CN104199102A - Method and device for acquiring full stratigraphic seism fractional frequency profile - Google Patents

Abstract

The invention provides a method and device for acquiring a full stratigraphic seism fractional frequency profile. The method includes (a) acquiring the bispectrum of a seismic record according to the acquired seismic record; (b) acquiring seismic wavelet according to the bispectrum of the seismic record; (c) acquiring the refraction reflection coefficient of the stratum according to the seismic wavelet; (d) acquiring the full stratigraphic seism fractional frequency profile with the predetermined resolution according to the refraction reflection coefficient of the stratum and the Yu wavelet with the predetermined dominant frequency.

Description

Obtain method and the device of full stratum earthquake frequency division section

Technical field

The present invention relates to the process field of seismologic record, more particularly, relate to a kind of method and device that obtains full stratum earthquake frequency division section.

Background technology

Frequency division seismic sequence analytical technology is an important technology of Seismic Sedimentology, and the analysis of frequency division seismic sequence is to carry out on the basis of the earthquake frequency division section obtaining according to seismologic record.Therefore, how to obtain the hot issue that the earthquake frequency division section with full formation information becomes research day by day.

At present, the method for obtaining earthquake frequency division section mainly contains frequency filtering method and time-frequency spectrum decomposition method.Yet, while utilizing frequency filtering method and time-frequency spectrum decomposition method to obtain earthquake frequency division section, need to from seismologic record, select a frequency range, thereby obtain the earthquake frequency division section of selected frequency range.Therefore in the earthquake frequency division section, obtaining, can lose a lot of formation informations that lineups in selected frequency range do not carry.And, according to the earthquake frequency division section obtaining, cannot learn contacting between the earthquake frequency division section that earthquake frequency division section corresponding to this frequency range is corresponding with other frequency ranges.

Therefore, need a kind of method of obtaining earthquake frequency division section, obtain having the earthquake frequency division section of full formation information.

Summary of the invention

The object of the present invention is to provide a kind of method and device that obtains full stratum earthquake frequency division section, thereby can obtain the earthquake frequency division section with full formation information.

An aspect of of the present present invention provides a kind of method of obtaining full stratum earthquake frequency division section, comprising: two spectrums of (a) obtaining seismologic record according to seismologic record; (b) according to two spectrums of seismologic record, obtain seismic wavelet; (c) according to seismic wavelet, obtain the sparse reflection coefficient on stratum; (d) according to the Yu Shi wavelet of the sparse reflection coefficient on stratum and predetermined dominant frequency, obtain the full stratum earthquake frequency division section of predetermined resolution.

Alternatively, step (b) comprising: the spectral amplitude that (b1) obtains seismic wavelet according to the spectral amplitude of two spectrums of seismologic record; (b2) according to the phase spectrum of two spectrums of seismologic record, obtain the phase spectrum of seismic wavelet; (b3) phase spectrum of the spectral amplitude of seismic wavelet and seismic wavelet is carried out to inversefouriertransform, obtain the seismic wavelet of time domain.

Alternatively, step (b1) comprising: the spectral amplitude that is obtained seismic wavelet by equation below: wherein, the vector forming for the natural logarithm of the spectral amplitude by seismic wavelet, for the vector that the natural logarithm of the spectral amplitudes of the two spectrums by seismologic record forms, A be according to the matrix of coefficients of the natural logarithm acquisition of two spectral amplitudes of composing of seismologic record.

Alternatively, vector element be the natural logarithm of the frequency of two seismic wavelets of the two spectrums sampling corresponding spectral amplitude in two spectrums of seismologic record in scheduled frequency range.

Alternatively, described scheduled frequency range is [N ₁hZ, N ₂hZ], sampling interval is MHZ, N ₁for being less than N ₂positive number, M is positive count, the sampling of the frequency of described two seismic wavelets in described scheduled frequency range is represented as respectively ω ₁and ω ₂, wherein, ω ₁with ω ₂some or all two spectrums that are combined in seismologic record in the natural logarithm of corresponding spectral amplitude as vector element.

Alternatively, ω ₁and ω ₂value be expressed as equation below: ω ₁(n)=N ₁+ (n-1) M; ω ₂(m)=ω ₁(n)+(m-1) M wherein, ω ₁(n) represent ω ₁n value, n is positive integer, ω ₂(m) represent ω ₂m value, m is positive integer, $m\∈[1,\frac{N_2+{3N}_1+M-{4\mathrm{\ω}}_1\left(n\right)}{4M}],$ And $\frac{N_2-N_1+M}{M}$ For can be by 4 positive integers that divide exactly.

Alternatively, the line number of coefficient matrices A is the columns of coefficient matrices A is the element a of the capable j row of any i in coefficient matrices A _{i, j}expression will with vector in the frequencies omega of two seismic wavelets corresponding to i element ₁and ω ₂in the time of in the formula of the natural logarithm of the spectral amplitude of the two spectrums of substitution, in formula coefficient, wherein, i is for being less than or equal to positive integer, j is for being less than or equal to positive integer, wherein, expression frequency is N ₁+ (j-1) natural logarithm of the spectral amplitude of the seismic wavelet of M.

Alternatively, the formula of the natural logarithm of the spectral amplitude of two spectrums is: wherein, $\tilde{W}\left({\mathrm{\ω}}_1\right)=\ln \left|W\left({\mathrm{\ω}}_1\right)\right|,\tilde{W}\left({\mathrm{\ω}}_2\right)=\ln \left|W\left({\mathrm{\ω}}_2\right)\right|,\tilde{W}({\mathrm{\ω}}_1+{\mathrm{\ω}}_2)=\ln \left|W({\mathrm{\ω}}_1+{\mathrm{\ω}}_2)\right|,$ | W (ω ₁) | for frequency is ω ₁the spectral amplitude of seismic wavelet, | W (ω ₂) | for frequency is ω ₂the spectral amplitude of seismic wavelet, | W (ω ₁+ ω ₂) | for frequency is ω ₁+ ω ₂the spectral amplitude of seismic wavelet.

Alternatively, step (b2) comprising: the phase spectrum that is obtained seismic wavelet by equation below: wherein, φ is the vector that the phase spectrum by seismic wavelet forms, for the vector that the two phase spectrums of composing by seismologic record form, H is the matrix of coefficients obtaining according to the phase spectrums of two spectrums of seismologic record.

Alternatively, vector element be the natural logarithm of the frequency of two seismic wavelets of the two spectrums sampling corresponding phase spectrum in two spectrums of seismologic record in scheduled frequency range.

Alternatively, described scheduled frequency range is [N ₁hZ, N ₂hZ], sampling interval is MHZ, N ₁for being less than N ₂positive number, M is positive count, the sampling of the frequency of described two seismic wavelets in described scheduled frequency range is represented as respectively ω ₃and ω ₄, wherein, ω ₃and ω ₄some or all two spectrums that are combined in seismologic record in the natural logarithm of corresponding phase spectrum as vector element.

Alternatively, ω ₃and ω ₄value be expressed as equation below: ω ₃(x)=N ₁+ (x-1) M; ω ₄(y)=ω ₃(x)+(y-1) M wherein, ω ₃(x) represent ω ₃x value, x is positive integer, ω ₄(y) represent ω ₄y value, y is positive integer, y ∈ $[1,\frac{N_2+N_1+M-{2\mathrm{\ω}}_3\left(x\right)}{2M}],$ And $\frac{N_2-N_1+M}{M}$ For can be by 2 positive integers that divide exactly.

Alternatively, the line number of matrix of coefficients H is the columns of matrix of coefficients H is the element h of the capable k row of any l in matrix of coefficients H _{l, k}expression will with vector in the frequencies omega of two seismic wavelets corresponding to l element ₃and ω ₄in the time of in the formula of the phase spectrum of the two spectrums of substitution, Ф (N in formula ₁+ (k-1) M) coefficient, wherein, l is for being less than or equal to positive integer, k is for being less than or equal to positive integer, wherein, Ф (N ₁+ (k-1) M) expression frequency is N ₁+ (k-1) phase spectrum of the seismic wavelet of M.

Alternatively, described scheduled frequency range is [1HZ, NHZ], and N is can be by 4 positive integers that divide exactly.

Alternatively, step (c) comprising: the seismic wavelet of time domain is carried out to sparse inversion, obtain the sparse reflection coefficient on the stratum of time domain.

Alternatively, step (d) comprising: the described predetermined dominant frequency of (d1) selecting Yu Shi wavelet; (d2) initial frequency of the Yu Shi wavelet based on selected described predetermined dominant frequency select time territory and termination frequency; (d3) will select the Yu Shi wavelet of the time domain of initial frequency and termination frequency to carry out Fourier transform, obtain the Yu Shi wavelet of frequency field, and the sparse reflection coefficient on the stratum of time domain is carried out to Fourier transform, obtain the sparse reflection coefficient on the stratum of frequency field; (d4) the sparse reflection coefficient on the stratum of the Yu Shi wavelet of frequency field and frequency field is carried out to convolution, obtain the full stratum earthquake frequency division section of predetermined resolution.

Alternatively, step (a) comprising: (a1) according to seismologic record, obtain skew stack geological data; (a2) window while skew stack geological data being divided into at least one; (a3) from described at least one time window while selecting one of them window, two spectrums of the seismologic record in the time of described in the skew stack seismic data acquisition when selected in window in window.

Alternatively, described predetermined resolution is directly proportional to described predetermined dominant frequency.

Another aspect of the present invention provides a kind of device that obtains full stratum earthquake frequency division section, comprising: two spectrum acquiring units, obtain the two of seismologic record according to seismologic record and compose; Wavelet acquiring unit, obtains seismic wavelet according to two spectrums of seismologic record; Reflection coefficient acquiring unit, obtains the sparse reflection coefficient on stratum according to seismic wavelet; Frequency division section acquiring unit, obtains the full stratum earthquake frequency division section of predetermined resolution according to the Yu Shi wavelet of the sparse reflection coefficient on stratum and predetermined dominant frequency.

Alternatively, wavelet acquiring unit comprises: wavelet amplitude getter, obtains the spectral amplitude of seismic wavelet according to the spectral amplitudes of two spectrums of seismologic record; Sub-wave phase is composed getter, obtains the phase spectrum of seismic wavelet according to the phase spectrums of two spectrums of seismologic record; Wavelet getter, carries out inversefouriertransform by the phase spectrum of the spectral amplitude of seismic wavelet and seismic wavelet, obtains the seismic wavelet of time domain.

Alternatively, use equation below obtains the spectral amplitude of seismic wavelet: wherein, the vector forming for the natural logarithm of the spectral amplitude by seismic wavelet, for the vector that the natural logarithm of the spectral amplitudes of the two spectrums by seismologic record forms, A be according to the matrix of coefficients of the natural logarithm acquisition of two spectral amplitudes of composing of seismologic record.

Alternatively, vector element be the natural logarithm of the frequency of two seismic wavelets of the two spectrums sampling corresponding spectral amplitude in two spectrums of seismologic record in scheduled frequency range.

Alternatively, described scheduled frequency range is [N ₁hZ, N ₂hZ], sampling interval is MHZ, N ₁for being less than N ₂positive number, M is positive count, the sampling of the frequency of described two seismic wavelets in described scheduled frequency range is represented as respectively ω ₁and ω ₂, wherein, ω ₁with ω ₂some or all two spectrums that are combined in seismologic record in the natural logarithm of corresponding spectral amplitude as vector element.

Alternatively, ω ₁and ω ₂value be expressed as equation below: ω ₁(n)=N ₁+ (n-1) M; ω ₂(m)=ω ₁(n)+(m-1) M wherein, ω ₁(n) represent ω ₁n value, n is positive integer, ω ₂(m) represent ω ₂m value, m is positive integer, m $\∈[1,\frac{N_2+{3N}_1+M-{4\mathrm{\ω}}_1\left(n\right)}{4M}],$ And $\frac{N_2-N_1+M}{M}$ For can be by 4 positive integers that divide exactly.

Alternatively, the line number of coefficient matrices A is the columns of coefficient matrices A is the element a of the capable j row of any i in coefficient matrices A _{i, j}expression will with vector in the frequencies omega of two seismic wavelets corresponding to i element ₁and ω ₂in the time of in the formula of the natural logarithm of the spectral amplitude of the two spectrums of substitution, in formula coefficient, wherein, i is for being less than or equal to positive integer, j is for being less than or equal to positive integer, wherein, expression frequency is N ₁+ (j-1) natural logarithm of the spectral amplitude of the seismic wavelet of M.

Alternatively, the formula of the natural logarithm of the spectral amplitude of two spectrums is: wherein, $\tilde{W}\left({\mathrm{\ω}}_1\right)=\ln \left|W\left({\mathrm{\ω}}_1\right)\right|,\tilde{W}\left({\mathrm{\ω}}_2\right)=\ln \left|W\left({\mathrm{\ω}}_2\right)\right|,\tilde{W}({\mathrm{\ω}}_1+{\mathrm{\ω}}_2)=\ln \left|W({\mathrm{\ω}}_1+{\mathrm{\ω}}_2)\right|,$ | W (ω ₁) | for frequency is ω ₁the spectral amplitude of seismic wavelet, | W (ω ₂) | for frequency is ω ₂the spectral amplitude of seismic wavelet, | W (ω ₁+ ω ₂) | for frequency is ω ₁+ ω ₂the spectral amplitude of seismic wavelet.

Alternatively, use equation below obtains the phase spectrum of seismic wavelet: wherein, φ is the vector that the phase spectrum by seismic wavelet forms, for the vector that the two phase spectrums of composing by seismologic record form, H is the matrix of coefficients obtaining according to the phase spectrums of two spectrums of seismologic record.

Alternatively, vector element be the natural logarithm of the frequency of two seismic wavelets of the two spectrums sampling corresponding phase spectrum in two spectrums of seismologic record in scheduled frequency range.

Alternatively, described scheduled frequency range is [N ₁hZ, N ₂hZ], sampling interval is MHZ, N ₁for being less than N ₂positive number, M is positive count, the sampling of the frequency of described two seismic wavelets in described scheduled frequency range is represented as respectively ω ₃and ω ₄, wherein, ω ₃and ω ₄some or all two spectrums that are combined in seismologic record in the natural logarithm of corresponding phase spectrum as vector element.

Alternatively, ω ₃and ω ₄value be expressed as equation below: ω ₃(x)=N ₁+ (x-1) M; ω ₄(y)=ω ₃(x)+(y-1) M wherein, ω ₃(x) represent ω ₃x value, x is positive integer, ω ₄(y) represent ω ₄y value, y is positive integer, y ∈ $[1,\frac{N_2+N_1+M-2{\mathrm{\ω}}_3\left(x\right)}{2M}],$ And $\frac{N_2-N_1+M}{M}$ For can be by 2 positive integers that divide exactly.

Alternatively, the line number of matrix of coefficients H is the columns of matrix of coefficients H is the element h of the capable k row of any l in matrix of coefficients H _l,kexpression will with vector in the frequencies omega of two seismic wavelets corresponding to l element ₃and ω ₄in the time of in the formula of the phase spectrum of the two spectrums of substitution, Φ (N in formula ₁+ (k-1) M) coefficient, wherein, l is for being less than or equal to positive integer, k is for being less than or equal to positive integer, wherein, Φ (N ₁+ (k-1) M) expression frequency is N ₁+ (k-1) phase spectrum of the seismic wavelet of M.

Alternatively, described scheduled frequency range is [1HZ, NHZ], and N is can be by 4 positive integers that divide exactly.

Alternatively, reflection coefficient acquiring unit carries out sparse inversion by the seismic wavelet of time domain, obtains the sparse reflection coefficient on the stratum of time domain.

Alternatively, frequency division section acquiring unit comprises: dominant frequency selector switch, the described predetermined dominant frequency of selection Yu Shi wavelet; Frequency selector, the initial frequency of the Yu Shi wavelet based on selected described predetermined dominant frequency select time territory and termination frequency; Fourier transformer, by selecting the Yu Shi wavelet of the time domain of initial frequency and termination frequency, carry out Fourier transform, obtain the Yu Shi wavelet of frequency field, and the sparse reflection coefficient on the stratum of time domain is carried out to Fourier transform, obtain the sparse reflection coefficient on the stratum of frequency field; Frequency division section getter, carries out convolution by the sparse reflection coefficient on the stratum of the Yu Shi wavelet of frequency field and frequency field, obtains the full stratum earthquake frequency division section of predetermined resolution.

Alternatively, two spectrum acquiring units comprise: seismic data acquisition device, obtains skew stack geological data according to seismologic record; Time window divide device, by the skew geological data window while being divided at least one that superposes; Two spectrum getters, from described at least one time window while selecting one of them window, two spectrums of the seismologic record in the time of described in the skew stack seismic data acquisition when selected in window in window.

Alternatively, described predetermined resolution is directly proportional to described predetermined dominant frequency.

In the method for obtaining full stratum earthquake frequency division section according to the present invention and device, can obtain the seismologic record section with full formation information according to seismologic record, thereby can improve the accuracy that frequency division seismic sequence is analyzed, and, can dominant frequency be set according to the needs of frequency division seismic sequence analysis, thereby can obtain the full stratum frequency division seismic section with corresponding resolution.

Accompanying drawing explanation

By the detailed description of carrying out below in conjunction with accompanying drawing, above and other objects of the present invention, feature and advantage will become apparent, wherein:

Fig. 1 illustrates according to the process flow diagram of the method for the full stratum earthquake of obtaining of exemplary embodiment of the present invention frequency division section.

Fig. 2 illustrates the example of the full stratum earthquake frequency division section obtaining while being 10HZ according to the predetermined dominant frequency of exemplary embodiment of the present invention.

Fig. 3 illustrates the example of the full stratum earthquake frequency division section obtaining while being 20HZ according to the predetermined dominant frequency of exemplary embodiment of the present invention.

Fig. 4 illustrates the example of the full stratum earthquake frequency division section obtaining while being 40HZ according to the predetermined dominant frequency of exemplary embodiment of the present invention.

Fig. 5 illustrates the example of the full stratum earthquake frequency division section obtaining while being 80HZ according to the predetermined dominant frequency of exemplary embodiment of the present invention.

Fig. 6 illustrates according to the block diagram of the device of the full stratum earthquake of obtaining of exemplary embodiment of the present invention frequency division section.

Embodiment

Now will be in detail with reference to embodiments of the invention, the example of described embodiment is shown in the drawings, and wherein, identical label refers to identical parts all the time.

Fig. 1 illustrates according to the process flow diagram of the method for the full stratum earthquake of obtaining of exemplary embodiment of the present invention frequency division section.

At step S10, according to seismologic record, obtain two spectrums of seismologic record.

The present invention does not limit obtaining the concrete mode of two spectrums, can obtain according to various two spectrum extractive techniques two spectrums of seismologic record.

In one example, can obtain skew stack geological data (that is, obtaining according to seismologic record the original earthquake frequency division section that precision is not high) according to seismologic record; Window when skew stack geological data is divided into at least one; From described at least one time window while selecting one of them window, two spectrums of the seismologic record in the time of described in the skew stack seismic data acquisition when selected in window in window.

Here, in two spectrums of seismologic record, retained the phase information of seismic wavelet, and two spectrums of seismologic record can represent with the spectral amplitude of two spectrums of seismologic record and the phase spectrum of two spectrums of seismologic record in frequency field.

At step S20, according to two spectrums of seismologic record, obtain seismic wavelet.

As example, the step of obtaining seismic wavelet according to two spectrums of seismologic record can comprise: obtain the spectral amplitude of seismic wavelet step, obtain seismic wavelet phase spectrum step and obtain the step of the seismic wavelet of time domain.

Specifically, the step of obtaining the spectral amplitude of seismic wavelet can comprise: the spectral amplitude that obtains seismic wavelet according to the spectral amplitudes of two spectrums of seismologic record.

Specifically, can obtain by equation (1) below the spectral amplitude of seismic wavelet:

$\tilde{w}={\left(A^{T}A\right)}^{-1}{A}^T\tilde{b}---\left(1\right)$

Wherein, the vector forming for the natural logarithm of the spectral amplitude by seismic wavelet, for the vector that the natural logarithm of the spectral amplitudes of the two spectrums by seismologic record forms, A is matrix of coefficients.

First by equation (1), obtain the expression formula of natural logarithm of the spectral amplitude of seismic wavelet, then the natural logarithm of the spectral amplitude of the seismic wavelet obtaining is taken to so (e) index, obtain the spectral amplitude of seismic wavelet.

Here, vector element can be the natural logarithm of the frequency of two seismic wavelets of the two spectrums sampling corresponding spectral amplitude in two spectrums of seismologic record in scheduled frequency range.

Suppose: described scheduled frequency range is [N ₁hZ, N ₂hZ], sampling interval is all M HZ, the sampling of the frequency of two seismic wavelets is represented as respectively ω ₁and ω ₂, can obtain a plurality of ω ₁with a plurality of ω ₂, the ω of acquisition ₁with the ω obtaining ₂some or all two spectrums that are combined in seismologic record in the natural logarithm of corresponding spectral amplitude can be used as vector element, N ₁for being less than N ₂positive number, M is positive count.

In the preferred embodiment that the frequency of two seismic wavelets is sampled, ω ₁and ω ₂value can be expressed as equation (2) below:

ω ₁(n)＝N ₁+(n-1)M；ω ₂(m)＝ω ₁(n)+(m-1)M (2)

Wherein, ω ₁(n) represent ω ₁n value, n is positive integer, ω ₂(m) represent ω ₂m value, m is positive integer, $m\∈[1,\frac{N_2+{3N}_1+M-{4\mathrm{\ω}}_1\left(n\right)}{4M}],$ And for can be by 4 positive integers that divide exactly.

The ω obtaining by the preferred embodiment ₁and ω ₂all two spectrums that are combined in seismologic record in the natural logarithm of corresponding spectral amplitude can be used as vector element.

Preferably, described scheduled frequency range is [1HZ, N HZ], and N is can be by 4 positive integers that divide exactly.

In addition,, in equation (1), the line number of coefficient matrices A is the columns of coefficient matrices A is the element a of the capable j row of any i in coefficient matrices A _{i, j}expression will with vector in i the ω that element is corresponding ₁and ω ₂the natural logarithm of the spectral amplitude of the two spectrums of substitution formula in time, in formula coefficient, here, i is for being less than or equal to positive integer, j is for being less than or equal to positive integer.

Here, the natural logarithm of the spectral amplitude of two spectrums can represent with equation (3) below:

${\tilde{B}}_s({\mathrm{\ω}}_1,{\mathrm{\ω}}_2)=\tilde{W}\left({\mathrm{\ω}}_1\right)+\tilde{W}\left({\mathrm{\ω}}_2\right)+\tilde{W}({\mathrm{\ω}}_1+{\mathrm{\ω}}_2)---\left(3\right)$

Wherein, ${\tilde{B}}_s({\mathrm{\ω}}_1,{\mathrm{\ω}}_2)=\ln \left|B_s({\mathrm{\ω}}_1,{\mathrm{\ω}}_2)\right|,\tilde{W}\left({\mathrm{\ω}}_1\right)=\ln \left|W\left({\mathrm{\ω}}_1\right)\right|,\tilde{W}\left({\mathrm{\ω}}_2\right)=\ln \left|W\left({\mathrm{\ω}}_2\right)\right|,$ $\tilde{W}({\mathrm{\ω}}_1+{\mathrm{\ω}}_2)=\ln \left|W({\mathrm{\ω}}_1+{\mathrm{\ω}}_2)\right|,$ | B _s(ω ₁, ω ₂) | be two spectral amplitudes of composing of seismologic record, | W (ω ₁) | for frequency is ω ₁the spectral amplitude of seismic wavelet, | W (ω ₂) | for frequency is ω ₂the spectral amplitude of seismic wavelet, | W (ω ₁+ ω ₂) | for frequency is ω ₁+ ω ₂the spectral amplitude of seismic wavelet.

In other words, the natural logarithm of the spectral amplitude of two spectrums formula can be expressed as:

$\tilde{W}\left({\mathrm{\ω}}_1\right)+\tilde{W}\left({\mathrm{\ω}}_2\right)+\tilde{W}({\mathrm{\ω}}_1+{\mathrm{\ω}}_2)---\left(4\right)$

The step of obtaining the phase spectrum of seismic wavelet can comprise: the phase spectrum that obtains seismic wavelet according to the phase spectrum of two spectrums of seismologic record.

Specifically, can obtain by equation (5) below the phase spectrum of seismic wavelet:

Wherein, φ is the vector that the phase spectrum by seismic wavelet forms, for the vector that the two phase spectrums of composing by seismologic record form, H is matrix of coefficients.

By equation (5), can obtain the phase spectrum of seismic wavelet.

Here, vector element can be the natural logarithm of the frequency of two seismic wavelets of the two spectrums sampling corresponding phase spectrum in two spectrums of seismologic record in scheduled frequency range.

Suppose: described scheduled frequency range is [N ₁hZ, N ₂hZ], sampling interval is all M HZ, the sampling of the frequency of two seismic wavelets is represented as respectively ω ₃and ω ₄, can obtain a plurality of ω ₃with a plurality of ω ₄, the ω of acquisition ₃with the ω obtaining ₄some or all two spectrums that are combined in seismologic record in the natural logarithm of corresponding phase spectrum can be used as vector element, N ₁for being less than N ₂positive number, M is positive count.

In the preferred embodiment that the frequency of two seismic wavelets is sampled, ω ₃and ω ₄value can be expressed as equation (6) below:

ω ₃(x)＝N ₁+(x-1)M；ω ₄(y)＝ω ₃(x)+(y-1)M (6)

Wherein, ω ₃(x) represent ω ₃x value, x is positive integer, ω ₄(y) represent ω ₄y value, y is positive integer, $y\∈[1,\frac{N_2+N_1+M-{2\mathrm{\ω}}_3\left(x\right)}{2M}],$ And $\frac{N_2-N_1+M}{M}$ For can be by 2 positive integers that divide exactly.

The ω obtaining by the preferred embodiment ₃and ω ₄all two spectrums that are combined in seismologic record in corresponding phase spectrum can be used as vector element.

Preferably, described scheduled frequency range is [1HZ, N HZ], and N is can be by 4 positive integers that divide exactly.

In addition,, in equation (5), the line number of matrix of coefficients H is the columns of matrix of coefficients H is the element h of the capable k row of any l in matrix of coefficients H _{l, k}expression will with vector in l the ω that element is corresponding ₃and ω ₄phase spectrum ψ (the ω of the two spectrums of substitution ₃, ω ₄) formula in time, φ (N in formula ₁+ (k-1) M) coefficient, here, l is for being less than or equal to positive integer, k is for being less than or equal to positive integer, here, φ (N ₁+ (k-1) M) expression frequency is N ₁+ (k-1) phase spectrum of the seismic wavelet of M.

The step that obtains the seismic wavelet of time domain can comprise: the phase spectrum of the spectral amplitude of seismic wavelet and seismic wavelet is carried out to inversefouriertransform, obtain the seismic wavelet of time domain.

As example, the phase spectrum of the spectral amplitude of the seismic wavelet being obtained by equation (1) and the seismic wavelet being obtained by equation (5) can be carried out to inversefouriertransform, obtain the seismic wavelet of time domain.

At step S30, according to seismic wavelet, obtain the sparse reflection coefficient on stratum.

When obtaining at step S20 after the seismic wavelet of time domain, the seismic wavelet of time domain can be carried out to sparse inversion, thereby obtain the sparse reflection coefficient on the stratum of time domain.

At step S40, according to the Yu Shi wavelet of the sparse reflection coefficient on stratum and predetermined dominant frequency, obtain the full stratum earthquake frequency division section of predetermined resolution.

Specifically, the Yu Shi wavelet of the sparse reflection coefficient on stratum and predetermined dominant frequency is carried out to convolution, can obtain the full stratum earthquake frequency division section of predetermined resolution.Here, Yu Shi wavelet has the advantages that waveform is simple, main lobe is narrow, secondary lobe is little, resolution is high and fidelity is high, and all phase informations that comprised seismic wavelet in the sparse reflection coefficient on stratum, therefore, the Yu Shi wavelet of the sparse reflection coefficient on stratum and predetermined dominant frequency is carried out to convolution and can obtain the full stratum earthquake frequency division section that comprises full formation information.

As example, step S40 can comprise: select predetermined dominant frequency step, select initial frequency and stop the step of frequency, the step of Fourier transform and obtain the step of full stratum earthquake frequency division section.

Specifically, select the step of predetermined dominant frequency to comprise: to select the described predetermined dominant frequency with the corresponding Yu Shi wavelet of described predetermined resolution.For example, when needs obtain the full stratum earthquake frequency division section of thicker lithologic character stratum (that is, resolution is lower), can select lower predetermined dominant frequency; During the full stratum earthquake frequency division section of sedimentary formation when needs obtain thinner grade (, resolution is higher), can select higher predetermined dominant frequency.

Select the step of initial frequency and termination frequency to comprise: the initial frequency of the Yu Shi wavelet based on selected described predetermined dominant frequency select time territory and termination frequency.Specifically, described predetermined dominant frequency is in the frequency separation being comprised of initial frequency and termination frequency.Preferably, described predetermined dominant frequency is initial frequency and the intermediate value that stops frequency.For example, when the predetermined dominant frequency of selecting is 10HZ, initial frequency can be elected as to 5HZ, stop frequency and elect 15HZ (should be appreciated that, initial frequency and to stop frequency can be also other frequencies, for example, initial frequency is elected 3HZ as, stops frequency and elects 17HZ as) as.

The step of Fourier transform can comprise: will select the Yu Shi wavelet of the time domain of initial frequency and termination frequency to carry out Fourier transform, obtain the Yu Shi wavelet of frequency field, and the sparse reflection coefficient on the stratum of time domain is carried out to Fourier transform, obtain the sparse reflection coefficient on the stratum of frequency field.

The step of obtaining full stratum earthquake frequency division section can comprise: the sparse reflection coefficient on the stratum of the Yu Shi wavelet of frequency field and frequency field is carried out to convolution, obtain the full stratum earthquake frequency division section of predetermined resolution.

For example, can obtain by equation (7) below full stratum earthquake frequency division section:

$X\left(\mathrm{\ω}\right)=\stackrel{\‾}{W}\left(\mathrm{\ω}\right)*R\left(\mathrm{\ω}\right)---\left(7\right)$

Wherein, X (ω) is the frequency spectrum of full stratum earthquake frequency division section, for the Yu Shi wavelet in frequency field, the sparse reflection coefficient that R (ω) is stratum, " * " represents convolution.Because the frequency of the sparse reflection coefficient in equation (7) is fixed, therefore can change the dominant frequency of X (ω) by changing the dominant frequency of Yu Shi wavelet, thereby change the resolution of full stratum earthquake frequency division section.

Fig. 2 illustrates the example of the full stratum earthquake frequency division section obtaining while being 10HZ according to the described predetermined dominant frequency of exemplary embodiment of the present invention.

Fig. 3 illustrates the example of the full stratum earthquake frequency division section obtaining while being 20HZ according to the described predetermined dominant frequency of exemplary embodiment of the present invention.

Fig. 4 illustrates the example of the full stratum earthquake frequency division section obtaining while being 40HZ according to the described predetermined dominant frequency of exemplary embodiment of the present invention.

Fig. 5 illustrates the example of the full stratum earthquake frequency division section obtaining while being 80HZ according to the described predetermined dominant frequency of exemplary embodiment of the present invention.

By Fig. 2 to Fig. 5, can be found out, described predetermined resolution is directly proportional to described predetermined dominant frequency.When the predetermined dominant frequency of selecting is higher, the resolution of the full stratum earthquake frequency division section of acquisition is higher, thereby is more convenient to stratum to carry out meticulous Sequence Analysis.

In the method for obtaining full stratum earthquake frequency division section according to the present invention, can obtain the earthquake frequency division section with full formation information according to seismologic record, thereby can improve the accuracy that frequency division seismic sequence is analyzed, and, can dominant frequency be set according to the needs of frequency division seismic sequence analysis, thereby can obtain the full stratum frequency division seismic section with corresponding resolution.

Fig. 6 illustrates according to the block diagram of the device of the full stratum earthquake of obtaining of exemplary embodiment of the present invention frequency division section.

As shown in Figure 6, according to the device of the full stratum earthquake of obtaining of exemplary embodiment of the present invention frequency division section, comprise: two spectrum acquiring units 10, wavelet acquiring unit 20, reflection coefficient acquiring unit 30, frequency division section acquiring unit 40.

Specifically, two spectrum acquiring units 10 are for obtaining two spectrums of seismologic record according to seismologic record.

The present invention does not limit obtaining the concrete mode of two spectrums, can obtain according to various two spectrum extractive techniques two spectrums of seismologic record.

In one example, two spectrum acquiring units 10 can comprise: seismic data acquisition device, time window divide device and two spectrum getter.

Specifically, seismic data acquisition device is for obtaining skew stack geological data (that is, obtaining according to seismologic record the original earthquake frequency division section that precision is not high) according to seismologic record.

Time window window when dividing device and being divided at least one for geological data that skew is superposeed, and from described at least one time window while selecting one of them window.

Two spectrums of two spectrum getters seismologic record in window for described in the skew stack seismic data acquisition of window when selected time.

Here, in two spectrums of seismologic record, retained the phase information of seismic wavelet, and two spectrums of seismologic record can represent with the spectral amplitude of two spectrums of seismologic record and the phase spectrum of two spectrums of seismologic record in frequency field.

Wavelet acquiring unit 20 obtains seismic wavelet for the two spectrums according to seismologic record.

As example, wavelet acquiring unit 20 can comprise: wavelet amplitude getter, sub-wave phase spectrum getter and wavelet getter.

Specifically, wavelet amplitude getter is for obtaining the spectral amplitude of seismic wavelet according to the spectral amplitude of two spectrums of seismologic record.

Specifically, can obtain by equation (1) spectral amplitude of seismic wavelet.

Owing to have been described in detail obtain the method for the spectral amplitude of seismic wavelet by equation (1) with reference to Fig. 1 above, do not repeat them here.

Sub-wave phase spectrum getter is for obtaining the phase spectrum of seismic wavelet according to the phase spectrum of two spectrums of seismologic record.

Specifically, can obtain by equation (5) phase spectrum of seismic wavelet.

Owing to have been described in detail obtain the method for the phase spectrum of seismic wavelet by equation (5) with reference to Fig. 1 above, do not repeat them here.

Wavelet getter, for the phase spectrum of the spectral amplitude of seismic wavelet and seismic wavelet is carried out to inversefouriertransform, obtains the seismic wavelet of time domain.

As example, the phase spectrum of the spectral amplitude of the seismic wavelet being obtained by equation (1) and the seismic wavelet being obtained by equation (5) can be carried out to inversefouriertransform, obtain the seismic wavelet of time domain.

Reflection coefficient acquiring unit 30 is for obtaining the sparse reflection coefficient on stratum according to seismic wavelet.

As example, when wavelet acquiring unit 20 obtains after the seismic wavelet of time domain, the seismic wavelet of time domain can be carried out to sparse inversion, thereby obtain the sparse reflection coefficient on the stratum of time domain.

Frequency division section acquiring unit 40 is for obtaining the full stratum earthquake frequency division section of predetermined resolution according to the Yu Shi wavelet of the sparse reflection coefficient on stratum and predetermined dominant frequency.

Specifically, the Yu Shi wavelet of the sparse reflection coefficient on stratum and predetermined dominant frequency is carried out to convolution, can obtain the full stratum earthquake frequency division section of predetermined resolution.Here, Yu Shi wavelet has the advantages that waveform is simple, main lobe is narrow, secondary lobe is little, resolution is high and fidelity is high, and all phase informations that comprised seismic wavelet in the sparse reflection coefficient on stratum, therefore, the Yu Shi wavelet of the sparse reflection coefficient on stratum and predetermined dominant frequency is carried out to convolution and can obtain the full stratum earthquake frequency division section that comprises full formation information.

As example, frequency division section acquiring unit can comprise: dominant frequency selector switch, frequency selector, Fourier transformer and frequency division section getter.

Specifically, dominant frequency selector switch is for selecting the described predetermined dominant frequency with the corresponding Yu Shi wavelet of described predetermined resolution.For example, when needs obtain the full stratum earthquake frequency division section of thicker lithologic character stratum (that is, resolution is lower), can select lower predetermined dominant frequency; During the full stratum earthquake frequency division section of sedimentary formation when needs obtain thinner grade (, resolution is higher), can select higher predetermined dominant frequency.

Frequency selector is for initial frequency and the termination frequency of the Yu Shi wavelet based on selected described predetermined dominant frequency select time territory.Specifically, described predetermined dominant frequency is in the frequency separation being comprised of initial frequency and termination frequency.Preferably, described predetermined dominant frequency is initial frequency and the intermediate value that stops frequency.For example, when the predetermined dominant frequency of selecting is 10HZ, initial frequency can be elected as to 5HZ, stop frequency and elect 15HZ (should be appreciated that, initial frequency and to stop frequency can be also other frequencies, for example, initial frequency is elected 3HZ as, stops frequency and elects 17HZ as) as.

Fourier transformer is for will selecting the Yu Shi wavelet of the time domain of initial frequency and termination frequency to carry out Fourier transform, obtain the Yu Shi wavelet of frequency field, and the sparse reflection coefficient on the stratum of time domain is carried out to Fourier transform, obtain the sparse reflection coefficient on the stratum of frequency field.

Frequency division section getter, for the sparse reflection coefficient on the stratum of the Yu Shi wavelet of frequency field and frequency field is carried out to convolution, obtains the full stratum earthquake frequency division section of predetermined resolution.

For example, can obtain full stratum earthquake frequency division section by the above-mentioned equation of mentioning (7).

Owing to having described the example of the full stratum earthquake frequency division section about obtaining with reference to Fig. 2 to Fig. 5, repeat no more.

In the device that obtains full stratum earthquake frequency division section according to the present invention, can obtain the earthquake frequency division section with full formation information according to seismologic record, thereby can improve the accuracy that frequency division seismic sequence is analyzed, and, can dominant frequency be set according to the needs of frequency division seismic sequence analysis, thereby can obtain the full stratum frequency division seismic section with corresponding resolution.

In addition, according to the said method of exemplary embodiment of the present invention, may be implemented as computer program, thereby when this program of operation, realize said method.

In addition, according to the unit in the said apparatus of exemplary embodiment of the present invention, can be implemented nextport hardware component NextPort.Those skilled in the art, according to the performed processing of unit limiting, can for example use field programmable gate array (FPGA) or special IC (ASIC) to realize unit.

Although specifically shown with reference to its exemplary embodiment and described the present invention, but it should be appreciated by those skilled in the art, in the situation that do not depart from the spirit and scope of the present invention that claim limits, can carry out the various changes in form and details to it.

Claims (36)

1. a method of obtaining full stratum earthquake frequency division section, comprising:

(a) according to seismologic record, obtain two spectrums of seismologic record;

(b) according to two spectrums of seismologic record, obtain seismic wavelet;

(c) according to seismic wavelet, obtain the sparse reflection coefficient on stratum;

(d) according to the Yu Shi wavelet of the sparse reflection coefficient on stratum and predetermined dominant frequency, obtain the full stratum earthquake frequency division section of predetermined resolution.

2. the method as described in right request 1, wherein, step (b) comprising:

(b1) according to the spectral amplitude of two spectrums of seismologic record, obtain the spectral amplitude of seismic wavelet;

(b2) according to the phase spectrum of two spectrums of seismologic record, obtain the phase spectrum of seismic wavelet;

(b3) phase spectrum of the spectral amplitude of seismic wavelet and seismic wavelet is carried out to inversefouriertransform, obtain the seismic wavelet of time domain.

3. method as claimed in claim 2, wherein, step (b1) comprising:

By equation below, obtained the spectral amplitude of seismic wavelet:

$\tilde{w}={\left(A^{T}A\right)}^{-1}{A}^T\tilde{b}$

Wherein, the vector forming for the natural logarithm of the spectral amplitude by seismic wavelet, for the vector that the natural logarithm of the spectral amplitudes of the two spectrums by seismologic record forms, A be according to the matrix of coefficients of the natural logarithm acquisition of two spectral amplitudes of composing of seismologic record.

4. method as claimed in claim 3, wherein, vector element be the natural logarithm of the frequency of two seismic wavelets of the two spectrums sampling corresponding spectral amplitude in two spectrums of seismologic record in scheduled frequency range.

5. method as claimed in claim 4, wherein, described scheduled frequency range is [N ₁hZ, N ₂hZ], sampling interval is M HZ, N ₁for being less than N ₂positive number, M is positive count, the sampling of the frequency of described two seismic wavelets in described scheduled frequency range is represented as respectively ω ₁and ω ₂,

Wherein, ω ₁with ω ₂some or all two spectrums that are combined in seismologic record in the natural logarithm of corresponding spectral amplitude as vector element.

6. method as claimed in claim 5, wherein, ω ₁and ω ₂value be expressed as equation below:

ω ₁(n)＝N ₁+(n-1)M；ω ₂(m)＝ω ₁(n)+(m-1)M

Wherein, ω ₁(n) represent ω ₁n value, n is positive integer, ω ₂(m) represent ω ₂m value, m is positive integer, $m\∈[1,\frac{N_2+{3N}_1+M-{4\mathrm{\ω}}_1\left(n\right)}{4M}],$ And $\frac{N_2-N_1+M}{M}$ For can be by 4 positive integers that divide exactly.

7. method as claimed in claim 6, wherein, the line number of coefficient matrices A is the columns of coefficient matrices A is the element a of the capable j row of any i in coefficient matrices A _{i, j}expression will with vector in the frequencies omega of two seismic wavelets corresponding to i element ₁and ω ₂in the time of in the formula of the natural logarithm of the spectral amplitude of the two spectrums of substitution, in formula coefficient,

Wherein, i is for being less than or equal to positive integer, j is for being less than or equal to positive integer,

Wherein, expression frequency is N ₁+ (j-1) natural logarithm of the spectral amplitude of the seismic wavelet of M.

8. method as claimed in claim 3, wherein, the formula of the natural logarithm of the spectral amplitude of two spectrums is:

$\tilde{W}\left({\mathrm{\ω}}_1\right)+\tilde{W}\left({\mathrm{\ω}}_2\right)+\tilde{W}({\mathrm{\ω}}_1+{\mathrm{\ω}}_2)$

Wherein, $\tilde{W}\left({\mathrm{\ω}}_1\right)=\ln \left|W\left({\mathrm{\ω}}_1\right)\right|,\tilde{W}\left({\mathrm{\ω}}_2\right)=\ln \left|W\left({\mathrm{\ω}}_2\right)\right|,\tilde{W}({\mathrm{\ω}}_1+{\mathrm{\ω}}_2)=\ln \left|W({\mathrm{\ω}}_1+{\mathrm{\ω}}_2)\right|,$ | W (ω ₁) | for frequency is ω ₁the spectral amplitude of seismic wavelet, | W (ω ₂) | for frequency is ω ₂the spectral amplitude of seismic wavelet, | W (ω ₁+ ω ₂) | for frequency is ω ₁+ ω ₂the spectral amplitude of seismic wavelet.

9. method as claimed in claim 2, wherein, step (b2) comprising:

By equation below, obtained the phase spectrum of seismic wavelet:

Wherein, φ is the vector that the phase spectrum by seismic wavelet forms, for the vector that the two phase spectrums of composing by seismologic record form, H is the matrix of coefficients obtaining according to the phase spectrums of two spectrums of seismologic record.

10. method as claimed in claim 9, wherein, vector element be the natural logarithm of the frequency of two seismic wavelets of the two spectrums sampling corresponding phase spectrum in two spectrums of seismologic record in scheduled frequency range.

11. methods as claimed in claim 10, wherein, described scheduled frequency range is [N ₁hZ, N ₂hZ], sampling interval is M HZ, N ₁for being less than N ₂positive number, M is positive count, the sampling of the frequency of described two seismic wavelets in described scheduled frequency range is represented as respectively ω ₃and ω ₄,

Wherein, ω ₃and ω ₄some or all two spectrums that are combined in seismologic record in the natural logarithm of corresponding phase spectrum as vector element.

12. methods as claimed in claim 11, wherein, ω ₃and ω ₄value be expressed as equation below:

ω ₃(x)＝N ₁+(x-1)M；ω ₄(y)＝ω ₃(x)+(y-1)M

Wherein, ω ₃(x) represent ω ₃x value, x is positive integer, ω ₄(y) represent ω ₄y value, y is positive integer, $y\∈[1,\frac{N_2+N_1+M-{2\mathrm{\ω}}_3\left(x\right)}{2M}],$ And $\frac{N_2-N_1+M}{M}$ For can be by 2 positive integers that divide exactly.

13. methods as claimed in claim 12, wherein, the line number of matrix of coefficients H is the columns of matrix of coefficients H is the element h of the capable k row of any l in matrix of coefficients H _{l, k}expression will with vector in the frequencies omega of two seismic wavelets corresponding to l element ₃and ω ₄in the time of in the formula of the phase spectrum of the two spectrums of substitution, Φ (N in formula ₁+ (k-1) M) coefficient,

Wherein, l is for being less than or equal to positive integer, k is for being less than or equal to positive integer,

Wherein, Φ (N ₁+ (k-1) M) expression frequency is N ₁+ (k-1) phase spectrum of the seismic wavelet of M.

14. methods as described in claim 5 or 11, wherein, described scheduled frequency range is [1HZ, NHZ], N is for can be by 4 positive integers that divide exactly.

15. methods as described in right request 2, wherein, step (c) comprising: the seismic wavelet of time domain is carried out to sparse inversion, obtain the sparse reflection coefficient on the stratum of time domain.

16. methods as described in right request 15, wherein, step (d) comprising:

(d1) select the described predetermined dominant frequency of Yu Shi wavelet;

(d2) initial frequency of the Yu Shi wavelet based on selected described predetermined dominant frequency select time territory and termination frequency;

(d3) will select the Yu Shi wavelet of the time domain of initial frequency and termination frequency to carry out Fourier transform, obtain the Yu Shi wavelet of frequency field, and the sparse reflection coefficient on the stratum of time domain is carried out to Fourier transform, obtain the sparse reflection coefficient on the stratum of frequency field;

(d4) the sparse reflection coefficient on the stratum of the Yu Shi wavelet of frequency field and frequency field is carried out to convolution, obtain the full stratum earthquake frequency division section of predetermined resolution.

17. the method for claim 1, wherein step (a) comprising:

(a1) according to seismologic record, obtain skew stack geological data;

(a2) window while skew stack geological data being divided into at least one;

(a3) from described at least one time window while selecting one of them window, two spectrums of the seismologic record in the time of described in the skew stack seismic data acquisition when selected in window in window.

18. the method for claim 1, wherein described predetermined resolution be directly proportional to described predetermined dominant frequency.

19. 1 kinds of devices that obtain full stratum earthquake frequency division section, comprising:

Two spectrum acquiring units, obtain the two of seismologic record according to seismologic record and compose;

Wavelet acquiring unit, obtains seismic wavelet according to two spectrums of seismologic record;

Reflection coefficient acquiring unit, obtains the sparse reflection coefficient on stratum according to seismic wavelet;

Frequency division section acquiring unit, obtains the full stratum earthquake frequency division section of predetermined resolution according to the Yu Shi wavelet of the sparse reflection coefficient on stratum and predetermined dominant frequency.

20. devices as described in right request 19, wherein, wavelet acquiring unit comprises:

Wavelet amplitude getter, obtains the spectral amplitude of seismic wavelet according to the spectral amplitudes of two spectrums of seismologic record;

Sub-wave phase is composed getter, obtains the phase spectrum of seismic wavelet according to the phase spectrums of two spectrums of seismologic record;

Wavelet getter, carries out inversefouriertransform by the phase spectrum of the spectral amplitude of seismic wavelet and seismic wavelet, obtains the seismic wavelet of time domain.

21. devices as claimed in claim 20, wherein, use equation below obtains the spectral amplitude of seismic wavelet:

$\tilde{w}={\left(A^{T}A\right)}^{-1}{A}^T\tilde{b}$

Wherein, the vector forming for the natural logarithm of the spectral amplitude by seismic wavelet, for the vector that the natural logarithm of the spectral amplitudes of the two spectrums by seismologic record forms, A be according to the matrix of coefficients of the natural logarithm acquisition of two spectral amplitudes of composing of seismologic record.

22. devices as claimed in claim 21, wherein, vector element be the natural logarithm of the frequency of two seismic wavelets of the two spectrums sampling corresponding spectral amplitude in two spectrums of seismologic record in scheduled frequency range.

23. devices as claimed in claim 22, wherein, described scheduled frequency range is [N ₁hZ, N ₂hZ], sampling interval is M HZ, N ₁for being less than N ₂positive number, M is positive count, the sampling of the frequency of described two seismic wavelets in described scheduled frequency range is represented as respectively ω ₁and ω ₂,

Wherein, ω ₁with ω ₂some or all two spectrums that are combined in seismologic record in the natural logarithm of corresponding spectral amplitude as vector element.

24. devices as claimed in claim 23, wherein, ω ₁and ω ₂value be expressed as equation below:

ω ₁(n)＝N ₁+(n-1)M；ω ₂(m)＝ω ₁(n)+(m-1)M

Wherein, ω ₁(n) represent ω ₁n value, n is positive integer, ω ₂(m) represent ω ₂m value, m is positive integer, $m\∈[1,\frac{N_2+3N_1+M-4{\mathrm{\ω}}_1\left(n\right)}{4M}],$ And for can be by 4 positive integers that divide exactly.

25. devices as claimed in claim 24, wherein, the line number of coefficient matrices A is the columns of coefficient matrices A is the element a of the capable j row of any i in coefficient matrices A _{i, j}expression will with vector in the frequencies omega of two seismic wavelets corresponding to i element ₁and ω ₂in the time of in the formula of the natural logarithm of the spectral amplitude of the two spectrums of substitution, in formula coefficient,

Wherein, i is for being less than or equal to positive integer, j is for being less than or equal to positive integer,

Wherein, expression frequency is N ₁+ (j-1) natural logarithm of the spectral amplitude of the seismic wavelet of M.

26. devices as claimed in claim 21, wherein, the formula of the natural logarithm of the spectral amplitude of two spectrums is:

$\tilde{W}\left({\mathrm{\ω}}_1\right)+\tilde{W}\left({\mathrm{\ω}}_2\right)+\tilde{W}({\mathrm{\ω}}_1+{\mathrm{\ω}}_2)$

Wherein, $\tilde{W}\left({\mathrm{\ω}}_1\right)=\ln \left|W\left({\mathrm{\ω}}_1\right)\right|,\tilde{W}\left({\mathrm{\ω}}_2\right)=\ln \left|W\left({\mathrm{\ω}}_2\right)\right|,\tilde{W}({\mathrm{\ω}}_1+{\mathrm{\ω}}_2)=\ln \left|W({\mathrm{\ω}}_1+{\mathrm{\ω}}_2)\right|,$ | W (ω ₁) | for frequency is ω ₁the spectral amplitude of seismic wavelet, | W (ω ₂) | for frequency is ω ₂the spectral amplitude of seismic wavelet, | W (ω ₁+ ω ₂) | for frequency is ω ₁+ ω ₂the spectral amplitude of seismic wavelet.

27. devices as claimed in claim 20, wherein, use equation below obtains the phase spectrum of seismic wavelet:

Wherein, φ is the vector that the phase spectrum by seismic wavelet forms, for the vector that the two phase spectrums of composing by seismologic record form, H is the matrix of coefficients obtaining according to the phase spectrums of two spectrums of seismologic record.

28. devices as claimed in claim 27, wherein, vector element be the natural logarithm of the frequency of two seismic wavelets of the two spectrums sampling corresponding phase spectrum in two spectrums of seismologic record in scheduled frequency range.

29. devices as claimed in claim 28, wherein, described scheduled frequency range is [N ₁hZ, N ₂hZ], sampling interval is M HZ, N ₁for being less than N ₂positive number, M is positive count, the sampling of the frequency of described two seismic wavelets in described scheduled frequency range is represented as respectively ω ₃and ω ₄,

Wherein, ω ₃and ω ₄some or all two spectrums that are combined in seismologic record in the natural logarithm of corresponding phase spectrum as vector element.

30. devices as claimed in claim 29, wherein, ω ₃and ω ₄value be expressed as equation below:

ω ₃(x)＝N ₁+(x-1)M；ω ₄(y)＝ω ₃(x)+(y-1)M

Wherein, ω ₃(x) represent ω ₃x value, x is positive integer, ω ₄(y) represent ω ₄y value, y is positive integer, $y\∈[1,\frac{N_2+N_1+M-2{\mathrm{\ω}}_3\left(x\right)}{2M}],$ And $\frac{N_2-N_1+M}{M}$ For can be by 2 positive integers that divide exactly.

31. devices as claimed in claim 30, wherein, the line number of matrix of coefficients H is the columns of matrix of coefficients H is the element h of the capable k row of any l in matrix of coefficients H _{l, k}expression will with vector in the frequencies omega of two seismic wavelets corresponding to l element ₃and ω ₄in the time of in the formula of the phase spectrum of the two spectrums of substitution, Φ (N in formula ₁+ (k-1) M) coefficient,

Wherein, l is for being less than or equal to positive integer, k is for being less than or equal to positive integer,

Wherein, Φ (N ₁+ (k-1) M) expression frequency is N ₁+ (k-1) phase spectrum of the seismic wavelet of M.

32. devices as described in claim 23 or 29, wherein, described scheduled frequency range is [1HZ, NHZ], N is for can be by 4 positive integers that divide exactly.

33. devices as described in right request 20, wherein, reflection coefficient acquiring unit carries out sparse inversion by the seismic wavelet of time domain, obtains the sparse reflection coefficient on the stratum of time domain.

34. devices as described in right request 33, wherein, frequency division section acquiring unit comprises:

Dominant frequency selector switch, the described predetermined dominant frequency of selection Yu Shi wavelet;

Frequency selector, the initial frequency of the Yu Shi wavelet based on selected described predetermined dominant frequency select time territory and termination frequency;

Fourier transformer, by selecting the Yu Shi wavelet of the time domain of initial frequency and termination frequency, carry out Fourier transform, obtain the Yu Shi wavelet of frequency field, and the sparse reflection coefficient on the stratum of time domain is carried out to Fourier transform, obtain the sparse reflection coefficient on the stratum of frequency field;

Frequency division section getter, carries out convolution by the sparse reflection coefficient on the stratum of the Yu Shi wavelet of frequency field and frequency field, obtains the full stratum earthquake frequency division section of predetermined resolution.

35. devices as claimed in claim 19, wherein, two spectrum acquiring units comprise:

Seismic data acquisition device, obtains skew stack geological data according to seismologic record;

Time window divide device, by the skew geological data window while being divided at least one that superposes;

Two spectrum getters, from described at least one time window while selecting one of them window, two spectrums of the seismologic record in the time of described in the skew stack seismic data acquisition when selected in window in window.

36. devices as claimed in claim 19, wherein, described predetermined resolution is directly proportional to described predetermined dominant frequency.