CN108241171A - A kind of complex value Gauss integration filters and seismic data is filtered and is extracted three wink attribute method - Google Patents

A kind of complex value Gauss integration filters and seismic data is filtered and is extracted three wink attribute method Download PDF

Info

Publication number
CN108241171A
CN108241171A CN201711467102.4A CN201711467102A CN108241171A CN 108241171 A CN108241171 A CN 108241171A CN 201711467102 A CN201711467102 A CN 201711467102A CN 108241171 A CN108241171 A CN 108241171A
Authority
CN
China
Prior art keywords
frequency
complex value
filter
gauss
filters
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.)
Granted
Application number
CN201711467102.4A
Other languages
Chinese (zh)
Other versions
CN108241171B (en
Inventor
姬战怀
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.)
XI'AN SHIWEN SOFTWARE Co Ltd
Xian University of Science and Technology
Original Assignee
XI'AN SHIWEN SOFTWARE Co Ltd
Xian 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 XI'AN SHIWEN SOFTWARE Co Ltd, Xian University of Science and Technology filed Critical XI'AN SHIWEN SOFTWARE Co Ltd
Priority to CN201711467102.4A priority Critical patent/CN108241171B/en
Publication of CN108241171A publication Critical patent/CN108241171A/en
Application granted granted Critical
Publication of CN108241171B publication Critical patent/CN108241171B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
    • G01V1/00—Seismology; Seismic or acoustic prospecting or detecting
    • G01V1/28—Processing seismic data, e.g. analysis, for interpretation, for correction
    • G01V1/30—Analysis
    • G01V1/307—Analysis for determining seismic attributes, e.g. amplitude, instantaneous phase or frequency, reflection strength or polarity
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
    • G01V1/00—Seismology; Seismic or acoustic prospecting or detecting
    • G01V1/28—Processing seismic data, e.g. analysis, for interpretation, for correction
    • G01V1/36—Effecting static or dynamic corrections on records, e.g. correcting spread; Correlating seismic signals; Eliminating effects of unwanted energy
    • G01V1/364—Seismic filtering
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
    • G01V2210/00—Details of seismic processing or analysis
    • G01V2210/30—Noise handling
    • G01V2210/32—Noise reduction
    • G01V2210/324—Filtering
    • 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
    • G01V2210/63—Seismic attributes, e.g. amplitude, polarity, instant phase

Abstract

The present invention propose a complex value Gauss integration filter and use it for extraction seismic signal three wink attribute, method constructs a complex value wave filter based on Gauss windows.This wave filter be zero phase complex value wave filter, frequency function programmable single-chip system ideal gate function, effectively inhibit Fourier frequency domain ideal door and window wave filters Gibbs phenomenons.With this complex value Gauss integration filters to Seismic signal filtering, output signal real part is original signal filter result, and imaginary part is converted for the Hilbert of solid part signal, thus result can facilitate calculate seismic signal three wink attribute.Thus method calculates the defects of three wink of seismic signal attribute is overcome when being calculated using Hilbert transform methods to noise-sensitive.

Description

A kind of complex value Gauss integration filters and it is filtered and extracts three to seismic data Wink attribute method
Technical field
The invention belongs to oil-gas seismic exploration fields, and in particular to the construction of complex digital wave filter and to seismic prospecting signal It filters and frequency-division filter is handled, to obtain the method for stable and accurate seismic data instantaneous attribute, including：Complex value Gauss is accumulated The construction of filter-divider；The setting of complex value Gauss integration filter height cut-off frequency and Numerical Implementation；At the filtering of seismic data Reason；Earthquake instantaneous attribute is extracted by filter result.
Background technology
In oil-gas seismic exploration, instantaneous attribute is widely used poststack seismic properties.Instantaneous phase is usually used in detecting The unconformability on stratum, tomography and stratum laterally vary；In journal article《Seismic attributes----A historical perspective》Chorpra etc. is used to instantaneous frequency identify seismic data in (Geophysics, 2005) Decay pattern of anomaly and thin layer tuning.Robertson etc. is respectively in document《Complex seismic trace analysis of thin beds》(Geophysics, 1984) and《Seismic interpretation 9-Complex seismic trace attributes》Instantaneous attribute is used for tlc analysis and earthquake common version in (The Leading Edge, 1988)；Hart In document《Channel detection in 3-D seismic data using sweetness》(AAPG Bulletin, 2008) in river is identified with instantaneous amplitude；Zeng is in document《Geologic significance of anomalous instantaneous frequency》Instantaneous frequency anomalous identification thin layer and lithofacies boundary are utilized in (Geophysics, 2010) Deng.
Complex seismic trace technology is to extract the basis of instantaneous seismic attributes, wherein, instantaneous amplitude and instantaneous phase are bases This complex seismic trace attribute, other instantaneous attributes can be generated by their differential, average, combination or transformation.Complex seismic trace is One analytic signal, it is made of two parts of real and imaginary parts, and real part is original signal, and imaginary part is the Hilbert of real part Transformation.The study found that Hilbert is converted to noise-sensitive, error is larger when calculating instantaneous attribute by seismic data with low signal-to-noise ratio, Difficulty is brought to seismic attributes analysis.To overcome this shortcoming, Barnes is in document《A tutorial on complex seismic trace analysis》It is pointed out in (Geophysics, 2007) flat using filtering and weighting during extraction instantaneous attribute Noise effect can effectively be overcome by carrying out processing.Luo etc. is in document《Generalized Hilbert transform and its applications in geophysics》The Hilbert transformation of broad sense is introduced in (The Leading Edge, 2003) To calculate instantaneous attribute, method overcomes the defects of Hilbert transformation is to noise-sensitive；Lu and Zhang are in document《Robust estimation of instantaneous phase using a time-frequency adaptive filter》 In (Geophysics, 2013) under the premise of assuming that the frequency content of high amplitude has higher signal-to-noise ratio in seismic data, A zero phase sef-adapting filter is constructed with the time-frequency spectrum of STFT, this wave filter can enhance the frequency content of high amplitude, press The frequency content of short arc processed, so as to fulfill the Robust Estimation to earthquake signal transient phase attributes.Lu and Zhang are in document《A robust instantaneous frequency estimation method》(EAGE,Extended Abstracts, 2011) sef-adapting filter is proposed to estimate the algorithm of instantaneous frequency.
The present invention constructs a complex value zero-phase filters, and Hilbert is replaced when being and multiple trace analysis is done to seismic data The new method of transformation.Method integrates warbled Gauss window functions in given frequency separation, generates a time domain zero Phase filter.With this wave filter to Seismic signal filtering, response is complex signal, and the imaginary part of response signal is real part Hilbert is converted, and can facilitate extraction signal transient attribute with this filter response.
Invention content
Technical problems to be solved
It needs to do seismic data Hilbert transformation using the multiple trace analysis technology extraction instantaneous attribute of seismic data, and Hilbert is converted to noise-sensitive, and the instantaneous attribute error for making extraction is larger, is unfavorable for seismic data interpretive analysis.The present invention To overcome the shortcomings of that Hilbert is converted, a complex value Gauss integration filter is proposed, substituting Hilbert transformation realizes earthquake Data answer trace analysis, and method conveniently steadily can extract instantaneous seismic attributes.
Technical solution
A kind of complex value Gauss integration filters, the frequency separation specified after frequency modulation(PFM) at one by a Gauss window Inner product it is mitogenetic into time domain filtering, it is characterised in that expression formula is as follows：
Wherein, f1And f2The respectively height cut-off frequency of wave filter, and 0≤f1＜ f2；σ be multiple Gauss windows resolution ratio because Son；J is imaginary unit, i.e.,T is the time.
A kind of discretization method of complex value Gauss integration filters, it is characterised in that step is as follows：
Step 1：The high and low cut-off frequency f of wave filter is set1And f2, and f2＞ f10；
Step 2：The resolution factor σ of the multiple Gauss windows of setting, takes σ >=2；
Step 3：The time sampling interval Δ t of digital signal to be filtered is obtained, complex value Gauss integration filter sequences are set Time sampling interval it is equal with the time sampling interval of digital signal to be filtered；
Step 4：The frequency sampling interval delta f of the numerical integration of integration filter is set, takes Δ f≤0.001Hz；
Step 5：Taking filter temporal section, wherein T ＞ 0 for each fixed time t, are accumulated using numerical value for [- T, T] Divide and calculate filter sequence in moment numerical value Fσ(t)。
It is a kind of using complex value Gauss integration filters seismic data is filtered and is extracted three wink attribute method, It is characterized in that step is as follows：
Step 1：To the seismic data to be filtered, several representative sections are extracted, are FFT by road, and ask for earthquake and cut open The average frequency spectrum in face determines the frequency range of seismic data；
Step 2：Generate complex value Gauss device digital wave filters：
Step 2a：According to the frequency range of seismic data and practical application needs, complex value Gauss integration filters are set Low cut-off frequency f1With high cut-off frequency f2, the frequency domain sample interval delta f of numerical integration is set, the resolution factor σ of frequency modulation Gauss windows is set, Often take σ >=2；
Step 2b：To it is any given at the time of t, calculate complex value Gauss integration filters moment t numerical value：
Wherein,Operator " [] " represents logarithm rounding, andFor imaginary unit；
Step 2c：The time domain of multiple trace analysis, filter temporal sampling interval Δ t values and seismic data is done to seismic data Sampling interval is equal；Filter temporal section is [- T, T], and wherein T ＞ 0, complex value Gauss integration filter sequence of values has 2M+ 1 complex value point, wherein,Operator " [] " represents logarithm rounding；
The value mode of T is as follows：If in moment t, have
||Fσ(t)||≤1.0e-6
So, if there are continuous 6 positive integers so that
||Fσ(t+k Δs t) | |≤1.0e-6, k=1,2 ..., 5
Method takes
T=t+5 Δs t
Gauss integration complex filter sequence of values is obtained, is denoted as
{Fσ(k) }, k=-M,-M+1 ..., -1,0,1, M-1 ..., M
Fσ(k) value of the wave filter in moment k Δ t is represented；
Step 3：The complex value Gauss device digital wave filters F generated using step 2σ(k) to seismic data Tr (n), n= 0,1 ..., N are filtered by road：
Wherein, Re (GTr(n)) it is GTr(n) real part is that seismic channel Tr (n) is filtered through complex value Gauss integration filters As a result；Im(GTr(n)) it is GTr(n) imaginary part is Re (GTr(n)) Hilbert transformation；
Step 4：Re (the G obtained using step 3TrAnd Im (G (n))Tr(n)) instantaneous amplitude, instantaneous of seismic data is calculated Phase and instantaneous frequency instantaneous attribute：
The instantaneous amplitude InAm (n) is
The instantaneous phase InPh (n) is
InPh (n)=arctan (Im (GTr(n))/Re(GTr(n))), n=0,1 ..., N
The instantaneous frequency InFr (n) is
Advantageous effect
A kind of complex value Gauss integration filters proposed by the present invention and its shake data over the ground and be filtered and extract three winks The method of attribute substitutes Hilbert transformation using complex value Gauss integration filters, realizes that seismic data answers trace analysis.It is deemed-to-satisfy4 It can stablize, overcome deficiency of the Hilbert transformation to noise-sensitive, can stablize, accurately extract three wink of earthquake attribute.By repairing Change the high and low cut-off frequency of wave filter, the low pass or bandpass filter of different frequency bands can be generated, needed with adapting to different seismic data analysis It asks.Filter result is complex value analytic signal, can facilitate three wink of extraction attribute.Method And Principle understands, realizes simple, user's operation It is few.
Description of the drawings
The low cut-off frequency f of Fig. 11=0Hz, high cut-off frequency f2The low-pass filter of=40Hz, resolution factor change to filters affect Comparison diagram：(a) of Fig. 1, (b), (c) and (d) is respectively resolution factor when being 1, the theoretical spectrum figure of low-pass filter, time domain The FFT spectrum of numerical filters, the real and imaginary parts of time domain numerical filters；(A), (B), (C) and (D) of Fig. 1 is respectively point When the resolution factor is 4, the theoretical spectrum figure of low-pass filter, the FFT spectrum of time domain numerical filters, time domain numerical filters Real and imaginary parts；It is compared from Fig. 1 as can be seen that increasing with resolution factor, filter spectrum function approximation ideal gate function.
The low cut-off frequency f of Fig. 21=10Hz, high cut-off frequency f2Bandpass filter during=20Hz, resolution factor change to wave filter shadow Ring comparison diagram：(a) of Fig. 2, (b), (c) and (d) is respectively resolution factor when being 1, the theoretical spectrum figure of bandpass filter, when The FFT spectrum of domain numerical filters, the real and imaginary parts of time domain numerical filters；(A), (B), (C) and (D) of Fig. 2 is respectively When resolution factor is 4, the theoretical spectrum figure of bandpass filter, the FFT spectrum of time domain numerical filters, time domain numerical filters Real and imaginary parts；It is compared from Fig. 2 as can be seen that increasing with resolution factor, filter spectrum function approximation ideal gate function.
Fig. 3 Hilbert are converted and complex value Gauss integration filters realize multiple trace analysis effect contrast figure：Subgraph in Fig. 3 (a), (b) and (c) is real part, imaginary part and the instantaneous phase figure of noiseless 10Hz sines analytic signal；Subgraph (A1), (B1) and (C1) it is respectively to scheme real signal shown in sub (a) signal-to-noise ratio generated after noise is added to convert for the signals and associated noises of 23dB, through Hilbert The orthogonal signalling of generation and the instantaneous phase attribute by its calculating；Subgraph (A2), (B2) and (C2) is respectively real shown in subgraph (a) The signal-to-noise ratio generated after signal plus noise is the signals and associated noises of 23dB, with the parsing of complex value Gauss integration filters filtering generation Real part, imaginary part and the instantaneous phase attribute by its signal calculated of signal.As can be seen that with complex value Gauss integration filters It is more accurate to make the instantaneous phase attribute that multiple trace analysis calculates.
The low cut-off frequency f of Fig. 41=0Hz, high cut-off frequency f2The low pass complex value Gauss integration filters processing original place shake of=125Hz Data, and calculate instantaneous phase, instantaneous frequency and instantaneous amplitude attribute：(a) former seismic channel section；(b) it is integrated by complex value Gauss The former seismic profile imaginary part section that filter filtering obtains；(c) original place extracted by complex value Gauss integration filters filter result Shake section instantaneous phase attribute section；(d) the instantaneous frequency of former seismic profile extracted by complex value Gauss integration filters filter result Rate attribute section；(e) the former seismic profile instantaneous amplitude attribute section extracted by complex value Gauss integration filters filter result； (f) each road frequency spectrum of former seismic profile (red curve is average frequency spectrum).
The low cut-off frequency f of Fig. 51=0Hz, high cut-off frequency f2The low pass complex value Gauss integration filters of=30Hz, to original earthquake Section scaling down processing and multiple trace analysis, and three wink of extraction seismic profile：(a) by low pass complex value Gauss integration filter filtering The former seismic profile of reason obtains dividing real seismic profile；(b) by low pass complex value Gauss integration filter filtering process original seismic profiles It obtains dividing empty seismic profile；(c) the former seismic profile of frequency dividing is filtered by low pass complex value Gauss integration filters and extracts it instantaneously Phase attributes section；(d) by the instantaneous frequency attribute section of low pass complex value Gauss integration filters frequency dividing extraction；(e) by low pass The instantaneous amplitude attribute section of complex value Gauss integration filters frequency dividing extraction；(f) it is filtered by complex value Gauss integration filters Each road frequency spectrum of frequency dividing reality seismic profile arrived.
The low cut-off frequency f of Fig. 61=20Hz, high cut-off frequency f2The band logical complex value Gauss integration filters of=35Hz, to original earthquake Section scaling down processing and multiple trace analysis, and three wink of extraction seismic profile：(a) by band logical complex value Gauss integration filter filtering The former seismic profile of reason obtains dividing real seismic profile；(b) by band logical complex value Gauss integration filter filtering process original seismic profiles It obtains dividing empty seismic profile；(c) by the instantaneous phase attribute section of band logical complex value Gauss integration filters frequency dividing extraction；(d) By the instantaneous frequency attribute section of band logical complex value Gauss integration filters frequency dividing extraction；(e) it is integrated and filtered by band logical complex value Gauss The instantaneous amplitude attribute section of wave device frequency dividing extraction；(f) frequency dividing filtered by band logical complex value Gauss integration filters is on the spot Shake each road frequency spectrum of section.
Specific embodiment
Contain abundant earth formation information in seismic data.But noisy seismic data reduces seismic data resolution, It is unfavorable for describing earth formation with seismic data；In addition, instantaneous seismic attributes are commonly used for stratigraphic structure analysis, but calculate instantaneous The Hilbert that attribute uses is converted to noise-sensitive so that the instantaneous attribute distortion of extraction, it is impossible to reflect true stratigraphic structure Information is unfavorable for describing earth formation.For this purpose, propose a complex value Gauss integration filter.
The principle of the invention：
1) complex value Gauss integration filters
Complex value Gauss integration filters are in the frequency separation specified after frequency modulation(PFM) at one by a Gauss window The time domain filtering of generation is integrated, is formed as follows
Wherein, f1And f2The respectively height cut-off frequency of wave filter, and 0≤f1＜ f2；σ be multiple Gauss windows resolution ratio because Son；J is imaginary unit, i.e.,As can be seen that complex value Gauss integration filters are a complex value filtering from formula (1) Device, its real part are time domain even function, and imaginary part is odd function.Thus, this wave filter is zero-phase filters, and filtering does not change Original signal phase information.Its frequency functionFor
The characteristics of complex value Gauss integration filters, is as follows：
A) complex value Gauss integration filters are a time domain zero phase complex filters, are to be by the basic frequency of multiple Gauss windows The integral function of integration variable；
B) appropriate integrating range is set, i.e., the high and low cut-off frequency of given wave filter can construct different low pass or band logical Complex value Gauss integration filters, can be used to signal filtering or scaling down processing；
C) resolution factor of multiple Gauss windows is adjusted, filter frequency domain response can be made to approach ideal rectangle gate function, suppressed Wave filter secondary lobe efficiently controls filter frequency domain resolution ratio；
D) with complex value Gauss integration filters to seismic channel filter, generate complex seismic trace, wherein the imaginary part of complex seismic trace with Real part is orthogonal, i.e., imaginary part is the Hilbert transformation of real part.
E) complex value Gauss integration filters, which have, well filters performance of making an uproar, and is calculated with its filter result to seismic channel instantaneous Seismic properties overcome Hilbert transform methods and calculate the defects of instantaneous attribute is to noise-sensitive.It can be easily by filter result 4 the empty seismic channel of extraction, instantaneous amplitude, instantaneous phase and instantaneous frequency seismic data attributes.
2) seismic data filtering and instantaneous attribute extraction
Seismic data is formed by basic composition unit of seismic channel, therefore seismic data filtering and instantaneous seismic attributes carry Take is also what is realized by road.If appointing and taking a seismic channel, Tr (t) is denoted as, utilizes complex value Gauss integration filters Fσ(t) to Tr (t) it filters, result is denoted as GTr(t), that is, have
GTr(t)=Fσ(t)*Tr(t)
=Re (Fσ(t))*Tr(t)+jIm(Fσ(t))*Tr(t)
=Re (GTr(t))+jIm(GTr(t)) (3)
In formula (3), operator " * " represents time domain convolution operation.Complex value Gauss integration filters Fσ(t) high and low cut-off frequency Value is different, can generate different low passes or bandpass filter, available for being filtered to former seismic channel or seismic channel being done at frequency dividing Reason.Instantaneous seismic attributes can be easily extracted by filter result, instantaneous amplitude InAm (t) is
Instantaneous phase InPh (t) is
InPh (t)=arctan (Im (GTr(t))/Re(GTr(t))) (5)
Instantaneous frequency InFr (t) is
Other instantaneous attributes can be derived from by attribute above, not repeated herein.
To frequency modulation Gauss windows in assigned frequency section upper integral, one zero phase time domain complex value wave filter of generation.With multiple The resolution parameter increase of Gauss windows, the frequency function of wave filter will approach preferable gate function, wave filter made to have high resolution Rate.The present invention realizes complex value Gauss integration filters with numerical method.Implementation method is as follows：
1) the high and low cut-off frequency f of setting wave filter1And f2, and f2＞ f1>=0. wherein, works as f1When=0, wave filter is low-pass filtering Device；Work as f1During ＞ 0, wave filter is bandpass filter；
2) the resolution factor σ of multiple Gauss windows is set, σ >=2. is taken to increase with resolution factor σ, the frequency letter of wave filter Number approaches preferable gate function；
3) the time sampling interval Δ t of digital signal for needing to filter is obtained, complex value Gauss integration filter sequences are set Time sampling interval it is equal with the digital time sampling interval to be filtered；
4) the frequency sampling interval delta f of the numerical integration of setting integration filter, takes Δ f≤0.001Hz；
5) filter temporal section is taken as [- T, T], and wherein T ＞ 0 using duration Δ t as time interval, obtain wave filter number Value sequence Fσ(k) (k=-N,-N+1 ..., -1,0,1 ..., N-1, N), Fσ(k) filter value in moment k Δ t is represented,Operator [] is represented to data rounding.For each fixed time t, filter sequence is calculated with numerical integration In moment numerical value Fσ(t)。
Pass through the complex value Gauss integration filter sequences F of prior art schemes generationσ(k), seismic signal is filtered with it Wave.Its implementation is as follows：
1) in the seismic data for needing to filter, one or several seismic profiles are taken out, each earthquake is calculated by road using FFT The frequency spectrum in road, and average frequency spectrum is calculated, to determine the frequency range of pending seismic data；
2) the height cut-off frequency of Gauss integration complex filters is determined according to actual needs, and the filter time domain sampling interval is set, When such as extracting the instantaneous attribute of frequency dividing seismic data, according to frequency dividing it needs to be determined that wave filter height cut-off frequency, is realized by aforementioned schemes Wave filter Fσ(t)；
If 3) Tr (t) is a seismic channel in seismic data, filter result is
GTr(t)=Fσ(t)*Tr(t) (7)
Formula (7) result is a complex-valued sequences；
4) complex-valued sequences generated by filter filtering can acquire the instantaneous of seismic channel by formula (4), (5) and (6) and shake Width, instantaneous phase and instantaneous frequency.
Specific embodiment：
The first step, data analysis.
To seismic data to be processed, several seismic profiles are extracted, FFT is by road, and counts and obtains seismic channel in section Average frequency spectrum, determine the frequency range and dominant frequency band of seismic data；
Second step realizes complex value Gauss integrated value wave filters
1) according to practical application request, the low cut-off frequency f of complex value Gauss integration filters is set1With high cut-off frequency f2, number is set It is worth the frequency domain sample interval delta f of integration, the resolution factor σ of frequency modulation Gauss windows is set, often takes σ >=2.
2) to it is any given at the time of t, using numerical integration according to formula (1) calculate complex value Gauss integration filters when The numerical value of t is carved, as a result as shown in formula (8)
In formula (8),Operator " [] " represents logarithm rounding, andFor imaginary unit；
3) time-domain sampling of multiple trace analysis, filter temporal sampling interval Δ t values and seismic data is done to seismic data It is spaced equal.Filter temporal section is [- T, T] (T ＞ 0), and complex value Gauss integration filter sequence of values has 2M+1 complex value Point, wherein,Operator " [] " represents logarithm rounding.
The value mode of T is as follows：If in moment t, have
||Fσ(t)||≤1.0e-6 (9)
So, if there are continuous 6 positive integers so that
||Fσ(t+k Δs t) | |≤1.0e-6, k=1,2 ..., 5 (10)
Method takes
T=t+5 Δs t (11)
Gauss integration complex filter sequence of values can obtain according to formula (8), be denoted as
{Fσ(k) }, k=-M,-M+1 ..., -1,0,1, M-1 ..., M (12)
In formula (12), Fσ(k) value of the wave filter in moment k Δ t is represented.
Third walks, and complex seismic trace is generated to seismic data filtering with complex value Gauss integration filters.
Seismic data is using seismic channel as basic storage unit, and it is by section sequence, by road that seismic data, which answers trace analysis, It realizes.A seismic channel is taken, is denoted as Tr (n), n=0,1 ..., N. filters seismic channel with complex value Gauss integration filters, filter Wave result is denoted as GTr(n), have
In formula (13), Re (GTr(n)) it is GTr(n) real part is that seismic channel Tr (n) is filtered through complex value Gauss integration filters The result of wave；Im(GTr(n)) it is GTr(n) imaginary part is Re (GTr(n)) Hilbert transformation.
4th step calculates instantaneous seismic attributes with the multiple road of filtering generation.
It is calculated using formula (4), (5) and (6), instantaneous amplitude InAm (n) is
Instantaneous phase InPh (n) is
InPh (n)=arctan (Im (GTr(n))/Re(GTr(n))), n=0,1 ..., N (15)
Instantaneous frequency InFr (n) is
Other instantaneous attributes can generate or consult pertinent literature by upper three instantaneous attributes.

Claims (3)

1. a kind of complex value Gauss integration filters, in the frequency separation specified after frequency modulation(PFM) at one by a Gauss window Integrate the time domain filtering of generation, it is characterised in that expression formula is as follows：
Wherein, f1And f2The respectively height cut-off frequency of wave filter, and 0≤f1＜ f2；σ is the resolution factor of multiple Gauss windows；J is Imaginary unit, i.e.,T is the time.
2. a kind of discretization method of complex value Gauss integration filters described in claim 1, it is characterised in that step is as follows：
Step 1：The high and low cut-off frequency f of wave filter is set1And f2, and f2＞ f1≥0；
Step 2：The resolution factor σ of the multiple Gauss windows of setting, takes σ >=2；
Step 3：Obtain the time sampling interval Δ t of digital signal to be filtered, setting complex value Gauss integration filter sequences when Between the sampling interval it is equal with the time sampling interval of digital signal to be filtered；
Step 4：The frequency sampling interval delta f of the numerical integration of integration filter is set, takes Δ f≤0.001Hz；
Step 5：Taking filter temporal section, wherein T ＞ 0 for each fixed time t, utilize numerical integration meter for [- T, T] Filter sequence is calculated in moment numerical value Fσ(t)。
3. a kind of be filtered seismic data and extract three winks using complex value Gauss integration filters described in claim 1 The method of attribute, it is characterised in that step is as follows：
Step 1：To the seismic data to be filtered, several representative sections are extracted, FFT is, and ask for seismic profile by road Average frequency spectrum determines the frequency range of seismic data；
Step 2：Generate complex value Gauss device digital wave filters：
Step 2a：According to the frequency range of seismic data and practical application needs, low section of setting complex value Gauss integration filters Frequency f1With high cut-off frequency f2, the frequency domain sample interval delta f of numerical integration is set, the resolution factor σ of frequency modulation Gauss windows is set, is often taken σ≥2；
Step 2b：To it is any given at the time of t, calculate complex value Gauss integration filters moment t numerical value：
Wherein,Operator " [] " represents logarithm rounding, andFor imaginary unit；
Step 2c：The time-domain sampling of multiple trace analysis, filter temporal sampling interval Δ t values and seismic data is done to seismic data It is spaced equal；Filter temporal section is [- T, T], wherein T ＞ 0, and complex value Gauss integration filter sequence of values has 2M+1 Complex value point, wherein,Operator " [] " represents logarithm rounding；
The value mode of T is as follows：If in moment t, have
‖Fσ(t)‖≤1.0e-6
So, if there are continuous 6 positive integers so that
‖Fσ(t+k Δ t) ‖≤1.0e-6, k=1,2 ..., 5
Method takes
T=t+5 Δs t
Gauss integration complex filter sequence of values is obtained, is denoted as
{Fσ(k) }, k=-M,-M+1 ..., -1,0,1, M-1 ..., M
Fσ(k) value of the wave filter in moment k Δ t is represented；
Step 3：The complex value Gauss device digital wave filters F generated using step 2σ(k) to seismic data Tr (n), n=0, 1 ..., N is filtered by road：
Wherein, Re (GTr(n)) it is GTr(n) real part is the result that seismic channel Tr (n) is filtered through complex value Gauss integration filters； Im(GTr(n)) it is GTr(n) imaginary part is Re (GTr(n)) Hilbert transformation；
Step 4：Re (the G obtained using step 3TrAnd Im (G (n))Tr(n)) instantaneous amplitude, the instantaneous phase of seismic data are calculated With instantaneous frequency instantaneous attribute：
The instantaneous amplitude InAm (n) is
The instantaneous phase InPh (n) is
InPh (n)=arctan (Im (GTr(n))/Re(GTr), (n)) n=0,1 ..., N
The instantaneous frequency InFr (n) is
CN201711467102.4A 2017-12-29 2017-12-29 A method of using complex value Gauss integration filter seismic data is filtered and is extracted three wink attribute Active CN108241171B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201711467102.4A CN108241171B (en) 2017-12-29 2017-12-29 A method of using complex value Gauss integration filter seismic data is filtered and is extracted three wink attribute

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201711467102.4A CN108241171B (en) 2017-12-29 2017-12-29 A method of using complex value Gauss integration filter seismic data is filtered and is extracted three wink attribute

Publications (2)

Publication Number Publication Date
CN108241171A true CN108241171A (en) 2018-07-03
CN108241171B CN108241171B (en) 2019-06-18

Family

ID=62700271

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201711467102.4A Active CN108241171B (en) 2017-12-29 2017-12-29 A method of using complex value Gauss integration filter seismic data is filtered and is extracted three wink attribute

Country Status (1)

Country Link
CN (1) CN108241171B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2021124329A1 (en) * 2019-12-17 2021-06-24 E.Q. Earthquake Ltd. Systems and methods for earthquake detection and alerts

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2004044615A2 (en) * 2002-11-09 2004-05-27 Geoenergy, Inc. Method and apparatus for seismic feature extraction
CN105487119A (en) * 2014-10-08 2016-04-13 中国石油化工股份有限公司 Time domain fan filter and application thereof

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2004044615A2 (en) * 2002-11-09 2004-05-27 Geoenergy, Inc. Method and apparatus for seismic feature extraction
CN105487119A (en) * 2014-10-08 2016-04-13 中国石油化工股份有限公司 Time domain fan filter and application thereof

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
姬战怀: "改进的Morlet小波及其变换", 《计算机工程与应用》 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2021124329A1 (en) * 2019-12-17 2021-06-24 E.Q. Earthquake Ltd. Systems and methods for earthquake detection and alerts

Also Published As

Publication number Publication date
CN108241171B (en) 2019-06-18

Similar Documents

Publication Publication Date Title
Ahrabian et al. Synchrosqueezing-based time-frequency analysis of multivariate data
CN108009347B (en) Time-frequency analysis method based on synchronous compression joint improvement generalized S transformation
CN102692647B (en) Stratum oil-gas possibility prediction method with high time resolution
CN106405654A (en) Seismic spectrum imaging method based on deconvolution generalized S transform
CN107179550B (en) A kind of seismic signal zero phase deconvolution method of data-driven
CN104849756A (en) Method for improving resolution ratio of seismic data and enhancing energy of valid weak signals
CN107144879A (en) A kind of seismic wave noise-reduction method combined based on adaptive-filtering with wavelet transformation
CN104297791A (en) Inversion method and system based on earthquake dominant frequency
CN108020863A (en) A kind of thin and interbedded reservoir porosity prediction method based on earthquake parity function
CN106680874A (en) Harmonic noise suppression method based on waveform morphology sparse modeling
CN104932018A (en) Method for enhancing resolution of seismic section through compensating variable resolution factor S transform complex time-frequency spectrum
CN104597502A (en) Novel petroleum seismic exploration data noise reduction method
CN104635264B (en) The processing method of earthquake data before superposition and equipment
CN107526016B (en) A kind of detection method and device for semiconductor devices 1/f noise bound frequency
CN102866429B (en) Method for determining groundwater occurrence
CN108241171B (en) A method of using complex value Gauss integration filter seismic data is filtered and is extracted three wink attribute
CN104422956B (en) A kind of high precision seismic spectral factorization method based on Sparse Pulse Inversion
CN105334532A (en) Seismic wavelet estimation method
CN108181486A (en) The processing method and processing device of acceleration signal
CN104027095A (en) Pulse data preprocessing method
CN102353991B (en) Method for analyzing seismic instantaneous frequency based on physical wavelet matched with seismic wavelet
Lin et al. Varying-window-length time-frequency peak filtering and its application to seismic data
CN104267257B (en) Signal amplitude spectrum detecting method and system based on dot frequency filter
CN114429157A (en) Method for analyzing terrestrial physical signal characteristics
Boudraa Instantaneous frequency estimation of fm signals by ψb-energy operator

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
GR01 Patent grant
GR01 Patent grant