CN114415241A - Tight sandstone reservoir fluid identification method based on logging curve generalized S transformation - Google Patents

Tight sandstone reservoir fluid identification method based on logging curve generalized S transformation Download PDF

Info

Publication number
CN114415241A
CN114415241A CN202210062143.XA CN202210062143A CN114415241A CN 114415241 A CN114415241 A CN 114415241A CN 202210062143 A CN202210062143 A CN 202210062143A CN 114415241 A CN114415241 A CN 114415241A
Authority
CN
China
Prior art keywords
generalized
time
frequency
signal
transformation
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202210062143.XA
Other languages
Chinese (zh)
Inventor
赖富强
蒋国强
黄兆辉
臧永钤
魏艳写
刘源琦
陈阵
寇小攀
王海涛
张晓树
夏小雪
刘粤蛟
欧发辉
谭先锋
朱章雄
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Chongqing University of Science and Technology
Original Assignee
Chongqing University of Science and Technology
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Chongqing University of Science and Technology filed Critical Chongqing University of Science and Technology
Priority to CN202210062143.XA priority Critical patent/CN114415241A/en
Publication of CN114415241A publication Critical patent/CN114415241A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
    • G01V1/00—Seismology; Seismic or acoustic prospecting or detecting
    • G01V1/40—Seismology; Seismic or acoustic prospecting or detecting specially adapted for well-logging
    • G01V1/44—Seismology; Seismic or acoustic prospecting or detecting specially adapted for well-logging using generators and receivers in the same well
    • G01V1/48—Processing data
    • G01V1/50—Analysing data
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
    • G01V1/00—Seismology; Seismic or acoustic prospecting or detecting
    • G01V1/40—Seismology; Seismic or acoustic prospecting or detecting specially adapted for well-logging
    • G01V1/44—Seismology; Seismic or acoustic prospecting or detecting specially adapted for well-logging using generators and receivers in the same well
    • G01V1/48—Processing data
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
    • G01V2210/00—Details of seismic processing or analysis
    • G01V2210/60—Analysis
    • G01V2210/62—Physical property of subsurface
    • G01V2210/624—Reservoir parameters
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
    • G01V2210/00—Details of seismic processing or analysis
    • G01V2210/70—Other details related to processing
    • G01V2210/74—Visualisation of seismic data

Landscapes

  • Physics & Mathematics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Acoustics & Sound (AREA)
  • Environmental & Geological Engineering (AREA)
  • Geology (AREA)
  • Remote Sensing (AREA)
  • General Life Sciences & Earth Sciences (AREA)
  • General Physics & Mathematics (AREA)
  • Geophysics (AREA)
  • Geophysics And Detection Of Objects (AREA)

Abstract

The method introduces a time-frequency analysis method of generalized S transformation into the identification of logging fluid, utilizes conventional logging data, reduces noise of the data, then carries out generalized S transformation on high-frequency signal parts, outputs a time-frequency matrix, draws a two-dimensional contour map on the time-frequency matrix, and further analyzes the properties of reservoir fluid through the characteristics of the shape of an energy cluster and the like. Compared with the conventional well logging fluid identification method, the method has the advantages that firstly, the noise in the conventional well logging curve is reduced, the distortion problem of time-frequency domain conversion is solved, secondly, fluids with different properties can be distinguished accurately in the form of energy clusters, and the identification accuracy of the high-resistance water layer is improved.

Description

Tight sandstone reservoir fluid identification method based on logging curve generalized S transformation
Technical Field
The invention belongs to the field of geological exploration, and particularly relates to a tight sandstone reservoir fluid identification method based on logging curve generalized S transformation
Background
The dense sandstone fluid identification has the advantages that the dense sandstone reservoir has strong heterogeneity, the gas-water distribution is uneven, and the high-resistance water layer and the low-resistance gas layer have obvious abnormal phenomena, so that great challenges are brought to the reservoir fluid identification, and the identification is particularly characterized in that the coincidence rate of well logging interpretation and gas testing results is low, and the production and research requirements cannot be met.
At present, common fluid identification technologies based on core analysis, well logging and gas testing data comprise an overlapping method, a cross-plot method, a mathematical statistics method and the like, wherein the overlapping method relies on a porosity curve to construct a fluid indicator factor and then cross the porosity curve to identify fluid, the method relies on a single porosity curve to identify fluid, and the identification accuracy and efficiency are required to be further improved; the cross plot method is to establish a linear equation between resistivity and porosity by using an Arzier formula, and distinguish oil, gas and water according to the change of water saturation; the mathematical statistics method is to scale some attributes of the logging curve by means of normal distribution and the like and distinguish the attributes according to the normal curve forms of different fluids, but the method needs to establish different plates according to different regions, is complex to operate, strongly depends on manual experience, and is poor in popularization.
In addition, reservoir fluid is identified through special logging data such as an array acoustic logging method and a nuclear magnetic resonance logging method, the special logging data are not easy to obtain, cost is greatly increased, and popularization and application are not facilitated.
Disclosure of Invention
The technical problem to be solved by the invention is as follows:
(1) the prior art signal transitions in the time-frequency domain may present distortion phenomena.
(2) Noise interference present in prior art logging signals affects the final fluid identification effect display.
(3) The prior art fluid identification routine curves are less different and strongly dependent on manual experience.
The invention provides a tight sandstone reservoir fluid identification method based on logging curve generalized S transformation, which comprises the following steps:
s1 loading a logging signal;
s2 decomposing the logging signal;
s3 performing a generalized S transform on the logging signal;
s4 generating a time-frequency matrix graph;
s5 plots a two-dimensional contour map.
Further, the step S2 includes,
s21: drawing an upper envelope line and a lower envelope line respectively according to upper and lower extreme points of an original signal, wherein the upper envelope line is defined as max (t), the lower envelope line is defined as min (t), and t represents a time signal;
s22: calculating the mean value m (t) of the upper envelope line and the lower envelope line, and drawing a mean value envelope line, wherein the mean value envelope line is obtained by the following formula:
m(t)＝(max(t)+min(t))/2；
s23: subtracting the envelope curve of the mean value of the original signal to obtain an intermediate signal h (t), wherein h (t) is obtained by the following formula:
h(t)＝x(t)-m(t)
x (t) represents the original signal;
s24: judging whether the intermediate signal meets two conditions of IMF components, wherein the IMF components represent each frequency component in the original signal;
s25 the original signal is decomposed to form a linear superposition of a series of eigenmode components and residuals, formulated as follows:
Figure BDA0003478802670000021
in the formula: imfi(t) denotes the i-th intrinsic modal component, r (t) denotes the residual component, and l denotes the total number of modal components.
Further, the step S3 includes,
the S transform is defined as:
Figure BDA0003478802670000031
in the formula: tau is time, and the position of the window function on the time axis is controlled; h (t) is an analytical signal; f is the frequency; s (tau, f) is a time-frequency spectrum matrix obtained by transformation, and i represents the serial number of the intrinsic mode;
the time duration and the decay speed of the window function are adjusted by introducing two factors of lambda and p into the originally constructed Gaussian window function, and the window function and the generalized S transformation are expressed as follows:
Figure BDA0003478802670000032
Figure BDA0003478802670000033
wherein h (t) is the original signal, λ and p are two adjustment factors, and the time-frequency resolution of the window function is controlled; tau is time, the position of the window function on a time axis is controlled, f is frequency, and GST (tau, f) is an obtained time-frequency spectrum matrix;
and mapping the one-dimensional well logging curve into a two-dimensional image through the generalized S transformation.
Further, the step S4 includes,
and in the time-frequency matrix diagram, a logging curve data information diagram and a generalized S transformation effect diagram are displayed in a contrast mode.
Further, two conditions of the IMF component include,
(1) the number of IMF extrema (sum of the number of maxima and minima) and the number of zero crossings must be equal or differ by at most 1;
(2) at any point in the IMF, the average of the envelope defined by the local maxima and the average of the envelope defined by the local minima should be equal to zero. If so, the signal is an IMF component; if not, the analysis of the first to fourth steps is repeated based on the signal.
The invention has the beneficial effects that:
(1) aiming at the phenomenon that distortion may exist in the time-frequency domain conversion of signals, the generalized S transformation is adopted to process the signals, and the generalized S transformation is developed on the basis of short-time Fourier transformation and wavelet transformation and is reversible lossless transformation, so that the distortion phenomenon cannot occur.
(2) Aiming at the problem that the noise interference in the logging signal can influence the display of the final fluid identification effect, the invention adopts a first algorithm to decompose the original logging signal, and because the first eigenmode component retains more complete high-frequency information and can be used for reflecting the integral logging characteristics and filtering some low-frequency noise, the first eigenmode component is selected for further processing.
(3) Aiming at the problems that the conventional curve difference of fluid identification is small and manual experience is strongly depended on, the generalized S transformation with adjustable parameters is adopted, the shape of a window function is changed by changing the size of parameters lambda and p, namely the time-frequency resolution of the window function is changed, the curve characteristic difference of the finally obtained logging signal is amplified, and the curve characteristic difference is displayed visually in the form of energy clusters, so that different fluids are distinguished.
Drawings
Fig. 1 is a schematic diagram of the upper and lower envelope of a signal.
Fig. 2 is a diagram illustrating the mean envelope of a signal.
Fig. 3 is a schematic diagram of intermediate signals.
Fig. 4 is a schematic diagram of the iterative processing of the first algorithm.
Fig. 5 is a diagram illustrating the effect of the first algorithm after signal decomposition.
FIG. 6 is a preferred diagram of p-parameter.
Fig. 7 is a preferred diagram of the lambda parameter.
Fig. 8 is a diagram illustrating the p parameter corresponding to the RLLD curve generalized S transform effect.
Fig. 9 is a diagram illustrating the effect of the generalized S transform of the RLLD curve corresponding to the λ parameter.
FIG. 10 is a cross-sectional view of shallow and deep resistivity in a certain area.
FIG. 11 is a schematic diagram of the effect (gas layer) of the three time-frequency analysis methods.
FIG. 12 is a schematic diagram of the effect (water layer) of the three time-frequency analysis methods.
FIG. 13 is a schematic diagram of the effect (gas difference layer) of the three time-frequency analysis methods.
FIG. 14 is a technical roadmap.
Detailed Description
The invention has the conception that the time-frequency analysis method of generalized S transformation is introduced into fluid identification, conventional logging data is utilized, after data noise reduction, high-frequency signal parts in the logging data are subjected to generalized S transformation, a time-frequency matrix is output and drawn into a two-dimensional contour map, and reservoir fluid properties are further analyzed through the characteristics of energy cluster morphology and the like.
The technical solution of the present invention is explained below.
(1) Extracting high frequency information using a first algorithm decomposition
The first algorithm can adaptively decompose the signal into the sum of a plurality of intrinsic mode functions according to the characteristics of the input signal without knowing any prior knowledge. After the original logging signal is decomposed by the first algorithm, the first eigenmode component not only retains relatively complete high-frequency information, but also filters low-frequency noise, and the display effect after the next signal transformation is optimized.
The first algorithm is specifically implemented by the following steps:
the first step is as follows: according to the upper and lower extreme points of the original signal, an upper envelope curve and a lower envelope curve are drawn respectively, wherein the upper envelope curve is defined as max (t), and the lower envelope curve is defined as min (t).
The second step is that: calculating the mean value m (t) of the upper envelope line and the lower envelope line, and drawing a mean value envelope line, wherein the mean value envelope line is obtained by the following formula:
m(t)＝(max(t)+min(t))/2 (1)
the third step: subtracting the envelope curve of the mean value of the original signal to obtain an intermediate signal h (t), wherein h (t) is obtained by the following formula:
h(t)＝x(t)-m(t) (2)
the fourth step: judging whether the intermediate signal meets two conditions of IMF, if so, the signal is an IMF component; if not, the analysis of the first to fourth steps is repeated based on the signal. The acquisition of the IMF components typically requires several iterations.
After the first IMF is obtained by the method, the IMF1 is subtracted from the original signal to be used as a new original signal, and the IMF2 can be obtained by the four steps of analysis, and the rest is done to complete the first algorithm decomposition. The original signal after being decomposed by a first algorithm forms a series of linear superpositions of eigenmode components and residuals, which are formulated as follows:
the original signal after being decomposed by a first algorithm forms a series of linear superpositions of eigenmode components and residuals, which are formulated as follows:
Figure BDA0003478802670000061
in the formula: imfi(t) denotes the ith eigenmode component, and r (t) denotes the residual component.
The first eigenmode component has complete high-frequency information, and certain low-frequency noise which may cause interference is filtered, so the first eigenmode component after signal decomposition is selected for subsequent operation.
(2) Generalized S-transform time-frequency diagram generation
The S transform is defined as:
Figure BDA0003478802670000062
in the formula: tau is time, and the position of the window function on the time axis is controlled; h (t) is an analytical signal; f is the frequency; and S (tau, f) is a time-frequency spectrum matrix obtained by transformation.
However, the window function of the S-transform is still fixed, so that the time-frequency focusing is limited, and a generalized S-transform is proposed to overcome this disadvantage. The generalized S transform combines the advantages of short-time Fourier transform and wavelet transform, overcomes the problem that the time-frequency window of the short-time Fourier transform is not changeable in a time-frequency plane, and does not need to meet the tolerance condition of the wavelet. The window function of the device has the self-adaptive variable characteristic depending on the frequency, and the time-frequency window characteristic can present various different variation trends along with the frequency through different parameter values, so that the device is more suitable for various complex signal analysis in practical application and is a good tool for non-stationary signal time-frequency analysis.
The generalized S-transform adjusts the window function time duration and decay rate by introducing two factors of λ and p into the originally constructed Gaussian window function, whose window function and generalized S-transform (GST) can be expressed as:
Figure BDA0003478802670000071
Figure BDA0003478802670000072
in the formula: h (t) is the original signal; λ and p are two adjustment factors, controlling the time-frequency resolution of the window function; tau is time, and the position of the window function on the time axis is controlled; f is the frequency; GST (τ, f) is the resulting time-frequency spectrum matrix.
According to the invention, one-dimensional logging curves are mapped into a two-dimensional image through generalized S transformation, so that the characteristics of curves corresponding to different fluid properties of different reservoirs are amplified and displayed in an intuitive energy cluster form.
The following is a specific implementation step of the generalized S transformation in the present invention:
optimization of generalized S transformation parameters
Since the Gaussian window of the S-transform varies inversely with frequency, lambda and p (lambda > 0, p > 0) are used as the embodiment of the flexibility of the generalized S-transform, and the rate of the change of the size of the window function along with the frequency, namely the time-frequency resolution of the window function, can be changed by adjusting the lambda and the p, so that the difference between different fluids of the time-frequency diagram can be amplified.
After λ is selected, the value of p is changed, and it can be seen that the rate of change of the width of the window function with frequency is first increased and then decreased, that is, the value of p is not as large as possible, and finally the value of p is selected to be 0.6.
Similarly, when p is selected, by changing the value of λ, it can be seen that the rate of change of the width of the window function with frequency is first increased and then decreased, that is, the value of λ is not as large as possible, and finally the value of λ is selected to be 1.4.
After multiple experiments, the optimal parameter value λ 1.4 and p 0.6 are selected comprehensively for subsequent experiments, but different λ and p values should be selected through experiments according to different data, so as to obtain the optimal time-frequency resolution of generalized S transform.
Analysis of generalized S transformation effect
The existing logging data are processed by respectively adopting the lambda parameters and the p parameters, and the following example is shown by comparing the generalized S transformation effect of the RLLD curve. It can be verified that under the window function when λ is 1.4 and p is 0.6, the energy clique time domain and the frequency domain of the signal are both optimal, so the generalized S-transform time-frequency resolution of the curve is both optimal. It can also be verified in reverse that the λ, p parameters are preferably correct.
In addition, the best effect of the generalized S transformation is taken as a reference, effect comparison analysis is carried out by combining short-time Fourier transformation and wavelet transformation, the same signal of the same well is selected for analysis, and the identification effect of the generalized S transformation can be verified under the condition that a gas testing result is known.
After selecting proper generalized S transformation window function parameters, recognizing the fluid by using a traditional intersection map method, and observing whether the effect can reach the expectation or not.
The depth resistivity curve is used as a characteristic curve which is sensitive to fluid properties in a logging curve, and the gas content of the stratum can be identified more sensitively, so that depth resistivity data is adopted for intersection, data points of different fluid properties are selected in the graph 10 for drawing, the data points of different fluid types can be found to be in linear distribution, but the distinction is not obvious, wherein the distribution range of a gas layer and a water layer is wide, and different fluid properties cannot be directly distinguished through the depth resistivity intersection graph. It is necessary to amplify the differences in fluid properties between different reservoirs by means of a generalized S-transform.
The following is the effect of processing data using the generalized S transform:
when the property of the selected section fluid is the gas layer, energy clusters in (middle) generalized S transform of the signal appear in 2hz and 8hz, and Fourier transform (left) is better when the time-frequency focusing property is shorter, and energy clusters with different frequency resolutions can be distinguished, which proves to be higher than the time-frequency resolution of wavelet transform (right). And when the property of the reservoir fluid is a gas layer, the energy mass is generally in an ellipsoid shape, the smoothness is better, and the enveloping ratio is centered.
When the property of the selected section fluid is a water layer, the energy cluster frequency of the generalized S transform of the signal is below 4hz, but the time focusing property and the frequency focusing property of the generalized S transform (middle) are better than those of the short-time Fourier transform (left) and the wavelet transform (right). And when the property of the reservoir fluid is a water layer, the energy group is more in a semicircular shape.
When the property of the selected section of fluid is a gas difference layer, the time-frequency resolution of the energy group of the generalized S transformation of the signal is similar to that of the other two methods. And when the property of the reservoir fluid is a poor gas layer, the energy cluster is in an oblong shape and is distributed in a strip shape.
Example 1
Fig. 14 is a flowchart of a specific implementation of the present invention, and the matlab2020a platform implements the above contents of the present invention, and develops corresponding modules, which are described in further detail below:
1. loading required conventional logging information;
2. processing the logging data curve, and using the obtained first intrinsic mode component for next processing;
3. processing the first eigenmode component by using a formula (6), and outputting a corresponding time-frequency matrix;
4. and (4) drawing and displaying the time-frequency matrix in the step (3) to obtain a result graph similar to the result graph in the figure 11 b. The left side of the diagram shows the data information of the logging curve, and the right side is a generalized S transformation effect diagram.
As can be seen from fig. 12, in the reservoir section, the reservoir section curve is identified from the well logging curve alone, the difference between the reservoir section curve and the surrounding rock curve is not obvious, and after the generalized S transformation, the property of the reservoir section can be visually identified through the energy clusters distributed in a strip shape, and the reservoir section is identified as a gas difference layer. It is therefore considered that the present invention achieves the effect of amplifying the difference in fluid properties.
Then, the other two time-frequency analysis methods mentioned above are used for comparative analysis, as shown in fig. 10, 11 and 12, the energy clusters obtained by the generalized S transform have excellent effects in the three time-frequency analysis methods, no matter from the time-frequency focusing or time-frequency resolution. And in reservoir fluid property identification, fluid properties can be preliminarily judged according to different energy cluster forms.
The invention has the beneficial effects that:
(1) aiming at the problem that the interference noise of the original logging signal possibly influences the recognition effect, the high-frequency information of the original logging signal is reserved, and meanwhile, the low-frequency noise is filtered, so that a foundation is laid for the next generalized S transformation processing.
(2) Through comparison and verification of a cross plot identification method and three time-frequency analysis methods, after the signals are subjected to generalized S transformation, the time domain and frequency domain resolutions of signal energy clusters are optimal among different fluid properties, and the energy cluster forms of different fluids are different, so that the fluids can be identified from another dimension, and the identification difficulty and the experience requirement are reduced.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims (5)

1. The tight sandstone reservoir fluid identification method based on the logging curve generalized S transform is characterized by comprising the following steps of:
s1 loading a logging signal;
s2 decomposing the logging signal;
s3 performing a generalized S transform on the logging signal;
s4 generating a time-frequency matrix graph;
s5 plots a two-dimensional contour map.
2. The tight sandstone reservoir fluid identification method based on the log generalized S-transform of claim 1, wherein the step S2 comprises,
s21: drawing an upper envelope line and a lower envelope line respectively according to upper and lower extreme points of an original signal, wherein the upper envelope line is defined as max (t), the lower envelope line is defined as min (t), and t represents a time signal;
s22: calculating the mean value m (t) of the upper envelope line and the lower envelope line, and drawing a mean value envelope line, wherein the mean value envelope line is obtained by the following formula:
m(t)＝(max(t)+min(t))/2；
s23: subtracting the envelope curve of the mean value of the original signal to obtain an intermediate signal h (t), wherein h (t) is obtained by the following formula:
h(t)＝x(t)-m(t)
x (t) represents the original signal;
s24: judging whether the intermediate signal meets two conditions of IMF components, wherein the IMF components represent each frequency component in the original signal;
s25 the original signal is decomposed to form a linear superposition of a series of eigenmode components and residuals, formulated as follows:
Figure FDA0003478802660000011
in the formula: imfi(t) denotes the i-th intrinsic modal component, r (t) denotes the residual component, and l denotes the total number of modal components.
3. The tight sandstone reservoir fluid identification method based on the log generalized S-transform of claim 1, wherein the step S3 comprises,
the S transform is defined as:
Figure FDA0003478802660000021
in the formula: tau is time, and the position of the window function on the time axis is controlled; h (t) is an analytical signal; f is the frequency; s (tau, f) is a time-frequency spectrum matrix obtained by transformation;
the time duration and the decay speed of the window function are adjusted by introducing two factors of lambda and p into the originally constructed Gaussian window function, and the window function and the generalized S transformation are expressed as follows:
Figure FDA0003478802660000022
Figure FDA0003478802660000023
wherein h (t) is the original signal, λ and p are two adjustment factors, and the time-frequency resolution of the window function is controlled; tau is time, the position of the window function on a time axis is controlled, f is frequency, and GST (tau, f) is an obtained time-frequency spectrum matrix;
and mapping the one-dimensional well logging curve into a two-dimensional image through the generalized S transformation.
4. The tight sandstone reservoir fluid identification method based on the log generalized S-transform of claim 1, wherein the step S4 comprises,
and in the time-frequency matrix diagram, a logging curve data information diagram and a generalized S transformation effect diagram are displayed in a contrast mode.
5. The tight sandstone reservoir fluid identification method based on the log generalized S-transform of claim 2, wherein the two conditions of the IMF component comprise,
(1) the number of IMF extrema (sum of the number of maxima and minima) and the number of zero crossings must be equal or differ by at most 1;
(2) at any point in the IMF, the average of the envelope defined by the local maxima and the average of the envelope defined by the local minima should be equal to zero. If so, the signal is an IMF component; if not, the analysis of the first to fourth steps is repeated based on the signal.
CN202210062143.XA 2022-01-19 2022-01-19 Tight sandstone reservoir fluid identification method based on logging curve generalized S transformation Pending CN114415241A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210062143.XA CN114415241A (en) 2022-01-19 2022-01-19 Tight sandstone reservoir fluid identification method based on logging curve generalized S transformation

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210062143.XA CN114415241A (en) 2022-01-19 2022-01-19 Tight sandstone reservoir fluid identification method based on logging curve generalized S transformation

Publications (1)

Publication Number Publication Date
CN114415241A true CN114415241A (en) 2022-04-29

Family

ID=81274490

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210062143.XA Pending CN114415241A (en) 2022-01-19 2022-01-19 Tight sandstone reservoir fluid identification method based on logging curve generalized S transformation

Country Status (1)

Country Link
CN (1) CN114415241A (en)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109557578A (en) * 2017-09-27 2019-04-02 中国石油化工股份有限公司 A kind of reservoir gas-bearing property detection method and device
CN111830559A (en) * 2019-04-19 2020-10-27 中国石油天然气股份有限公司 Method and device for frequency extension processing of seismic data
US20210181365A1 (en) * 2019-12-12 2021-06-17 King Fahd University Of Petroleum And Minerals Adaptive noise estimation and removal method for microseismic data
CN113608259A (en) * 2021-07-16 2021-11-05 长江大学 Seismic thin layer detection method based on ICEEMDAN constraint generalized S transformation

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109557578A (en) * 2017-09-27 2019-04-02 中国石油化工股份有限公司 A kind of reservoir gas-bearing property detection method and device
CN111830559A (en) * 2019-04-19 2020-10-27 中国石油天然气股份有限公司 Method and device for frequency extension processing of seismic data
US20210181365A1 (en) * 2019-12-12 2021-06-17 King Fahd University Of Petroleum And Minerals Adaptive noise estimation and removal method for microseismic data
CN113608259A (en) * 2021-07-16 2021-11-05 长江大学 Seismic thin layer detection method based on ICEEMDAN constraint generalized S transformation

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
WEI HUANG 等: "Application of generalized S-transform based on empirical mode decomposition in seismic data frequency division", 《INTERNATIONAL GEOPHYSICAL CONFERENCE》 *

Similar Documents

Publication Publication Date Title
Zhu et al. Intelligent logging lithological interpretation with convolution neural networks
CN104849757B (en) Eliminate random noise system and method in seismic signal
CN103487835A (en) Multi-resolution wave impedance inversion method based on model constraints
CN109581487B (en) Sedimentary body waveform space structure analysis and body-controlled inversion modeling method
CN108550116A (en) The self-adapting random resonant denoising method of silicon monocrystal growth image under low signal-to-noise ratio
Zhu et al. Rapid identification of high-quality marine shale gas reservoirs based on the oversampling method and random forest algorithm
CN110687597B (en) Wave impedance inversion method based on joint dictionary
CN114638255A (en) Tight sandstone fracture logging comprehensive identification method combined with deep learning
CN104730576A (en) Curvelet transform-based denoising method of seismic signals
CN112016477A (en) Logging deposition microphase identification method based on deep learning
CN111046328A (en) Rock type identification method based on logging curve wavelet Mallet algorithm
CN110927817B (en) Method for optimally designing and controlling thin-layer superposed riverway horizontal well track
CN112800590B (en) Grid coarsening method for machine learning-assisted two-phase flow oil reservoir random modeling
CN112796738A (en) Stratum permeability calculation method combining array acoustic logging and conventional logging
CN114415241A (en) Tight sandstone reservoir fluid identification method based on logging curve generalized S transformation
CN111598340A (en) Thin sand body plane spread prediction method based on fractional order Hilbert transform
CN114035226B (en) Microseism data noise suppression method and system based on noise semantic correlation depth migration noise removal self-coding network
CN111273346A (en) Method, device, computer equipment and readable storage medium for removing deposition background
GB2345776A (en) Method of determining geological facies
CN111366976A (en) Seismic attribute self-adaptive median filtering method based on scale guide
CN113325474B (en) Method for discriminating biological reef
CN117407841B (en) Shale layer seam prediction method based on optimization integration algorithm
CN112882092B (en) Well-seismic joint inversion method, system, medium and application based on multi-attribute CNN
Li et al. Hybrid Swin Transformer-CNN Model for Pore-crack Structure Identification
CN115238766A (en) Well-seismic combined oil layer prediction method based on full-sequence convolutional neural network

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination