint of the extrapolation model, which results in strong multi-resolution of inversion results and poor prediction effect on areas with few wells, strong reservoir heterogeneity and fast change. Aiming at the problems in the prior art, the inventor provides a reservoir prediction method by fully mining the information of the earthquake based on the earthquake data. The process of reservoir prediction may be a seismic inversion process, so the reservoir prediction method may be referred to as a seismic inversion method.

Fig. 2 is a schematic flow chart of a reservoir prediction method according to an embodiment of the invention. As shown in fig. 2, the reservoir prediction method of the present embodiment may include:

step S110: performing phase control modeling based on the seismic data to obtain an initial impedance model of a reservoir to be predicted;

step S120: synthesizing the initial impedance model to obtain a seismic record corresponding to the initial impedance model;

step S130: and judging whether the reservoir prediction result of the reservoir to be predicted can be output according to the initial impedance model or not according to the seismic record corresponding to the initial impedance model, and outputting the reservoir prediction result according to the initial impedance model under the condition that the judgment result is yes.

In step S110, the seismic data may be raw seismic data or seismic data subjected to high resolution processing. Phased modeling may be performed using existing methods or improved methods. The initial impedance model may include initial impedances corresponding to different seismic traces. In the phased modeling process, geological depositional background and seismic reflection intrinsic relations can be fully considered.

In step S120, the initial impedances of the different seismic traces are synthesized to obtain a seismic record, i.e., a synthetic record. The seismic records obtained from different initial impedance models are different.

In the step S130, an existing method may be used, that is, a difference between the seismic record and the actual seismic record is used as an error, and whether to output the reservoir prediction result of the reservoir to be predicted according to the initial impedance model is determined according to the error. Or, the seismic records are subjected to attribute analysis, an error is determined according to the attribute analysis result, and whether the reservoir prediction result of the reservoir to be predicted is output according to the initial impedance model or not is judged according to the error. The reservoir predictions may include reservoir thickness, reservoir property parameters, which may include, for example, porosity, pore size, and the like. Reservoir prediction results may be obtained directly or indirectly from the initial impedance model.

In the embodiment, the initial impedance model of the reservoir to be predicted is obtained by performing phase-controlled modeling based on the seismic data, rather than performing simple extrapolation and interpolation modeling under the seismic horizon constraint by using logging data, so that the problem of over dependence on the logging data can be avoided. In the phase-controlled modeling process, the internal relation between the geological deposition background and the seismic reflection can be fully considered, so that the phase-controlled modeling has higher modeling precision, and the high-precision impedance model can improve the resolution of reservoir prediction.

FIG. 3 is a schematic flow chart of a method for performing phase-controlled modeling based on seismic data to obtain an initial impedance model of a reservoir to be predicted according to an embodiment of the present invention. As shown in fig. 3, the method for performing phase-controlled modeling based on seismic data to obtain an initial impedance model of a reservoir to be predicted in step S110 may include:

step S111: performing attribute analysis by using the seismic data to obtain a plurality of seismic attributes;

step S112: based on the well logging lithofacies, performing lithofacies prediction by using the plurality of seismic attributes to obtain seismic lithofacies of the reservoir to be predicted;

step S113: and establishing an initial impedance model of the reservoir to be predicted under the phase control constraint of the seismic lithofacies.

In step S111, the plurality of seismic attributes may include an amplitude class, a frequency class, and the like. In step S112, the well-logging lithofacies may refer to lithofacies of known wells. Based on the well logging lithofacies, the lithofacies of the reservoir to be predicted, such as the lithofacies of the pseudo-well position, can be deduced by utilizing the plurality of seismic attributes, and the seismic lithofacies can be obtained. In the step S113, the internal relation between the geological depositional background and the seismic reflection is established by establishing an initial impedance model of the reservoir to be predicted under the phased constraint of the seismic facies.

In the embodiment, the lithofacies are predicted by using various seismic attributes based on the well logging lithofacies, the variation trends of different lithofacies at different depths are analyzed, and the high-resolution and high-precision impedance model is established under the phase control constraint, so that the phase control modeling method by using various information of the applied seismic data is realized. The process of performing the phase-controlled modeling may be a process of performing reservoir modeling, and a reservoir parameter model may be obtained from the impedance model.

In other embodiments, in step S110, the method for performing phase-controlled modeling based on the seismic data to obtain the initial impedance model of the reservoir to be predicted may be implemented by using methods such as deconvolution processing and frequency broadening processing.

FIG. 4 is a schematic flow chart of a method for obtaining seismic facies of a reservoir to be predicted by facies prediction based on well-logging facies and using a plurality of seismic attributes in one embodiment of the invention. As shown in fig. 4, in step S112, the method for performing facies prediction by using the plurality of seismic attributes based on the well-logging facies to obtain the seismic facies of the reservoir to be predicted may include:

step S1121: based on geological background, carrying out logging lithofacies analysis by using logging data and interpretation data thereof to obtain logging lithofacies;

step S1122: analyzing by utilizing the plurality of seismic attributes based on the logging lithofacies, and establishing a relation between the logging lithofacies and the seismic attributes;

step S1123: and outputting the seismic facies of the reservoir to be predicted according to the relation between the logging facies and the seismic attributes.

In step S1121, the geological context may include the geological conditions of the reservoir, such as rocks, rock formations, formation environments, etc. The well log lithofacies are lithofacies of the reservoir where the well is known to be located. In step S1122 above, the relationship between the log facies and the seismic attributes may be a linear or non-linear relationship. In step S1123, the seismic facies at the pseudo well location may be inferred from the relationship between the well log facies and the seismic attributes.

In the embodiment, the phased modeling based on seismic multi-attributes is conducted by taking geological background as guidance, logging lithofacies analysis is conducted through various logging and interpretation data, linear or nonlinear relation between the logging lithofacies and seismic attributes is established by utilizing various seismic attribute analysis, and impedance modeling under the phased guidance is conducted on the seismic lithofacies based on seismic attribute prediction.

Fig. 5 is a schematic flow chart of a reservoir prediction method according to another embodiment of the present invention. As shown in fig. 5, before the step S110, that is, before performing phase-controlled modeling based on seismic data to obtain an initial impedance model of a reservoir to be predicted, the reservoir prediction method shown in fig. 2 may further include:

step S140: and carrying out high-resolution processing on the initial seismic data based on the reservoir characteristics to obtain the seismic data.

In step S140, the high resolution processing is performed based on the reservoir characteristics, and the high resolution processing is performed on the seismic data of the target interval, not on the entire seismic section, so that the method has stronger pertinence. The high resolution processing may be implemented using various existing high resolution processing methods, or using modified high resolution processing. The initial seismic data may have a higher signal quality after being processed with high resolution. Under the condition that the original seismic data is low in resolution, high-resolution processing is carried out on the original seismic data based on reservoir characteristic constraint, and the resolution of the seismic data can be improved.

FIG. 6 is a schematic flow chart of a method for performing high resolution processing on initial seismic data based on reservoir characteristics to obtain seismic data according to an embodiment of the invention. As shown in fig. 6, the step S140 of performing high resolution processing on the initial seismic data based on the reservoir characteristics to obtain the seismic data may include:

step S141: establishing a relation between reservoir thickness and dominant frequency and bandwidth by performing time-frequency analysis on initial seismic data of a target layer in a reservoir to be predicted and performing variable reservoir thickness model forward modeling on the target layer;

step S142: and adjusting frequency broadening parameters based on the relation between the reservoir thickness and the dominant frequency and the frequency width so as to suppress interference noise in the initial seismic data and improve the frequency width of effective signals in the initial seismic data.

In the step S141, the thickness of the reservoir may be preliminarily determined by performing time-frequency analysis on the initial seismic data of the target layer in the reservoir to be predicted, the thickness of the reservoir may be further predicted by performing variable reservoir thickness model forward modeling on the target layer, and a reasonable relationship between the reservoir thickness and the dominant frequency and bandwidth is established by integrating the time-frequency analysis result and the variable reservoir thickness model forward modeling result. The time-frequency analysis and the forward modeling of the variable reservoir thickness model are carried out aiming at a target layer, but not aiming at the whole seismic section, so the high-resolution processing method of the embodiment considers the reservoir characteristics and has more pertinence.

In step S142, the relationship between the reservoir thickness and the dominant frequency and the reservoir width includes a relationship between the reservoir thickness and the dominant frequency and a relationship between the reservoir thickness and the reservoir width. The frequency extending parameters may include, for example, dominant frequency, bandwidth, etc. Because of the adjustment basis, the resolution of the initial seismic data can be theoretically improved by increasing the frequency extension parameter, but in practice, the frequency extension parameter cannot be increased without limit due to the interference noise. And adjusting the frequency broadening parameters based on the relation between the reservoir thickness and the dominant frequency and the frequency width, and simultaneously considering suppressing the interference noise in the initial seismic data and improving the frequency width of the effective signal in the initial seismic data, so that the aim of reasonably adjusting the frequency broadening parameters can be achieved.

In the embodiment, high-resolution processing is performed based on reservoir characteristic constraints, and mainly time-frequency analysis is performed on an original seismic data target layer, variable thickness model forward modeling is performed on the target layer, the relation between reservoir thickness and main frequency and bandwidth is established, and frequency extension processing parameters are adjusted under the condition that a signal-to-noise ratio is considered, so that the purpose of improving effective signal bandwidth on the premise of suppressing interference noise as much as possible is achieved. The effective seismic frequency band range suitable for local reservoir prediction is determined through reservoir variable thickness model seismic forward technical analysis, the energy of effective signals is improved as much as possible by using a frequency extension processing principle on the basis of keeping the signal-to-noise ratio of original seismic data, and the main frequency and the frequency width of the processed seismic data can be obviously improved. Therefore, high resolution processing of seismic data is one of the main reasons that inversion results of reservoir predictions have high resolution. The problem of high frequency components of high resolution mainly from logging data can be avoided by high resolution processing.

Fig. 7 is a schematic flow chart of a method for determining whether a reservoir prediction result of a reservoir to be predicted can be output according to an initial impedance model in an embodiment of the present invention. As shown in fig. 7, the method for determining whether the reservoir prediction result of the reservoir to be predicted can be output according to the initial impedance model according to the seismic record corresponding to the initial impedance model in step S130 may include:

step S131: performing seismic attribute analysis by using the seismic records corresponding to the initial impedance model to obtain a plurality of initial seismic attributes;

step S132: weighting and summing a plurality of effective seismic attributes in the plurality of initial seismic attributes according to corresponding weight coefficients to obtain weighted predicted seismic attributes;

step S133: calculating a seismic attribute error according to the weighted predicted seismic attribute and the weighted actual seismic attribute; the weighted actual seismic attributes are obtained by weighting and summing the effective seismic attributes in the actual seismic attributes according to the corresponding weight coefficients, and the actual seismic attributes are obtained by performing attribute analysis according to the seismic data;

step S134: and judging whether the reservoir prediction result of the reservoir to be predicted can be output according to the initial impedance model or not according to whether the seismic attribute error is smaller than a set error value or not.

And judging whether the reservoir prediction result of the reservoir to be predicted can be output according to the initial impedance model or not based on the weighting and setting constraint conditions of the seismic attributes, instead of setting the constraint conditions by only considering a single seismic attribute or independently considering a plurality of seismic attributes, so that the accuracy of reservoir prediction can be improved, and meanwhile, the multi-solution of the inversion result can be effectively reduced.

In this embodiment, based on the initial impedance model under the phase control guidance, a synthetic seismic profile is manufactured by using an alignment algorithm, an effective seismic attribute combination is extracted, and the attributes are combined and weighted with attributes corresponding to actual seismic traces to obtain an error. The error is determined through weighted comparison of various attribute information, various seismic attributes are comprehensively considered, and the error is not determined only according to synthetic records, so that the error determined by the method is more accurate, the accuracy of reservoir prediction can be further improved, and the multi-solution of the inversion result is effectively reduced.

In the embodiment, the initial impedance model can be repeatedly corrected until the weighted error of the combination of the seismic attributes of the synthesized seismic records and the combination of the attributes of the actual seismic traces is minimum, and the obtained impedance model is used as a final inversion result, so that the reservoir prediction can meet the requirement of the set error, and the resolution of the reservoir prediction can be further improved.

Fig. 8 is a schematic flow chart of a reservoir prediction method according to yet another embodiment of the present invention. As shown in fig. 8, the reservoir prediction method shown in fig. 2 may further include:

step S150: modifying the initial impedance model to obtain a modified impedance model under the condition that the reservoir prediction result of the reservoir to be predicted cannot be output according to the initial impedance model;

step S120': synthesizing the modified impedance model to obtain a seismic record corresponding to the modified impedance model;

step S130': and judging whether the reservoir prediction result of the reservoir to be predicted can be output according to the modified impedance model or not according to the seismic record corresponding to the modified impedance model, and if so, outputting the reservoir prediction result according to the modified impedance model.

The above-mentioned steps S120 'and S130' are processes of repeatedly performing the steps S120 and S130 by replacing the initial impedance model with the modified impedance model. The above-described embodiments of step S120 'and step S130' may be performed with reference to the embodiments of step S120 and step S130. In the process of obtaining the initial impedance model and modifying the initial impedance model, the reservoir parameter model has been quantified, so that the reservoir prediction of quantification can be performed by the embodiment.

In this embodiment, the reservoir prediction result of the reservoir to be predicted can be output according to the modified impedance model by continuously modifying the impedance model, re-synthesizing the seismic records, and re-judging whether the reservoir prediction result can be output according to the modified impedance model, and sequentially iterating until the judgment result is yes, and outputting the reservoir prediction result according to the modified impedance model. And the impedance model meeting the requirements can be found by modifying the impedance model through optimization iteration.

Fig. 9 is a schematic flow chart of a reservoir prediction method according to still another embodiment of the present invention. As shown in fig. 9, the reservoir prediction method shown in fig. 2 may further include:

step S160: determining the plurality of valid seismic attributes and the corresponding weight coefficients based on well log data and petrophysical theory.

In the step S160, obtaining the effective seismic attributes and the corresponding weighting coefficients based on the log data and the rock physics theory may include:

step S161: generating a geological model of the reservoir to be predicted according to the logging data;

step S162: establishing a geophysical model of the reservoir to be predicted according to the geological model;

step S163: establishing a reservoir parameter model of the reservoir to be predicted based on a rock physics theory and the geophysical model;

step S164: and screening the plurality of effective seismic attributes sensitive to the reservoir according to the reservoir parameter model, and setting the weight coefficient for each effective seismic attribute according to the degree of sensitivity to the reservoir.

In the step S163, specifically, under the guidance of petrophysical, a variation relationship between the reservoir property and the reservoir propagation velocity may be established based on the geophysical model, a relationship between the property, the velocity, and the density may be further obtained, and an impedance model may be established according to the relationship between the velocity and the density, so as to obtain the reservoir parameter model.

In step S164, a synthetic record may be obtained according to the impedance model, an attribute analysis may be performed according to the synthetic record to obtain a seismic attribute, an intersection of the seismic attribute and the reservoir physical property is used to determine which seismic attribute is sensitive to the reservoir thickness, and a corresponding weight coefficient may be assigned to the seismic attribute according to the degree of sensitivity of the seismic attribute to the reservoir thickness. The seismic attributes may be, for example, amplitude-based attributes, frequency-based attributes, and the like. Determining the main frequency and bandwidth of the seismic data after high-resolution processing, and then determining a variable thickness and variable physical property parameter model to obtain an attribute set sensitive to a reservoir.

In the embodiment, a petrophysical theory is used as a guide, physical parameter models of reservoir thickness variation, porosity variation and the like are developed to forward the earthquake, forward seismic channel attributes, reservoir thickness and physical parameter variation rules are analyzed, multiple effective seismic attribute combinations are preferably selected according to the correlation, different weight coefficients are given to the preferably selected seismic attributes, the larger the correlation, the larger the weight coefficient is, and the smaller the correlation is. Due to the consideration of the rock physics theory, the established reservoir parameter model of the reservoir to be predicted is more accurate. On the basis of reservoir characteristic-based model seismic forward modeling, various effective attributes are preferably selected, inversion convergence conditions are jointly constrained by using weighted iteration of the various effective seismic attributes, and the limit that conventional impedance model inversion only uses seismic amplitude information as convergence constraints is improved. Due to the fact that seismic information related to various attributes and reservoir changes is considered, the multi-solution of the inversion result can be effectively reduced while the inversion accuracy is improved.

The purpose, implementation and effect of the present invention will be described below with reference to a specific embodiment.

The exploration practice results show that: the high frequency part of the existing seismic inversion is mostly the result of interpolation, extrapolation of the high frequency components of the logging data, or random simulation through geology statistics. The seismic data have little effect on the high-resolution constraint of the model, so that the inversion result has strong multi-solution performance, and the prediction effect on areas with few wells, strong reservoir heterogeneity and quick change is poor. Based on this, the present embodiment provides a reservoir quantitative prediction method, that is, a high-resolution seismic inversion method based on the multidimensional information constraint of reservoir characteristics. Based on seismic data, the information of the earthquake is fully mined, and a seismic inversion method which does not depend on logging data excessively is developed, so that the multi-resolution of the inversion result is reduced, the longitudinal resolution capability and the transverse change prediction capability of the thin layer are improved, and the quantitative characterization capability of the physical properties of the reservoir layer has important theoretical significance and practical value. The method of the embodiment inherits the basic flow of the conventional reservoir prediction method and simultaneously greatly improves and perfects some key technical aspects. For example, under the condition that the original seismic data is low in resolution, high-resolution processing is performed on the seismic data based on reservoir characteristic constraints; in the aspect of reservoir modeling, phase control modeling is carried out by applying a plurality of information of seismic data; and on the condition of inversion iterative convergence, comprehensively weighting and summing seismic attribute classes such as amplitude classes, frequency classes and the like to calculate errors, and taking the errors as iterative constraint conditions. Finally, by the improved technologies, the reservoir prediction resolution and precision can be effectively improved, the dependence of the inversion result on logging data is greatly reduced, and the method is suitable for areas with less logging data, strong reservoir non-uniformity and quick change.

Fig. 10 is a schematic flow chart of a reservoir prediction method according to an embodiment of the invention. As shown in fig. 10, the reservoir prediction method of the present embodiment is a high-resolution seismic inversion method based on the multidimensional information constraint of reservoir features, and its improvement point mainly includes four aspects: firstly, high-resolution processing based on reservoir characteristic constraint is performed, wherein the content of the aspect is mainly to perform time-frequency analysis on an original seismic data target layer, develop variable thickness model forward modeling aiming at the target layer, establish the relation between reservoir thickness and main frequency and bandwidth, and adjust frequency expansion processing parameters under the condition of considering signal-to-noise ratio so as to achieve the purpose of improving effective signal bandwidth on the premise of suppressing interference noise as much as possible; secondly, based on the earthquake multi-attribute phased modeling technology, the main content of the aspect is to use geological background as guidance, carry out logging lithofacies analysis through various logging and interpretation data (known logging data), utilize various earthquake attribute analyses to establish the linear or nonlinear relation between logging lithofacies and earthquake attributes, and carry out impedance modeling under phased guidance based on earthquake lithofacies predicted by earthquake attributes; thirdly, using a rock physics theory as a guide, developing a reservoir variable thickness and variable porosity parameter model earthquake forward modeling, analyzing forward seismic trace attributes and reservoir thickness and physical parameter change rules, preferably selecting various effective earthquake attribute combinations according to the magnitude of the correlation, and endowing different weight coefficients to the selected earthquake attributes, wherein the larger the correlation is, the larger the weight coefficient is, and the smaller the correlation is otherwise; and fourthly, based on the initial impedance model under the phase control guidance, manufacturing a synthetic seismic section by using a forward algorithm and extracting an effective seismic attribute combination, performing weighted comparison on the attributes and the attribute combination corresponding to the actual seismic channel, and repeatedly correcting the initial model until the weighted error between the synthetic recorded seismic attribute combination and the actual seismic channel attribute combination is minimum, wherein the obtained impedance model is a final inversion result.

The method of the embodiment inherits the advantages of the conventional model-based inversion, and simultaneously provides corresponding technical improvement aiming at the problems commonly existing in the current model inversion, and achieves better effects mainly in the following three aspects:

first, the method of this embodiment has a higher seismic inversion resolution. The effective seismic frequency band range suitable for local reservoir prediction is determined through reservoir variable thickness model seismic forward technical analysis, the energy of effective signals is improved as much as possible by using a frequency extension processing principle on the basis of keeping the signal-to-noise ratio of original seismic data, and the main frequency and the frequency width of the processed seismic data are obviously improved. Therefore, the high-resolution processing of the seismic data is one of the main reasons for the high resolution of the inversion result of the method, and the problem that the high resolution of the conventional inversion mainly comes from the high-frequency components of the logging data is avoided.

Second, the method of this embodiment has a higher modeling accuracy. The high accuracy of the initial impedance model is the key to the model inversion. In the embodiment, the lithofacies is predicted by utilizing various seismic attributes on the basis of the logging lithofacies, the variation trends of different lithofacies at different depths are analyzed, and a high-resolution and high-precision impedance model is established under the phase control constraint. In the modeling process, geological deposition background and seismic reflection internal relation are fully considered, and the problems of simple extrapolation and interpolation modeling under seismic horizon constraint by using logging information are avoided, so that the method has higher modeling precision and lays a foundation for high-resolution reservoir prediction.

Third, the method of this embodiment can effectively reduce the ambiguity of the inversion result. According to the method, on the basis of reservoir characteristic-based model seismic forward modeling, various effective attributes are preferably selected, a condition of using the various effective seismic attributes for weighting, iterating and jointly constraining inversion convergence is innovatively provided, and the limitation that conventional impedance model inversion only uses seismic amplitude information as convergence constraint is improved. Due to the fact that various seismic information related to reservoir changes are considered, the inversion method of the embodiment can improve inversion accuracy and effectively reduce the multi-solution of inversion results.

The method of the embodiment comprises the following main steps: firstly, establishing a reservoir characteristic parameter earthquake forward modeling by a geological model, and preferably selecting a reservoir parameter sensitive earthquake attribute set; and secondly, performing time-frequency analysis and high-resolution processing on the seismic data, on the basis of establishing an initial impedance model under phase control constraint, making a synthetic record and analyzing a seismic attribute set of the synthetic record by a seismic forward modeling method, performing weighted comparison with an actual seismic data attribute set, and modifying the impedance model by using an iterative optimization algorithm until the weighted error of the two attribute sets is minimum. On the basis of the traditional seismic inversion thought, a plurality of seismic attributes such as amplitude, frequency, phase and the like and reservoir related information are introduced as constraint conditions, through an iterative optimization and attribute weighting method, the seismic attribute weighting corresponding to the actual seismic data is optimally matched with the amplitude and other attribute information of the model forward synthetic seismic data, and the final model data is an inversion result. Therefore, the method of the embodiment can effectively improve the inversion resolution and prediction accuracy by using the new seismic inversion method under the constraint of the multidimensional information based on the reservoir characteristics, and is suitable for areas with less logging data, strong reservoir heterogeneity and fast change.

In an application embodiment, the method provided by the embodiment of the invention is used for carrying out a reservoir prediction test on a quaternary clastic rock thin interbed in a certain block of an oil field in China. The buried depth of a target layer of the block is 200-1000 m, and the deposition environment is the deposition of a shore lake phase, and the storage and cover combination of the sand body of a beach dam of the shore lake and the shore lake mud is developed. Reservoir lithologic siltstone is mainly used, the sand shale thin interbed is very developed in the longitudinal direction, but the reservoir single-layer thickness is mainly 0.5-2 m, and the change of the transverse reservoir is fast. Fig. 11 is a graph comparing prior reservoir predictions for blind well a with reservoir predictions of the method of the present invention. As shown in fig. 11, (a) is the conventional model seismic inversion result of the blind well a, and (b) is the high-resolution seismic inversion result of the blind well a under the multi-dimensional information constraint based on reservoir characteristics. The comparison in fig. 11 shows that the inversion resolution of the conventional model is low, the reservoir distribution in the region cannot be predicted in the transverse direction, and the well point prediction result is poorly matched with the actual drilling; the method of the invention can show that the inversion resolution is obviously higher, the thickness and position of the reservoir predicted by the earthquake at the well point are better matched with the actual drilling, the reservoir distribution in the transverse direction is clear, the whole earthquake inversion section finely describes the law of strong transverse non-uniformity and quick change of the reservoir, and the prediction result is better matched with the actual geological knowledge. Inversion effect test is carried out on 15 blind wells in the whole area by using the method, and statistical results show that the predicted thickness coincidence rate of the reservoir is up to more than 92%.

In another application example, the method of the invention is used for carrying out reservoir prediction tests on a Jurassic carbonatite reservoir in a certain area of a foreign oil field. The buried depth of a target layer of the block is 3000-4000 m, the deposition environment is the slope phase deposition from the edge of a terrace to the front edge, and the biological reef and the beach are developed. Mainly using lithologic granular limestone of a target layer, and developing a set of thin mudstone with the thickness of 1-10 m close to the top of the target layer; three sets of huge thick paste rock layers developed above mudstone sandwich two sets of salt rocks, and the change of the thickness of the stratum is large. Affected by gypsum rock and salt rock, the seismic data quality of the local area is poor, the resolution and the signal-to-noise ratio are low, and the interpretation difficulty of the top structure of the target layer is high. Because the top mudstone is thin but develops in the whole area, the high-resolution seismic inversion technology is utilized to predict the set of mudstone, and the top structure of the target layer in the area is further realized.

FIG. 12 is a graph comparing prior art reservoir predictions for a posterior B-well and reservoir predictions for the method of the present invention. As shown in fig. 12, (a) part of the diagram is a conventional seismic profile of a posterior B well and the explanation of the top and bottom structures of a target layer, and (B) part of the diagram is a high-resolution mudstone prediction profile based on the multi-dimensional information constraint of reservoir characteristics. From analysis on the mudstone prediction section, the top structure of the target layer is red in the figure, and the difference from the prior explanation is large. And the well B is a posterior well, after well seismic calibration, the predicted position and thickness of the mudstone earthquake are well matched with actual drilling, and the thickness of the mudstone at the top of the target layer of the well is only 2 m. Inversion effect tests are carried out on 25 blind wells and 5 known wells in the whole area by using the method, and statistical results show that the coincidence rate of the predicted positions and thicknesses of the mudstones is up to more than 90%.

Therefore, the results of the two application example effect analyses show that the novel high-resolution seismic inversion method based on the reservoir characteristic under the multi-dimensional information constraint effectively improves the longitudinal resolution capability, the transverse change prediction capability and the reservoir physical property quantitative characterization capability of the thin layer, and has important theoretical significance and practical value for guiding the next exploration and development.

Based on the same inventive concept as the reservoir prediction method shown in fig. 2, the embodiment of the present application further provides a reservoir prediction apparatus, as described in the following embodiments. Because the principle of solving the problems of the reservoir prediction device is similar to that of the reservoir prediction method, the implementation of the reservoir prediction device can be referred to the implementation of the reservoir prediction method, and repeated details are not repeated.

Fig. 13 is a schematic structural diagram of a reservoir prediction apparatus according to an embodiment of the present invention. As shown in fig. 13, the reservoir prediction apparatus of the present embodiment may include: the phase control modeling unit 210, the seismic record generation unit 220, and the prediction result determination unit 230 are connected in this order.

A phase control modeling unit 210 for: performing phase control modeling based on the seismic data to obtain an initial impedance model of a reservoir to be predicted;

a seismic record generation unit 220 for: synthesizing the initial impedance model to obtain a seismic record corresponding to the initial impedance model;

a prediction result determination unit 230 for: and judging whether the reservoir prediction result of the reservoir to be predicted can be output according to the initial impedance model or not according to the seismic record corresponding to the initial impedance model, and outputting the reservoir prediction result according to the initial impedance model under the condition that the judgment result is yes.

FIG. 14 is a schematic diagram of the structure of a phased modeling unit in an embodiment of the invention. As shown in fig. 14, the phased modeling unit 220 may include: a seismic attribute generation module 221, a seismic lithofacies generation module 222, and an impedance model generation module 223, which are connected in sequence.

A seismic attribute generation module 221 for: performing attribute analysis by using the seismic data to obtain a plurality of seismic attributes;

a seismic lithofacies generation module 222 to: based on the well logging lithofacies, performing lithofacies prediction by using the plurality of seismic attributes to obtain seismic lithofacies of the reservoir to be predicted;

an impedance model generation module 223 for: and establishing an initial impedance model of the reservoir to be predicted under the phase control constraint of the seismic lithofacies.

FIG. 15 is a schematic diagram of the structure of a seismic lithofacies generation module in an embodiment of the present invention. As shown in fig. 15, the seismic lithofacies generation module 222 may include: a logging lithofacies generation module 2221, a lithofacies and attribute relationship establishment module 2222, and a seismic lithofacies output module 2223, which are connected in sequence.

A well log lithofacies generation module 2221 to: based on geological background, carrying out logging lithofacies analysis by using logging data and interpretation data thereof to obtain logging lithofacies;

a lithofacies and attribute relationship establishing module 2222 for: analyzing by utilizing the plurality of seismic attributes based on the logging lithofacies, and establishing a relation between the logging lithofacies and the seismic attributes;

a seismic lithofacies output module 2223 to: and outputting the seismic facies of the reservoir to be predicted according to the relation between the logging facies and the seismic attributes.

Fig. 16 is a schematic structural diagram of a reservoir prediction apparatus according to another embodiment of the present invention. As shown in fig. 16, the reservoir prediction apparatus shown in fig. 13 may further include: and the high-resolution processing unit 240 is connected with the phase control modeling unit 210.

A high resolution processing unit 240 for: and carrying out high-resolution processing on the initial seismic data based on the reservoir characteristics to obtain the seismic data.

Fig. 17 is a schematic structural diagram of a high resolution processing unit according to an embodiment of the invention. As shown in fig. 17, the high resolution processing unit 240 may include: a thickness and dominant frequency and bandwidth relation establishing module 241 and a frequency extending parameter adjusting module 242, which are connected with each other.

A thickness and dominant frequency and bandwidth relation establishing module 241, configured to: establishing a relation between reservoir thickness and dominant frequency and bandwidth by performing time-frequency analysis on initial seismic data of a target layer in a reservoir to be predicted and performing variable reservoir thickness model forward modeling on the target layer;

a frequency extending parameter adjusting module 242, configured to: and adjusting frequency broadening parameters based on the relation between the reservoir thickness and the dominant frequency and the frequency width so as to suppress interference noise in the initial seismic data and improve the frequency width of effective signals in the initial seismic data.

Fig. 18 is a schematic structural diagram of a prediction result determining unit according to an embodiment of the present invention. As shown in fig. 18, the prediction result determining unit 230 may include: an initial attribute generation module 231, a weighted prediction attribute generation module 232, an attribute error generation module 233, and a prediction result determination module 234, which are connected in sequence.

An initial attribute generation module 231, configured to: performing seismic attribute analysis by using the seismic records corresponding to the initial impedance model to obtain a plurality of initial seismic attributes;

a weighted prediction attribute generation module 232 configured to: weighting and summing a plurality of effective seismic attributes in the plurality of initial seismic attributes according to corresponding weight coefficients to obtain weighted predicted seismic attributes;

an attribute error generation module 233 for: calculating a seismic attribute error according to the weighted predicted seismic attribute and the weighted actual seismic attribute; the weighted actual seismic attributes are obtained by weighting and summing the effective seismic attributes in the actual seismic attributes according to the corresponding weight coefficients, and the actual seismic attributes are obtained by performing attribute analysis according to the seismic data;

a prediction result determination module 234 for: and judging whether the reservoir prediction result of the reservoir to be predicted can be output according to the initial impedance model or not according to whether the seismic attribute error is smaller than a set error value or not.

Fig. 19 is a schematic structural diagram of a reservoir prediction apparatus according to still another embodiment of the present invention. As shown in fig. 19, the reservoir prediction apparatus shown in fig. 13 may further include: an impedance model modification unit 240, a seismic record regeneration unit 220 'and a prediction result re-determination unit 230', which are connected in sequence.

An impedance model modification unit 240 for: modifying the initial impedance model to obtain a modified impedance model under the condition that the reservoir prediction result of the reservoir to be predicted cannot be output according to the initial impedance model;

a seismic record regeneration unit 220' for: synthesizing the modified impedance model to obtain a seismic record corresponding to the modified impedance model;

a prediction result re-determination unit 230' for: and judging whether the reservoir prediction result of the reservoir to be predicted can be output according to the modified impedance model or not according to the seismic record corresponding to the modified impedance model, and if so, outputting the reservoir prediction result according to the modified impedance model.

Fig. 20 is a schematic structural diagram of a reservoir prediction apparatus according to still another embodiment of the present invention. As shown in fig. 20, the reservoir prediction apparatus shown in fig. 13 may further include: the effective attribute and weight determination unit 250 is connected between the seismic record generation unit 220 and the prediction result determination unit 230.

A valid attribute and weight determination unit 250 configured to: determining the plurality of valid seismic attributes and the corresponding weight coefficients based on well log data and petrophysical theory.

The valid attribute and weight determining unit 250 may include: a geological model generation module 251, a geophysical model generation module 252, a reservoir parameter model building module 253, and an effective attribute and weight generation module 254, which are connected in sequence.

A geological model generation module 251 for: generating a geological model of the reservoir to be predicted according to the logging data;

a geophysical model generation module 252 to: establishing a geophysical model of the reservoir to be predicted according to the geological model;

a reservoir parametric model building module 253 for: establishing a reservoir parameter model of the reservoir to be predicted based on a rock physics theory and the geophysical model;

a valid attribute and weight generation module 254 to: and screening the plurality of effective seismic attributes sensitive to the reservoir according to the reservoir parameter model, and setting the weight coefficient for each effective seismic attribute according to the degree of sensitivity to the reservoir.

The embodiment of the present invention further provides a computer device, which includes a memory, a processor, and a computer program stored in the memory and capable of running on the processor, and when the processor executes the computer program, the steps of the method described in the above embodiments are implemented.

In summary, the reservoir prediction method, the reservoir prediction apparatus, the computer-readable storage medium, and the computer device according to the embodiments of the present invention perform phase-controlled modeling based on the seismic data to obtain the initial impedance model of the reservoir to be predicted, instead of performing simple extrapolation and interpolation modeling under the seismic horizon constraint by using the logging data. In the phase-controlled modeling process, the internal relation between the geological deposition background and the seismic reflection can be fully considered, so that the problem that the logging data is excessively depended when the logging data is used for carrying out extrapolation and interpolation modeling under the constraint of the seismic horizon can be solved, the phase-controlled modeling has higher modeling precision, and the high-precision impedance model can improve the resolution ratio of reservoir prediction. Furthermore, the initial seismic data are subjected to high-resolution processing based on reservoir characteristics to obtain the seismic data, so that the resolution of the seismic data can be improved. And further, whether the reservoir prediction result of the reservoir to be predicted can be output according to the initial impedance model or not is judged based on the weighted sum of the multiple attributes, so that the multi-solution of the inversion result in the prediction process can be effectively reduced.

In the description herein, reference to the description of the terms "one embodiment," "a particular embodiment," "some embodiments," "for example," "an example," "a particular example," or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. The sequence of steps involved in the various embodiments is provided to schematically illustrate the practice of the invention, and the sequence of steps is not limited and can be suitably adjusted as desired.

As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.

The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

The above-mentioned embodiments are intended to illustrate the objects, technical solutions and advantages of the present invention in further detail, and it should be understood that the above-mentioned embodiments are only exemplary embodiments of the present invention, and are not intended to limit the scope of the present invention, and any modifications, equivalent substitutions, improvements and the like made within the spirit and principle of the present invention should be included in the scope of the present invention.

Claims

1. A method of reservoir prediction, comprising:

performing phase control modeling based on the seismic data to obtain an initial impedance model of a reservoir to be predicted;

synthesizing the initial impedance model to obtain a seismic record corresponding to the initial impedance model;

judging whether a reservoir prediction result of the reservoir to be predicted can be output according to the initial impedance model or not according to the seismic records corresponding to the initial impedance model, and outputting the reservoir prediction result according to the initial impedance model under the condition that the judgment result is yes;

the step of judging whether the reservoir prediction result of the reservoir to be predicted can be output according to the initial impedance model according to the seismic records corresponding to the initial impedance model comprises the following steps:

performing seismic attribute analysis by using the seismic records corresponding to the initial impedance model to obtain a plurality of initial seismic attributes;

weighting and summing a plurality of effective seismic attributes in the plurality of initial seismic attributes according to corresponding weight coefficients to obtain weighted predicted seismic attributes;

calculating a seismic attribute error according to the weighted predicted seismic attribute and the weighted actual seismic attribute; the weighted actual seismic attributes are obtained by weighting and summing the effective seismic attributes in the actual seismic attributes according to the corresponding weight coefficients, and the actual seismic attributes are obtained by performing attribute analysis according to the seismic data;

and judging whether the reservoir prediction result of the reservoir to be predicted can be output according to the initial impedance model or not according to whether the seismic attribute error is smaller than a set error value or not.

2. The reservoir prediction method of claim 1, wherein performing phase-controlled modeling based on the seismic data to obtain an initial impedance model of the reservoir to be predicted comprises:

performing attribute analysis by using the seismic data to obtain a plurality of seismic attributes;

based on the well logging lithofacies, performing lithofacies prediction by using the plurality of seismic attributes to obtain seismic lithofacies of the reservoir to be predicted;

and establishing an initial impedance model of the reservoir to be predicted under the phase control constraint of the seismic lithofacies.

3. The reservoir prediction method of claim 2, wherein performing facies prediction using the plurality of seismic attributes based on well-log facies to obtain seismic facies of the reservoir to be predicted, comprises:

based on geological background, carrying out logging lithofacies analysis by using logging data and interpretation data thereof to obtain logging lithofacies;

analyzing by utilizing the plurality of seismic attributes based on the logging lithofacies, and establishing a relation between the logging lithofacies and the seismic attributes;

and outputting the seismic facies of the reservoir to be predicted according to the relation between the logging facies and the seismic attributes.

4. The method of reservoir prediction of claim 1, wherein prior to performing phase-controlled modeling based on seismic data to obtain an initial impedance model of the reservoir to be predicted, further comprising:

and carrying out high-resolution processing on the initial seismic data based on the reservoir characteristics to obtain the seismic data.

5. The method of reservoir prediction of claim 4, wherein the step of performing high resolution processing on the initial seismic data based on reservoir characteristics to obtain the seismic data comprises:

establishing a relation between reservoir thickness and dominant frequency and bandwidth by performing time-frequency analysis on initial seismic data of a target layer in a reservoir to be predicted and performing variable reservoir thickness model forward modeling on the target layer;

and adjusting frequency broadening parameters based on the relation between the reservoir thickness and the dominant frequency and the frequency width so as to suppress interference noise in the initial seismic data and improve the frequency width of effective signals in the initial seismic data.

6. The reservoir prediction method of claim 1, further comprising:

modifying the initial impedance model to obtain a modified impedance model under the condition that the reservoir prediction result of the reservoir to be predicted cannot be output according to the initial impedance model;

synthesizing the modified impedance model to obtain a seismic record corresponding to the modified impedance model;

and judging whether the reservoir prediction result of the reservoir to be predicted can be output according to the modified impedance model or not according to the seismic record corresponding to the modified impedance model, and if so, outputting the reservoir prediction result according to the modified impedance model.

7. The reservoir prediction method of claim 1, further comprising:

determining the plurality of effective seismic attributes and the corresponding weight coefficients based on well log data and petrophysical theory;

wherein obtaining the plurality of effective seismic attributes and the corresponding weight coefficients based on log data and petrophysical theory comprises:

generating a geological model of the reservoir to be predicted according to the logging data;

establishing a geophysical model of the reservoir to be predicted according to the geological model;

establishing a reservoir parameter model of the reservoir to be predicted based on a rock physics theory and the geophysical model;

and screening the plurality of effective seismic attributes sensitive to the reservoir according to the reservoir parameter model, and setting the weight coefficient for each effective seismic attribute according to the degree of sensitivity to the reservoir.

8. A reservoir prediction apparatus, comprising:

a phase control modeling unit to: performing phase control modeling based on the seismic data to obtain an initial impedance model of a reservoir to be predicted;

a seismic record generation unit to: synthesizing the initial impedance model to obtain a seismic record corresponding to the initial impedance model;

a prediction result determination unit configured to: judging whether a reservoir prediction result of the reservoir to be predicted can be output according to the initial impedance model or not according to the seismic records corresponding to the initial impedance model, and outputting the reservoir prediction result according to the initial impedance model under the condition that the judgment result is yes;

the prediction result determination unit includes:

an initial attribute generation module to: performing seismic attribute analysis by using the seismic records corresponding to the initial impedance model to obtain a plurality of initial seismic attributes;

a weighted prediction attribute generation module to: weighting and summing a plurality of effective seismic attributes in the plurality of initial seismic attributes according to corresponding weight coefficients to obtain weighted predicted seismic attributes;

an attribute error generation module to: calculating a seismic attribute error according to the weighted predicted seismic attribute and the weighted actual seismic attribute; the weighted actual seismic attributes are obtained by weighting and summing the effective seismic attributes in the actual seismic attributes according to the corresponding weight coefficients, and the actual seismic attributes are obtained by performing attribute analysis according to the seismic data;

a prediction result determination module to: and judging whether the reservoir prediction result of the reservoir to be predicted can be output according to the initial impedance model or not according to whether the seismic attribute error is smaller than a set error value or not.

9. The reservoir prediction apparatus of claim 8, wherein the phased modeling unit comprises:

a seismic attribute generation module to: performing attribute analysis by using the seismic data to obtain a plurality of seismic attributes;

a seismic lithofacies generation module to: based on the well logging lithofacies, performing lithofacies prediction by using the plurality of seismic attributes to obtain seismic lithofacies of the reservoir to be predicted;

an impedance model generation module to: and establishing an initial impedance model of the reservoir to be predicted under the phase control constraint of the seismic lithofacies.

10. The reservoir prediction apparatus of claim 9, wherein the seismic facies generation module comprises:

a log facies generation module to: based on geological background, carrying out logging lithofacies analysis by using logging data and interpretation data thereof to obtain logging lithofacies;

a lithofacies and attribute relationship establishing module for: analyzing by utilizing the plurality of seismic attributes based on the logging lithofacies, and establishing a relation between the logging lithofacies and the seismic attributes;

a seismic lithofacies output module to: and outputting the seismic facies of the reservoir to be predicted according to the relation between the logging facies and the seismic attributes.

11. The reservoir prediction apparatus of claim 8, further comprising:

a high resolution processing unit to: and carrying out high-resolution processing on the initial seismic data based on the reservoir characteristics to obtain the seismic data.

12. The reservoir prediction apparatus of claim 11, wherein the high resolution processing unit comprises:

the thickness, main frequency and bandwidth relation establishing module is used for: establishing a relation between reservoir thickness and dominant frequency and bandwidth by performing time-frequency analysis on initial seismic data of a target layer in a reservoir to be predicted and performing variable reservoir thickness model forward modeling on the target layer;

a frequency extension parameter adjusting module for: and adjusting frequency broadening parameters based on the relation between the reservoir thickness and the dominant frequency and the frequency width so as to suppress interference noise in the initial seismic data and improve the frequency width of effective signals in the initial seismic data.

13. The reservoir prediction apparatus of claim 8, further comprising:

an impedance model modification unit for: modifying the initial impedance model to obtain a modified impedance model under the condition that the reservoir prediction result of the reservoir to be predicted cannot be output according to the initial impedance model;

a seismic record regeneration unit to: synthesizing the modified impedance model to obtain a seismic record corresponding to the modified impedance model;

a prediction result re-determination unit configured to: and judging whether the reservoir prediction result of the reservoir to be predicted can be output according to the modified impedance model or not according to the seismic record corresponding to the modified impedance model, and if so, outputting the reservoir prediction result according to the modified impedance model.

14. The reservoir prediction apparatus of claim 8, further comprising:

an effective attribute and weight determination unit for: determining the plurality of effective seismic attributes and the corresponding weight coefficients based on well log data and petrophysical theory;

wherein the effective attribute and weight determining unit includes:

a geological model generation module to: generating a geological model of the reservoir to be predicted according to the logging data;

a geophysical model generation module to: establishing a geophysical model of the reservoir to be predicted according to the geological model;

a reservoir parametric model building module to: establishing a reservoir parameter model of the reservoir to be predicted based on a rock physics theory and the geophysical model;

an effective attribute and weight generation module to: and screening the plurality of effective seismic attributes sensitive to the reservoir according to the reservoir parameter model, and setting the weight coefficient for each effective seismic attribute according to the degree of sensitivity to the reservoir.

15. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method of claims 1 to 7.

16. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the steps of the method of claims 1 to 7 are implemented when the processor executes the program.