CN113900140B - Seismic data optimization method and device based on space-time combination - Google Patents
Seismic data optimization method and device based on space-time combination Download PDFInfo
- Publication number
- CN113900140B CN113900140B CN202111157271.4A CN202111157271A CN113900140B CN 113900140 B CN113900140 B CN 113900140B CN 202111157271 A CN202111157271 A CN 202111157271A CN 113900140 B CN113900140 B CN 113900140B
- Authority
- CN
- China
- Prior art keywords
- seismic
- time
- dimensional
- channel
- space
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 49
- 238000005457 optimization Methods 0.000 title claims abstract description 32
- 238000001914 filtration Methods 0.000 claims abstract description 197
- 238000005259 measurement Methods 0.000 claims description 43
- 239000011159 matrix material Substances 0.000 claims description 27
- 238000004364 calculation method Methods 0.000 claims description 19
- 150000001875 compounds Chemical class 0.000 claims description 16
- 238000012545 processing Methods 0.000 claims description 14
- 239000011435 rock Substances 0.000 claims description 12
- 238000006243 chemical reaction Methods 0.000 claims description 10
- 230000009466 transformation Effects 0.000 claims description 9
- 238000005070 sampling Methods 0.000 claims description 7
- 230000007704 transition Effects 0.000 claims description 7
- 230000008569 process Effects 0.000 claims description 6
- 238000004590 computer program Methods 0.000 claims description 5
- 238000003384 imaging method Methods 0.000 claims description 3
- 230000005012 migration Effects 0.000 claims description 3
- 238000013508 migration Methods 0.000 claims description 3
- 230000015572 biosynthetic process Effects 0.000 claims description 2
- 238000005755 formation reaction Methods 0.000 claims description 2
- 230000000694 effects Effects 0.000 description 6
- 238000004422 calculation algorithm Methods 0.000 description 5
- 238000004321 preservation Methods 0.000 description 4
- 238000012937 correction Methods 0.000 description 3
- 238000001514 detection method Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000009499 grossing Methods 0.000 description 3
- 230000006872 improvement Effects 0.000 description 3
- 230000001629 suppression Effects 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000001035 drying Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
- G01V1/00—Seismology; Seismic or acoustic prospecting or detecting
- G01V1/28—Processing seismic data, e.g. for interpretation or for event detection
- G01V1/282—Application of seismic models, synthetic seismograms
Landscapes
- Engineering & Computer Science (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Acoustics & Sound (AREA)
- Environmental & Geological Engineering (AREA)
- Geology (AREA)
- General Life Sciences & Earth Sciences (AREA)
- General Physics & Mathematics (AREA)
- Geophysics (AREA)
- Geophysics And Detection Of Objects (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a method and a device for seismic data optimization based on space-time combination, wherein the method comprises the following steps: acquiring earthquake post-stack pure wave data and earthquake post-stack result data; establishing a three-dimensional seismic Kalman filtering model, which comprises a three-dimensional seismic transverse space Kalman filtering model and a three-dimensional seismic longitudinal space Kalman filtering model; performing three-dimensional seismic Kalman iterative filtering of a time-space combination; and (3) performing three-dimensional seismic Kalman iterative filtering for multiple time-space combinations. According to the seismic data optimization method based on the space-time combination, the resolution of the filtered seismic section is not obviously reduced, the wave group relation and the longitudinal and transverse seismic amplitude energy relation of the seismic section and the seismic data before filtering are kept unchanged, the signal-to-noise ratio of the inversion result of the filtered seismic data is obviously improved, and the transverse tracking of the geologic body is more efficient.
Description
Technical Field
The invention relates to the technical field of oil-gas exploration, in particular to a method and a device for seismic data optimization based on space-time combination, which can be used in a deep and open sea marine observation system.
Background
With the increasing development of ocean engineering to deep and far sea oceans, lithology and unconventional oil and gas reservoirs are gradually turned to by the constructed oil and gas reservoirs in the current oil and gas exploration direction, and the lithology and unconventional oil and gas reservoir reservoirs have the characteristics of thin longitudinal and fast transverse change and low signal to noise ratio. However, seismic data with high signal-to-noise ratio is the primary condition for effective exploration and evaluation of oil and gas reservoirs, and seismic filtering can recover effective information from polluted seismic signals so as to improve the signal-to-noise ratio of the seismic data and realize the improvement of the accuracy and efficiency of the seismic data.
In the seismic data digital processing stage, frequency-up processing is often required to be performed on the acquired seismic records, which can aggravate the problems that the high-frequency component of the seismic signals contains noise, the transverse continuity of the seismic data is poor, the signal-to-noise ratio is low, and the like, and further reduce the seismic structure interpretation and the reservoir inversion precision. The common frequency domain low-pass filtering considers that the noise of the seismic data is mainly concentrated in high-frequency components, the seismic signals are firstly subjected to Fourier transform to the frequency domain, and then the high-frequency components containing the noise are filtered out in the frequency domain, so that effective signals in the high-frequency components of the signals can be filtered out together, and the existing smoothing filtering technology has an unobvious noise suppression effect and is difficult to ensure the fidelity and amplitude preservation of the seismic data.
Disclosure of Invention
The invention aims to provide a method and a device for seismic data optimization based on space-time combination, which are used for solving the problems that the noise suppression effect is not obvious and the fidelity and the amplitude of seismic data are difficult to maintain in the prior art.
The invention provides a method for optimizing seismic data based on space-time combination, which comprises the following steps:
acquiring earthquake post-stack pure wave data and earthquake post-stack result data;
establishing a three-dimensional seismic Kalman filtering model, which comprises a three-dimensional seismic transverse space Kalman filtering model and a three-dimensional seismic longitudinal space Kalman filtering model;
substituting the earthquake post-stack pure wave data and the earthquake post-stack result data into the three-dimensional earthquake transverse space Kalman filtering model, respectively completing two filtering iterations of a three-dimensional earthquake transverse space, and then respectively substituting the three-dimensional earthquake transverse space Kalman filtering model for performing one filtering iteration of a three-dimensional earthquake longitudinal space, so as to respectively obtain a primary space-time combination value of the earthquake post-stack pure wave data and a primary space-time combination value of the earthquake post-stack result data;
substituting the primary time-space combination value of the pure wave data after the earthquake superposition and the primary time-space combination value of the result data after the earthquake superposition into the three-dimensional earthquake Kalman filtering model, and performing multiple iteration to respectively obtain multiple time-space combination values of the pure wave data after the earthquake superposition and multiple time-space combination values of the result data after the earthquake superposition.
Specifically, the acquiring of the seismic post-stack pure wave data and the seismic post-stack result data specifically comprises the following steps:
step A1: acquiring seismic data, and acquiring a common reflection point gather by applying a seismic migration imaging method to the acquired seismic data;
step A2: carrying out full or partial superposition on the seismic data of the common reflection point gather to obtain post-seismic superposition pure wave data;
step A3: and performing channel equalization and filtering processing on the seismic post-stack pure wave data to obtain seismic post-stack result data.
Specifically, the establishment of the three-dimensional seismic longitudinal space kalman filtering model specifically comprises the following steps:
step B11: calculating the state value of the k channel through the state conversion equation of the seismic channel on the transverse space
Step B12: calculating the measurement value of the k channel of the seismic channel through the measurement equation of the seismic channel
Step B13: computing the kth trace X of a seismic section k Corresponding standard deviation P of k-th channel of filtering k ；
Step B14: according to the state value of the k-th channel
Seismic trace kth measurement
And the k channel X of seismic section k Corresponding standard deviation P of k-th track k For the kth trace X of the seismic section k Performing optimization calculation;
wherein, the k-th channel X of the seismic section k The system of optimization calculation equations of (1) is:
in the formula (I), the compound is shown in the specification,
for the k channel X of seismic section k The optimized seismic channels are obtained by the method,
to filter the state values of the k-th channel,
representing the filtered k-th channel measurement;
Kg k kalman gain, which is a lateral filtering;
Q k the standard deviation between the kth seismic data of the seismic channel and the adjacent seismic data channel is shown, wherein the difference between the seismic adjacent channels of the sedimentary rock stratum can be approximately regarded as stable random white Gaussian noise;
std denotes taking the standard deviation, X k+1 Is k +1 trace, X in seismic section on main survey line k-1 Is the k-1 st trace, X, in the seismic section on the main survey line k-numcmp The k-numcmp trace, X, of a main survey line adjacent to the k-th trace of the seismic section k+numcmp Respectively is the k + numcmp channel of the adjacent survey line of the main survey line where the k channel of the seismic profile is located; x k A kth trace representing a seismic section;
step B15: updating the standard deviation P of the k-th channel of the seismic section corresponding to the k-th channel of the filtering k ；
Step B16: and repeating the steps B11 to B15 until the establishment of the three-dimensional seismic transverse space Kalman filtering model is completed.
Specifically, the establishment of the three-dimensional seismic longitudinal space kalman filtering model specifically comprises the following steps:
step B21: through the longitudinal directionCalculating the data state of the three-dimensional seismic data volume at the t-th moment by using the state conversion equation of the slice at the seismic sampling time point in space
Step B22: calculating the measurement value of the slice at the t moment of the earthquake by the measurement equation of the slice at the t moment of the earthquake data
Step B23: calculating seismic data S of t-time slice of three-dimensional seismic data volume t Standard deviation of (PT) t ；
Step B24: according to the t-th time data state of the three-dimensional seismic data volume
Seismic data time t slice measurement
Seismic data S sliced at time t from three-dimensional seismic data volume t Seismic data S of corresponding t-time slice t Standard deviation of (PT) t Slicing seismic data S of a three-dimensional seismic data volume at time t t Performing optimization calculation;
wherein, the seismic data S of the t-th time slice of the three-dimensional seismic data volume t The optimization calculation equation set is as follows:
in the formula (I), the compound is shown in the specification,
seismic time slicing after seismic data slicing at the t-th time of the three-dimensional seismic data volume is optimized;
KgT t kalman gain, which is longitudinal filtering;
QT t the standard deviation between the tth moment of the seismic data and the time slice of the adjacent seismic data is shown, wherein the difference between the seismic adjacent channels of the sedimentary rock stratum can be approximately regarded as stable random white Gaussian noise;
stdT denotes standard deviation, S t-1 Is a time slice of the t-1 th time of the three-dimensional seismic data volume; s. the t+1 Is a time slice of the three-dimensional seismic data volume at time t + 1; s t Time slicing is carried out on the t-th time of the three-dimensional seismic data body;
step B25: updating seismic data S of t-time slice of three-dimensional seismic data volume t Standard deviation of (PT) t ；
Step B26: and C, repeating the steps B21 to B25 until the establishment of the three-dimensional seismic longitudinal space Kalman filtering model is completed.
Specifically, the multiple time-space combination values of the seismic post-stack pure wave data and the multiple time-space combination values of the seismic post-stack result data specifically include the following steps:
step D1: substituting the primary space-time combination value of the earthquake post-stack result data and the primary space-time combination value of the earthquake post-stack result data into the three-dimensional earthquake transverse space Kalman filtering model to respectively complete two filtering iterations of the three-dimensional earthquake transverse space, and substituting the filtering iterations into the three-dimensional earthquake longitudinal space Kalman filtering model to perform primary filtering iteration of the three-dimensional earthquake longitudinal space to respectively obtain a secondary space-time combination value of the earthquake post-stack result data and a secondary space-time combination value of the earthquake post-stack result data;
step D2: substituting the secondary space-time combination value of the earthquake post-stack result data and the secondary space-time combination value of the earthquake post-stack result data into the three-dimensional earthquake transverse space Kalman filtering model to respectively complete two filtering iterations of the three-dimensional earthquake transverse space, and substituting the filtering iterations into the three-dimensional earthquake longitudinal space Kalman filtering model to perform one filtering iteration of the three-dimensional earthquake longitudinal space to respectively obtain a three-time space-time combination value of the earthquake post-stack result data and a three-time space-time combination value of the earthquake post-stack result data;
step D3: and substituting the three-time-space combination value of the earthquake post-stack result data and the three-time-space combination value of the earthquake post-stack result data into the three-dimensional earthquake transverse space Kalman filtering model to respectively complete two filtering iterations of the three-dimensional earthquake transverse space, and substituting the three-time-space combination value into the three-dimensional earthquake longitudinal space Kalman filtering model to perform one filtering iteration of the three-dimensional earthquake longitudinal space to respectively obtain the four-time-space combination value of the earthquake post-stack result data and the four-time-space combination value of the earthquake post-stack result data.
Further, the step D3 is followed by the following steps:
and making a synthetic record by using the logging sound waves and the density curve, calibrating and comparing the quartic time-space combination value of the post-seismic-stack result data with the quartic time-space combination value of the post-seismic-stack result data, determining the fidelity amplitude of the filtered seismic data, and performing seismic structure interpretation and reservoir prediction work such as horizon tracking, fault interpretation, stratum elastic parameter inversion and the like on the filtered seismic data on the basis.
The invention also relates to a device for seismic data optimization based on space-time combination, which comprises
The first processing unit is used for acquiring seismic post-stack pure wave data and seismic post-stack result data;
the second processing unit is used for establishing a three-dimensional earthquake Kalman filtering model, and the three-dimensional earthquake transverse space Kalman filtering model and the three-dimensional earthquake longitudinal space Kalman filtering model are included;
the third processing unit is used for substituting the post-seismic-stack pure wave data and the post-seismic-stack result data into the three-dimensional seismic transverse space Kalman filtering model, respectively completing two filtering iterations of a three-dimensional seismic transverse space, and then respectively substituting the two filtering iterations into the three-dimensional seismic longitudinal space Kalman filtering model to perform one filtering iteration of a three-dimensional seismic longitudinal space, so as to respectively obtain a primary space-time combination value of the post-seismic-stack pure wave data and a primary space-time combination value of the post-seismic-stack result data;
and the fourth processing unit is used for substituting the primary space-time combination value of the earthquake post-stack pure wave data and the primary space-time combination value of the earthquake post-stack result data into the three-dimensional earthquake Kalman filtering model, and performing multiple iterations to respectively obtain multiple time-time combination values of the earthquake post-stack pure wave data and multiple time-time combination values of the earthquake post-stack result data.
The invention also relates to a computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the above-mentioned method.
The invention also relates to a computer arrangement comprising a memory, a processor and a computer program stored on the memory and executable on the processor, which when executed by the processor implements the steps of the method described above.
Compared with the prior art, the invention has the beneficial effects that:
the invention discloses a method and a device for optimizing seismic data based on space-time combination, which are based on a three-dimensional seismic data body, establish a seismic channel state equation by a plurality of adjacent seismic channels in the transverse direction, mainly realize the transverse space constraint and denoising of a geologic body, establish a state equation of a seismic signal at a certain moment by adjacent time slices in the longitudinal direction, mainly realize the longitudinal space constraint and denoising of the geologic body, continuously and iteratively correct the state equation through a Kalman algorithm to recover the original purpose of effective information, only suppress high-frequency noise components, have no obvious reduction of resolution ratio, do not change the structural characteristics of seismic event axes and the relative relation on amplitude space, and overcome the problems that the suppression effect of the existing filtering technology on noise is not obvious and the fidelity and the amplitude preservation of seismic data are difficult. According to the seismic data optimization method based on the space-time combination, the resolution of the filtered seismic section is not obviously reduced, the wave group relation and the longitudinal and transverse seismic amplitude energy relation of the seismic section and the seismic data before filtering are kept unchanged, the signal-to-noise ratio of the inversion result of the filtered seismic data is obviously improved, and the transverse tracking of a geologic body is more efficient.
Drawings
FIG. 1 is a flow chart of a method for spatio-temporal combination-based seismic data optimization as provided in example 1 of the present invention;
FIG. 2 is a synthetic seismic recording section without Gaussian random noise to which the present invention relates;
FIG. 3 is a synthetic seismic recording section involving Gaussian random noise in accordance with the present invention;
FIG. 4 shows a two-dimensional median filter profile of the prior art;
FIG. 5 is a filtering result of the method and apparatus for seismic data optimization based on spatio-temporal combinations according to the present invention;
FIG. 6 is a plane view of a conventional frequency domain low-pass filtering curvature property fracture detection;
FIG. 7 is a plan view of a filtered curvature attribute fracture detection for a method of spatiotemporal combination-based seismic data optimization according to the present invention;
FIG. 8 is a prior art partially stacked, wide angle seismic section;
FIG. 9 is a partially folded, increased angle seismic section after filtering in accordance with the present invention;
FIG. 10 is a graph of seismic data filtering inversion results according to the present invention;
FIG. 11 is a diagram of inversion results after seismic data filtering according to the present invention.
Detailed Description
The following examples are intended to illustrate the invention but are not intended to limit the scope of the invention.
The Kalman filtering is an optimized autoregressive signal filtering processing algorithm, a real effective signal is recovered from a digital signal polluted by noise by recursion in a sequence of 'prediction-measurement-correction', the calculation process of the Kalman filtering comprises two steps of prediction and correction, the prediction is to estimate the state of the current moment based on the state of the last moment, and the correction is to integrate the estimation state and the observation state of the current moment and estimate the optimal state of the signal. The Kalman filtering algorithm has great advantages in improving signal-to-noise ratio of signals.
The invention provides a three-dimensional seismic Kalman filtering iteration calculation method based on space-time combination, which is used for calculating three-dimensional seismic Kalman filtering iteration filtering, and has the advantages of removing high-frequency noise on the premise of retaining seismic high-frequency effective signal components, improving the signal-to-noise ratio and the transverse continuity of a same-phase axis of seismic signals, not changing the information of a geologic body in the seismic signals and having great significance for improving the structural interpretation of subsequent seismic data and the reservoir prediction precision and efficiency.
The low-pass filtering is to cut off the high-frequency component of the seismic signal, because the noise is mainly concentrated on the high-frequency component, the signal-to-noise ratio of the seismic data can be obviously improved, but the seismic resolution can be reduced, the identification of the thin geologic body is not favorable, the median smoothing filtering is to eliminate the noise by utilizing the stacking principle of adjacent channels, and the effect on the denoising effect and the fidelity and amplitude preservation is poor. Compared with the low-pass filtering and median smoothing filtering of the frequency domain in the prior art, the seismic data optimization method based on the space-time combination has obvious advantages in denoising effect and fidelity and amplitude preservation.
Example 1
step A: acquiring earthquake post-stack pure wave data and earthquake post-stack result data;
the method specifically comprises the following steps:
step A1: acquiring seismic data, and acquiring a common reflection point gather by applying a seismic migration imaging method to the acquired seismic data;
step A2: carrying out full or partial superposition on the seismic data of the common reflection point gather to obtain post-seismic superposition pure wave data;
step A3: and performing channel equalization and filtering processing on the seismic post-stack pure wave data to obtain seismic post-stack result data.
And B, step B: establishing a three-dimensional seismic Kalman filtering model, which comprises a three-dimensional seismic transverse space Kalman filtering model and a three-dimensional seismic longitudinal space Kalman filtering model;
the method specifically comprises the following steps of:
step B11: calculating the state value of the k-th channel through the state conversion equation of the seismic channel on the transverse space
Considering that sedimentary rocks and geologic bodies have high similarity of adjacent seismic channels in transverse space due to the transverse spatial similarity, the adjacent seismic channels in transverse space can estimate the optimal state of the channel, and the expression of the state value of the k-th filtering channel is calculated according to a Kalman filtering algorithm as follows:
in the formula (I), the compound is shown in the specification,
filtering state values of a k channel in a matrix form;
A k the matrix is a transverse filtering state transformation matrix which is an identity matrix and is used as an intermediate quantity for connecting the kth track with the k-1, k +1, k-numcmp and k + numcmp tracks;
the upper corner mark S is a mark of the state; the lower corner mark k is the serial number of the seismic channel, and k is an integer greater than 1;
X k-1 and X k+1 Respectively the k-1 th and k +1 th tracks of the seismic section on the main survey line;
X k-numcmp and X k+numcmp The k-numcmp channel and the k + numcmp channel are respectively adjacent survey lines of a main survey line where the k channel of the seismic profile is located;
numcmp is the number of contact lines, and k > numcmp;
W k the error is translated for the horizontal filter state.
Step B12: calculating the measurement value of the k channel of the seismic channel through the measurement equation of the seismic channel
The measurement equation for a seismic trace is:
in the formula (I), the compound is shown in the specification,
a measurement representing a kth trace of a seismic trace;
V k for the transverse filtering of the measurement error, wherein the transverse filtering of the measurement error V k Transverse filter state transition error W k ；
Due to the similarity between the seismic adjacent traces of the sedimentary rock stratum, the transverse filtering state transition error W k With transverse filtering of the measurement error V k Approximately equal.
Step B13: computing the kth trace X of a seismic section k Corresponding standard deviation P of k-th channel of filtering k ；
When k is 2, trace 2P of seismic section 2 For filtering state values of the 2 nd track
Standard deviation of (d);
when k is more than or equal to 3, the k-th channel X of the seismic section k Corresponding standard deviation P of k-th channel k The update equation set of (1) is:
in the formula, P k The standard deviation of a k-th filtering channel corresponding to the k-th channel of the seismic section is in a matrix form;
P k-1 the standard deviation of a k-1 channel of the seismic section corresponding to the k-1 channel of the filtering;
R k for filtering the standard deviation P of the k-th track k Standard deviation of error in the update process;
std denotes taking the standard deviation, X k+1 Is k +1 trace, X of seismic section on main survey line k-1 Is the k-1 st trace, X of the seismic section on the main survey line k-numcmp The k-numcmp trace, X, of a main survey line adjacent to the k-th trace of the seismic section k+numcmp Respectively is the k + numcmp track of the adjacent measuring line of the main measuring line where the k track is positioned;
step B14: according to the state value of the k-th channel
Seismic trace kth measurement
And the k channel X of seismic section k Corresponding standard deviation P of k-th channel k For the k channel X of seismic section k Performing optimization calculation;
wherein, the k-th channel X of the seismic section k The optimization calculation equation set is as follows:
in the formula (I), the compound is shown in the specification,
for the k channel X of seismic section k OptimizationThe seismic traces that follow are then processed,
to filter the state values of the k-th channel,
representing the filtered k-th channel measurement;
Kg k kalman gain, which is a lateral filtering;
Q k the standard deviation between the kth seismic data of the seismic channel and the adjacent seismic data channel is set as the standard deviation, wherein the difference between the seismic adjacent channels of the sedimentary rock stratum can be approximately regarded as stable random white Gaussian noise;
std denotes taking the standard deviation, X k+1 Is k +1 trace, X in the seismic section on the main survey line k-1 Is the k-1 st trace, X, in the seismic section on the main survey line k-numcmp The k-numcmp trace, X, of a main survey line adjacent to the k-th trace of the seismic section k+numcmp Respectively is the k + numcmp channel of the adjacent survey line of the main survey line where the k channel of the seismic profile is located; x k Representing the kth trace of the seismic section.
Step B15: updating the standard deviation P of the k-th channel of the seismic section corresponding to the k-th channel of the filtering k ；
The method specifically comprises the following steps:
step B151: let k be i, update the standard deviation P of k trace of seismic section k As the standard deviation P of the corresponding ith seismic trace i Wherein i is a positive integer greater than or equal to 2, and the updating formula is as follows:
P i ＝P k (1-Kg k ) (formula 5)
In the formula, P i Is the standard deviation P of the k-th trace of the seismic section k Standard deviation, P, corresponding to seismic trace after completion of updating k Is the standard deviation, Kg, of the kth trace corresponding to the kth trace of the seismic section k Kalman gain, which is a lateral filtering;
step B152: let P k-1 ＝P i And substituting into the k channel X of the seismic profile when k is more than or equal to 3 k Corresponding standard deviation P of k-th channel k Is moreCalculating in the new equation set to obtain the i +1 th seismic trace X of the seismic section i+1 Standard deviation P of corresponding i +1 th seismic trace i+1 ；
Step B153: carrying out the self-increment 1 operation of the seismic channel k, and assigning k which completes the self-increment 1 operation to i;
step B154: repeating the steps B51 to B53, and performing loop iteration to obtain the kth trace X of the seismic section k Corresponding standard deviation P of k-th channel k Updating of (1);
step B16: and repeating the steps B11 to B15 until the establishment of the three-dimensional seismic transverse space Kalman filtering model is completed.
Specifically, the establishment of the three-dimensional seismic longitudinal space kalman filtering model specifically comprises the following steps:
step B21: calculating the data state value of the t-th time of the three-dimensional seismic data volume by the state conversion equation of the slice at the seismic sampling time point in the longitudinal space
Considering that sedimentary rocks and geologic bodies have certain time thickness in the longitudinal direction and have similar sedimentary environments in the longitudinal adjacent geologic time, the seismic data of the longitudinal spatial adjacent sampling time points can estimate the optimal state of the sampling time point, and then a state conversion equation of a slice at the seismic sampling time point in the longitudinal space can be established according to a Kalman filtering algorithm, wherein the expression of the state conversion equation is as follows:
in the formula (I), the compound is shown in the specification,
the state value of the three-dimensional seismic data volume at the t moment is t, and t is an integer greater than 1;
S t-1 is a time slice of the t-1 th time of the three-dimensional seismic data volume;
S t+1 is threeTime slicing at the t +1 th moment of the dimension seismic data volume;
AT t the longitudinal filtering state transformation matrix is an identity matrix and is used for constraining the position of the geologic body at the longitudinal real time as an intermediate quantity for linking the tth time with t-1 and t + 1;
WT t is the state transition error of the vertical filtering.
Step B22: calculating the measurement value of the slice at the t moment of the seismic data through the measurement equation of the slice at the t moment of the seismic data
The measurement equation of the t-time slice of the seismic data is
In the formula (I), the compound is shown in the specification,
measurements representing slices at time t of the seismic data,
representing the real value of the seismic data slice at the t moment;
VT t a longitudinally filtered measurement error, wherein the state transition error WT of the longitudinal filtering can be considered due to the similarity between seismic adjacent moments of the sedimentary rock formations t Measurement error VT with longitudinal filtering t Are equal.
Step B23: calculating seismic data S of t-time slice of three-dimensional seismic data volume t Standard deviation of (PT) t ；
Time slice PT at time 2 of the three-dimensional seismic data volume when k is 2 2 State values for the 2 nd time of the three-dimensional seismic data volume
Standard deviation of (d);
when k is greater than or equal to 3, threeSeismic data S of seismic slice at t-th time of dimensional seismic data volume t Standard deviation of (PT) t The update equation set is:
in the formula, PT t Seismic data S being seismic slices of a three-dimensional seismic data volume at time t t Standard deviation of (d);
PT t-1 seismic data S being seismic slices of a three-dimensional seismic data volume at time t-1 t-1 The standard deviation of (a);
RT t seismic data S for seismic slices at time t of a three-dimensional seismic data volume t Standard deviation of (PT) t Standard deviation of error in the update process;
stdT represents taking the standard deviation;
S t-1 seismic data for a time slice at time t-1 of a three-dimensional seismic data volume;
S t+1 is the seismic data of the time slice at time t +1 of the three-dimensional seismic data volume.
Step B24: according to the t-th time data state of the three-dimensional seismic data volume
Measurements of time t slices of seismic data
And seismic data S sliced at time t of three-dimensional seismic data volume t Seismic data S of corresponding t-time slice t Standard deviation of (PT) t Slicing seismic data S of a three-dimensional seismic data volume at time t t Performing optimization calculation;
wherein the three-dimensional seismic data volume is sliced at time tSeismic data S t The optimization calculation equation set is as follows:
in the formula (I), the compound is shown in the specification,
seismic time slicing after seismic data slicing at the t-th time of the three-dimensional seismic data volume is optimized;
KgT t kalman gain, which is longitudinal filtering;
QT t the standard deviation between the t-th time of the seismic data and the time slice of the adjacent seismic data is obtained, wherein the difference between the seismic adjacent channels of the sedimentary rock stratum can be approximately regarded as stable random white Gaussian noise;
stdT denotes standard deviation, S t-1 Is a time slice of the t-1 th time of the three-dimensional seismic data volume; s t+1 Is a time slice of the three-dimensional seismic data volume at time t + 1; s t Time slicing is carried out on the t-th time of the three-dimensional seismic data body.
Step B25: updating seismic data S of t-time slice of three-dimensional seismic data body t Standard deviation of (PT) t ；
The method specifically comprises the following steps:
step B251: and (d) updating the seismic data S of the three-dimensional seismic data volume slice at the time t by taking t as j t Standard deviation of (PT) j Wherein j is a positive integer greater than or equal to 2, and the updating formula is as follows:
PT j ＝PT t (1-KgT t ) (formula 10)
In the formula, PT j Seismic data S sliced for time t of three-dimensional seismic data volume t Standard deviation of (PT) i Standard deviation, PT, corresponding to the updated seismic data t Seismic data S being a t-time slice of a three-dimensional seismic data volume t Standard deviation of (A), KgT t Kalman gain, which is longitudinal filtering;
step B252: let PT t-1 ＝PT j Substituting into seismic data S of t-th time slice of three-dimensional seismic data volume when k is more than or equal to 3 t Standard deviation of (PT) t Updating the equation set for calculation to obtain the standard deviation P of the seismic data of the seismic slice at the t +1 th time of the three-dimensional seismic data volume i+1 ；
Step B253: carrying out self-increment 1 operation on the seismic channel t, and assigning t which finishes the self-increment 1 operation to i;
step B254: repeating the steps B51 to B53, and performing loop iteration to obtain seismic data S of the t-time slice of the three-dimensional seismic data volume t Standard deviation of (PT) t Updating of (1);
step B26: and repeating the steps B21 to B25 until the establishment of the three-dimensional seismic longitudinal space Kalman filtering model is completed.
Step C: performing three-dimensional seismic Kalman iterative filtering of a time-space combination;
and substituting the earthquake post-stack pure wave data and the earthquake post-stack result data into a three-dimensional earthquake transverse space Kalman filtering model, respectively completing two filtering iterations of the three-dimensional earthquake transverse space, and then respectively substituting the three-dimensional earthquake transverse space Kalman filtering model into a three-dimensional earthquake longitudinal space Kalman filtering model to perform one filtering iteration of the three-dimensional earthquake longitudinal space.
Specifically, the post-seismic-stack pure wave data and post-seismic-stack result data acquired in the step A are substituted into the three-dimensional seismic transverse space Kalman filtering model established in the step B, and two filtering iterations of the three-dimensional seismic transverse space are respectively completed to obtain two transverse filtering values of the post-seismic-stack pure wave and two transverse filtering values of the post-seismic-stack result data; and B, respectively substituting the two-time transverse filtering value of the earthquake overlapped pure wave and the two-time transverse filtering value of the earthquake overlapped result data into the three-dimensional earthquake longitudinal space Kalman filtering model established in the step B, and performing one-time filtering iteration of the three-dimensional earthquake longitudinal space to respectively obtain a one-time space-time combination value of the earthquake overlapped pure wave data and a one-time space-time combination value of the earthquake overlapped result data.
Step D: performing three-dimensional seismic Kalman iterative filtering for multiple time-space combinations;
substituting the primary space-time combination value of the earthquake post-stack pure wave data and the primary space-time combination value of the earthquake post-stack result data into the three-dimensional earthquake Kalman filtering model, and performing multiple iterations to respectively obtain multiple time-time combination values of the earthquake post-stack pure wave data and multiple time-time combination values of the earthquake post-stack result data. The method specifically comprises the following steps:
step D1: substituting the primary space-time combination value of the earthquake post-stack result data and the primary space-time combination value of the earthquake post-stack result data into the three-dimensional earthquake transverse space Kalman filtering model to respectively complete two filtering iterations of the three-dimensional earthquake transverse space, and substituting the filtering iterations into the three-dimensional earthquake longitudinal space Kalman filtering model to perform primary filtering iteration of the three-dimensional earthquake longitudinal space to respectively obtain a secondary space-time combination value of the earthquake post-stack result data and a secondary space-time combination value of the earthquake post-stack result data;
step D2: substituting the secondary space-time combination value of the earthquake post-stack result data and the secondary space-time combination value of the earthquake post-stack result data into the three-dimensional earthquake transverse space Kalman filtering model to respectively complete two filtering iterations of the three-dimensional earthquake transverse space, and substituting the filtering iterations into the three-dimensional earthquake longitudinal space Kalman filtering model to perform one filtering iteration of the three-dimensional earthquake longitudinal space to respectively obtain a three-time space-time combination value of the earthquake post-stack result data and a three-time space-time combination value of the earthquake post-stack result data;
step D3: and substituting the three-time-space combination value of the earthquake post-stack result data and the three-time-space combination value of the earthquake post-stack result data into the three-dimensional earthquake transverse space Kalman filtering model to respectively complete two filtering iterations of the three-dimensional earthquake transverse space, and substituting the three-time-space combination value into the three-dimensional earthquake longitudinal space Kalman filtering model to perform one filtering iteration of the three-dimensional earthquake longitudinal space to respectively obtain the four-time-space combination values of the earthquake post-stack result data and the four-time-space combination values of the earthquake post-stack result data.
The three times of combined filtering are adopted to realize that the filtering of the earthquake is three-dimensional earthquake Kalman iterative filtering combined in transverse and longitudinal spaces.
And E, step E: reliability test
The method comprises the following steps of determining the reliability of filtered data by using well data and known geological information, and performing construction explanation and reservoir prediction on the four time space-time combination values of seismic post-stack result data and the four time space-time combination values of the seismic post-stack result data, wherein the method specifically comprises the following steps: and making a synthetic record by using the logging sound waves and the density curve, calibrating and comparing the quartic time-space combination value of the filtered seismic post-stack result data with the quartic time-space combination value of the seismic post-stack result data, determining the fidelity amplitude of the filtered seismic data, and performing seismic structure interpretation and reservoir prediction work such as horizon tracking, fault interpretation, stratum elastic parameter inversion and the like on the filtered seismic data on the basis.
The drying effect of the process is explained below:
as can be seen from FIGS. 2 and 3, the profile gas reservoir and the oil reservoir without noise can be clearly imaged, and the seismic event and the oil reservoir cannot be identified after Gaussian random noise is added.
As can be seen from the graphs in FIGS. 4 and 5, the two-dimensional median filtering denoising effect is not obvious, the filtering of the invention can better recover the original view of the seismic section, and the seismic characteristics of the oil and gas reservoir are effectively recovered.
As can be seen from FIGS. 6 and 7, the fracture detection result of the seismic data filtered by the method disclosed by the invention is more clear and reliable.
As can be seen from FIGS. 8 and 9, the signal-to-noise ratio of the seismic profile after filtering by the method disclosed by the invention is obviously improved.
As can be seen from FIGS. 10 and 11, the signal-to-noise ratio of the filtered inversion result is obviously improved by applying the method disclosed by the invention, the well-to-seismic coincidence of the filtered inversion result is better, and the geologic body characteristics are clearer.
In summary, compared with the existing two-dimensional median filtering technology, the seismic data optimization method based on the space-time combination has the advantages that the filtering effect is better, and the signal-to-noise ratio and the horizontal continuity of the same phase axis of the seismic data are obviously improved. The resolution of the filtered seismic profile is not obviously reduced, the wave group relation and the longitudinal and transverse seismic amplitude energy relation of the seismic profile and the seismic data before filtering are kept unchanged, the signal-to-noise ratio of the inversion result of the filtered seismic data is obviously improved, and the transverse tracking of the geologic body is more efficient.
Although the invention has been described in detail above with reference to a general description and specific examples, it will be apparent to one skilled in the art that modifications or improvements may be made thereto based on the invention. Accordingly, such modifications and improvements are intended to be within the scope of the invention as claimed.
Claims (7)
1. A method for spatio-temporal combination-based seismic data optimization, comprising:
acquiring pure wave data after earthquake superposition and result data after earthquake superposition;
establishing a three-dimensional seismic Kalman filtering model, which comprises a three-dimensional seismic transverse space Kalman filtering model and a three-dimensional seismic longitudinal space Kalman filtering model;
the method specifically comprises the following steps of establishing the three-dimensional seismic transverse space Kalman filtering model:
step B11: calculating the state value of the k-th channel through the state conversion equation of the seismic channel on the transverse space
The expression is as follows:
in the formula (I), the compound is shown in the specification,
filtering state values of a k channel in a matrix form; a. the k The transverse filtering state transformation matrix is an identity matrix and is used as an intermediate quantity for connecting the k-th track with the k-1, the k +1, the k-numcmp and the k + numcmp tracks; the upper corner mark S is a mark of the state; the lower corner mark k is the serial number of the seismic channel, and k is an integer greater than 1; x k-1 And X k+1 Respectively is the k-1 and k +1 channels of the seismic section on the main survey line; x k-numcmp And X k+numcmp The k-numcmp channel and the k + numcmp channel are respectively adjacent survey lines of a main survey line where the k channel of the seismic profile is located; numcmp is the number of contact lines, and k > numcmp; w is a group of k Switching the error for the transverse filtering state;
step B12: calculating the k-th channel measurement value of the seismic channel through the measurement equation of the seismic channel
The expression is
In the formula (I), the compound is shown in the specification,
a measurement representing a kth trace of a seismic trace;
representing the true value of the kth channel of the seismic channel; v k For transversely filtering the measurement error, wherein the transversely filtered measurement error V k Transverse filter state transition error W k ；
Step B13: computing the kth trace X of a seismic section k Corresponding standard deviation P of k-th channel k The expression is as follows:
when k is 2, trace 2P of seismic section 2 For filtering state values of the 2 nd track
Standard deviation of (d);
when k is more than or equal to 3, the k-th channel X of the seismic section k Corresponding standard deviation P of k-th channel k The update equation set of (1) is:
in the formula, P k The standard deviation of a k-th filtering channel corresponding to the k-th channel of the seismic section is in a matrix form; p k-1 The standard deviation of a k-1 channel of the seismic section corresponding to the k-1 channel of the filtering;
is a transverse filter state transformation matrix A k The transposed matrix of (2); r k For filtering the standard deviation P of the k-th channel k Standard deviation of error in the update process; std denotes taking the standard deviation, X k+1 Is k +1 trace, X of seismic section on main survey line k-1 Is the k-1 st trace, X of the seismic section on the main survey line k-numcmp K-numcmp trace, X, of a survey line adjacent to the main survey line in which the k-trace of the seismic section is located k+numcmp Respectively is the k + numcmp track of the adjacent measuring line of the main measuring line where the k track is positioned;
step B14: according to the state value of the k-th channel
Seismic trace kth measurement
And the k channel X of seismic section k Corresponding standard deviation P of k-th track k For the kth trace X of the seismic section k Performing optimization calculation;
wherein, the k-th channel X of the seismic section k The optimization calculation equation set is as follows:
in the formula (I), the compound is shown in the specification,
for the k channel X of seismic section k The optimized seismic channels are obtained by the method,
to filter the state values of the k-th track,
representing the filtered k-th trace measurement; kg k Kalman gain which is a lateral filtering; q k The standard deviation between the kth seismic data of the seismic channel and the adjacent seismic data channel is shown, wherein the difference between the seismic adjacent channels of the sedimentary rock stratum can be approximately regarded as stable random white Gaussian noise; std denotes taking the standard deviation, X k+1 Is k +1 trace, X in seismic section on main survey line k-1 Is the k-1 st trace, X, in the seismic section on the main survey line k-numcmp The k-numcmp trace, X, of a main survey line adjacent to the k-th trace of the seismic section k+numcmp Respectively is the k + numcmp channel of the adjacent survey line of the main survey line where the k channel of the seismic profile is located; x k A kth trace representing a seismic section;
step B15: updating the standard deviation P of the k-th channel of the seismic section corresponding to the k-th channel of the filtering k ；
Step B16: repeating the step B11 to the step B15 until the establishment of the three-dimensional seismic transverse space Kalman filtering model is completed;
the method for establishing the three-dimensional seismic longitudinal space Kalman filtering model specifically comprises the following steps:
step B21: calculating the data state of the t-th time of the three-dimensional seismic data volume by the state conversion equation of the slice at the seismic sampling time point in the longitudinal space
Step B22: through seismic data t timeCalculating the measurement value of the slice at the t moment of the earthquake by using the slicing measurement equation
Step B23: calculating seismic data S of t-time slice of three-dimensional seismic data volume t Standard deviation of (PT) t ；
Step B24: according to the t-th time data state of the three-dimensional seismic data volume
Seismic data time t slice measurement
Seismic data S sliced at time t from three-dimensional seismic data volume t Seismic data S of corresponding t-time slice t Standard deviation of (PT) t Slicing seismic data S of a three-dimensional seismic data volume at time t t Performing optimization calculation;
wherein, the seismic data S of the t-th time slice of the three-dimensional seismic data volume t The optimization calculation equation set is as follows:
in the formula (I), the compound is shown in the specification,
slicing the seismic data sliced at the t-th time of the three-dimensional seismic data volume by using the optimized seismic time slice;
the state value of seismic data sliced at the t-th time of the three-dimensional seismic data volume;
the measured value of the slice at the t moment of the three-dimensional seismic data volume is obtained; KgT t Kalman gain, which is longitudinal filtering; AT t T Is a longitudinal filter state transformation matrix AT t The transposed matrix of (2); QT t The standard deviation between the tth moment of the seismic data and the time slice of the adjacent seismic data is shown, wherein the difference between the seismic adjacent channels of the sedimentary rock stratum can be approximately regarded as stable random white Gaussian noise; stdT denotes standard deviation, S t-1 Is a time slice of the t-1 th time of the three-dimensional seismic data volume; s t+1 Is a time slice of the three-dimensional seismic data volume at time t + 1; s t Time slicing is carried out on the t-th time of the three-dimensional seismic data body;
step B25: updating seismic data S of t-time slice of three-dimensional seismic data volume t Standard deviation of (PT) t ；
Step B26: repeating the steps B21 to B25 until the establishment of the three-dimensional seismic longitudinal space Kalman filtering model is completed;
substituting the earthquake post-stack pure wave data and the earthquake post-stack result data into the three-dimensional earthquake transverse space Kalman filtering model, respectively completing two filtering iterations of a three-dimensional earthquake transverse space, and then respectively substituting the three-dimensional earthquake transverse space Kalman filtering model for performing one filtering iteration of a three-dimensional earthquake longitudinal space, so as to respectively obtain a primary space-time combination value of the earthquake post-stack pure wave data and a primary space-time combination value of the earthquake post-stack result data;
substituting the primary time-space combination value of the pure wave data after the earthquake superposition and the primary time-space combination value of the result data after the earthquake superposition into the three-dimensional earthquake Kalman filtering model, and performing multiple iteration to respectively obtain multiple time-space combination values of the pure wave data after the earthquake superposition and multiple time-space combination values of the result data after the earthquake superposition.
2. The method of claim 1, wherein the acquiring of the seismic post-stack pure wave data and the seismic post-stack result data specifically comprises the steps of:
step A1: acquiring seismic data, and acquiring a common reflection point gather by applying a seismic migration imaging method to the acquired seismic data;
step A2: carrying out full or partial superposition on the seismic data of the common reflection point gather to obtain post-seismic superposition pure wave data;
step A3: and performing channel equalization and filtering processing on the seismic post-stack pure wave data to obtain seismic post-stack result data.
3. The method of claim 1,
the method specifically comprises the following steps of:
step D1: substituting the primary space-time combination value of the earthquake post-stack result data and the primary space-time combination value of the earthquake post-stack result data into the three-dimensional earthquake transverse space Kalman filtering model to respectively complete two filtering iterations of the three-dimensional earthquake transverse space, and substituting the filtering iterations into the three-dimensional earthquake longitudinal space Kalman filtering model to perform primary filtering iteration of the three-dimensional earthquake longitudinal space to respectively obtain a secondary space-time combination value of the earthquake post-stack result data and a secondary space-time combination value of the earthquake post-stack result data;
step D2: substituting the secondary space-time combination value of the earthquake post-stack result data and the secondary space-time combination value of the earthquake post-stack result data into the three-dimensional earthquake transverse space Kalman filtering model to respectively complete two filtering iterations of the three-dimensional earthquake transverse space, and substituting the filtering iterations into the three-dimensional earthquake longitudinal space Kalman filtering model to perform one filtering iteration of the three-dimensional earthquake longitudinal space to respectively obtain a three-time space-time combination value of the earthquake post-stack result data and a three-time space-time combination value of the earthquake post-stack result data;
step D3: and substituting the three-time-space combination value of the earthquake post-stack result data and the three-time-space combination value of the earthquake post-stack result data into the three-dimensional earthquake transverse space Kalman filtering model to respectively complete two times of filtering iteration of the three-dimensional earthquake transverse space, and substituting the three-time-space combination value into the three-dimensional earthquake longitudinal space Kalman filtering model to perform one time of filtering iteration of the three-dimensional earthquake longitudinal space to respectively obtain the four-time-space combination values of the earthquake post-stack result data and the four-time-space combination values of the earthquake post-stack result data.
4. The method of claim 3, wherein said step D3 is further followed by the steps of:
and making a quartic time-space combination value of the synthetic record and the post-seismic-stack result data and a quartic time-space combination value of the post-seismic-stack result data by using the logging sound waves and the density curves, calibrating and comparing, determining the fidelity amplitude of the filtered seismic data, and performing seismic structure interpretation and reservoir prediction on the filtered seismic data on the basis, wherein the seismic structure interpretation comprises horizon tracking, fault interpretation and stratum elasticity parameter inversion.
5. An apparatus for spatiotemporal combination-based seismic data optimization, comprising:
the first processing unit is used for acquiring seismic post-stack pure wave data and seismic post-stack result data;
the second processing unit is used for establishing a three-dimensional earthquake Kalman filtering model, and the three-dimensional earthquake transverse space Kalman filtering model and the three-dimensional earthquake longitudinal space Kalman filtering model are included;
the method for establishing the three-dimensional seismic transverse space Kalman filtering model specifically comprises the following steps:
step B11: calculating the state value of the k channel through the state conversion equation of the seismic channel on the transverse space
The expression is as follows:
in the formula (I), the compound is shown in the specification,
filtering state values of a k channel in a matrix form; a. the k For transverse filteringThe state transformation matrix is an identity matrix and is used as an intermediate quantity for connecting the kth track with the k-1, k +1, k-numcmp and k + numcmp tracks; the upper corner mark S is a mark of the state; the lower corner mark k is the serial number of the seismic channel, and k is an integer greater than 1; x k-1 And X k+1 Respectively is the k-1 and k +1 channels of the seismic section on the main survey line; x k-numcmp And X k+numcmp Respectively a k-numcmp channel and a k + numcmp channel of a main survey line adjacent to the k channel of the seismic profile; numcmp is the number of contact lines, and k > numcmp; w k Switching the error for the transverse filtering state;
step B12: calculating the k-th channel measurement value of the seismic channel through the measurement equation of the seismic channel
The expression is
In the formula (I), the compound is shown in the specification,
a measurement representing a kth trace of a seismic trace;
representing the true value of the kth channel of the seismic channel; v k For transversely filtering the measurement error, wherein the transversely filtered measurement error V k Transverse filter state transition error W k ；
Step B13: computing the kth trace X of a seismic section k Corresponding standard deviation P of k-th channel k The expression is as follows:
when k is 2, trace 2P of seismic section 2 For filtering state values of the 2 nd track
Standard deviation of (d);
when k is more than or equal to 3, the k-th channel X of the seismic section k Corresponding filtered kth trackStandard deviation P of k The update equation set of (1) is:
in the formula, P k The standard deviation of a k-th filtering channel corresponding to the k-th channel of the seismic section is in a matrix form; p k-1 The standard deviation of the k-1 channel seismic channel of the filtering corresponding to the k-1 channel of the seismic section is obtained;
is a transverse filter state transformation matrix A k The transposed matrix of (2); r k For filtering the standard deviation P of the k-th channel k Standard deviation of error in the update process; std denotes taking the standard deviation, X k+1 Is the seismic section k +1 track, X on the main survey line k-1 Is the k-1 st trace, X of the seismic section on the main survey line k-numcmp The k-numcmp trace, X, of a main survey line adjacent to the k-th trace of the seismic section k+numcmp Respectively is a k + numcmp track of a measuring line adjacent to the main measuring line where the k track is located;
step B14: according to the state value of the k-th channel
Seismic trace kth measurement
And the k channel X of seismic section k Corresponding standard deviation P of k-th track k For the kth trace X of the seismic section k Performing optimization calculation;
wherein, the k-th channel X of the seismic section k The optimization calculation equation set is as follows:
in the formula (I), the compound is shown in the specification,
for the k channel X of seismic section k The optimized seismic channels are obtained by the method,
to filter the state values of the k-th channel,
representing the filtered k-th channel measurement; kg k Kalman gain, which is a lateral filtering; q k The standard deviation between the kth seismic data of the seismic channel and the adjacent seismic data channel is set as the standard deviation, wherein the difference between the seismic adjacent channels of the sedimentary rock stratum can be approximately regarded as stable random white Gaussian noise; std denotes taking the standard deviation, X k+1 Is k +1 trace, X in seismic section on main survey line k-1 Is the k-1 st trace, X, in the seismic section on the main survey line k-numcmp The k-numcmp trace, X, of a main survey line adjacent to the k-th trace of the seismic section k+numcmp Respectively a k + numcmp trace of a main survey line adjacent to the k trace of the seismic profile; x k A kth trace representing a seismic section;
step B15: updating the standard deviation P of the k-th channel of the seismic section corresponding to the k-th channel of the filtering k ；
Step B16: repeating the steps B11 to B15 until the establishment of the three-dimensional seismic transverse space Kalman filtering model is completed;
the method for establishing the three-dimensional seismic longitudinal space Kalman filtering model specifically comprises the following steps:
step B21: calculating the data state of the t-th time of the three-dimensional seismic data volume by the state conversion equation of the slice at the seismic sampling time point in the longitudinal space
Step B22: calculating the measurement value of the slice at the t moment of the earthquake by the measurement equation of the slice at the t moment of the earthquake data
Step B23: computing seismic data S of a time t slice of a three-dimensional seismic data volume t Standard deviation of (PT) t ；
Step B24: according to the t-th time data state of the three-dimensional seismic data volume
Seismic data time t slice measurement
Seismic data S sliced at time t from three-dimensional seismic data volume t Seismic data S of corresponding t-time slice t Standard deviation of (PT) t Slicing seismic data S of a three-dimensional seismic data volume at time t t Performing optimization calculation;
wherein, the seismic data S of the t-th time slice of the three-dimensional seismic data volume t The optimization calculation equation set is as follows:
in the formula (I), the compound is shown in the specification,
seismic time slicing after seismic data slicing at the t-th time of the three-dimensional seismic data volume is optimized;
the state value of seismic data sliced at the t-th time of the three-dimensional seismic data volume;
the measured value of the slice at the t moment of the three-dimensional seismic data volume is obtained; KgT t Kalman gain, which is longitudinal filtering; AT t T Is a longitudinal filter state transformation matrix AT t The transposed matrix of (2); QT t For the t-th time of seismic data andstandard deviation between adjacent seismic data time slices, wherein the difference between seismic adjacent traces of sedimentary rock formations can be approximately viewed as stationary random white gaussian noise; stdT denotes taking the standard deviation, S t-1 Is a time slice of the t-1 th time of the three-dimensional seismic data volume; s. the t+1 Is a time slice of the three-dimensional seismic data volume at time t + 1; s t Time slicing is carried out on the t-th time of the three-dimensional seismic data body;
step B25: updating seismic data S of t-time slice of three-dimensional seismic data volume t Standard deviation of (PT) t ；
Step B26: repeating the steps B21 to B25 until the establishment of the three-dimensional seismic longitudinal space Kalman filtering model is completed;
the third processing unit is used for substituting the post-seismic-stack pure wave data and the post-seismic-stack result data into the three-dimensional seismic transverse space Kalman filtering model, respectively completing two filtering iterations of a three-dimensional seismic transverse space, and then respectively substituting the two filtering iterations into the three-dimensional seismic longitudinal space Kalman filtering model to perform one filtering iteration of a three-dimensional seismic longitudinal space, so as to respectively obtain a primary space-time combination value of the post-seismic-stack pure wave data and a primary space-time combination value of the post-seismic-stack result data;
and the fourth processing unit is used for substituting the primary space-time combination value of the earthquake post-stack pure wave data and the primary space-time combination value of the earthquake post-stack result data into the three-dimensional earthquake Kalman filtering model, and performing multiple iterations to respectively obtain multiple time-time combination values of the earthquake post-stack pure wave data and multiple time-time combination values of the earthquake post-stack result data.
6. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 4.
7. A computer arrangement comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the steps of the method as claimed in any one of claims 1 to 4 are implemented by the processor when executing the computer program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111157271.4A CN113900140B (en) | 2021-09-30 | 2021-09-30 | Seismic data optimization method and device based on space-time combination |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111157271.4A CN113900140B (en) | 2021-09-30 | 2021-09-30 | Seismic data optimization method and device based on space-time combination |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113900140A CN113900140A (en) | 2022-01-07 |
| CN113900140B true CN113900140B (en) | 2022-08-23 |
Family
ID=79189443
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111157271.4A Active CN113900140B (en) | 2021-09-30 | 2021-09-30 | Seismic data optimization method and device based on space-time combination |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113900140B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107247290A (en) * | 2017-07-06 | 2017-10-13 | 西安交通大学 | A kind of seismic data noise drawing method filtered based on space-time fractional order |
| CN110857997A (en) * | 2018-08-23 | 2020-03-03 | 中国石油化工股份有限公司 | Step-by-step prestack elastic parameter inversion method and system based on transverse constraint |
| CN110873898A (en) * | 2018-08-30 | 2020-03-10 | 中国石油天然气股份有限公司 | Filtering method and device for seismic data |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8265876B1 (en) * | 2008-11-08 | 2012-09-11 | Ihs Global Inc. | Seismic horizon autopicking using orientation vector field |
| US10670753B2 (en) * | 2014-03-03 | 2020-06-02 | Saudi Arabian Oil Company | History matching of time-lapse crosswell data using ensemble kalman filtering |
| CN112698392A (en) * | 2020-11-20 | 2021-04-23 | 中国石油天然气股份有限公司 | Multilayer system reservoir three-dimensional drawing and oil-gas spatial distribution and constant volume method |
| CN113419278B (en) * | 2021-06-21 | 2022-03-08 | 大庆油田有限责任公司 | Well-seismic joint multi-target simultaneous inversion method based on state space model and support vector regression |
-
2021
- 2021-09-30 CN CN202111157271.4A patent/CN113900140B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107247290A (en) * | 2017-07-06 | 2017-10-13 | 西安交通大学 | A kind of seismic data noise drawing method filtered based on space-time fractional order |
| CN110857997A (en) * | 2018-08-23 | 2020-03-03 | 中国石油化工股份有限公司 | Step-by-step prestack elastic parameter inversion method and system based on transverse constraint |
| CN110873898A (en) * | 2018-08-30 | 2020-03-10 | 中国石油天然气股份有限公司 | Filtering method and device for seismic data |
Non-Patent Citations (1)
| Title |
|---|
| 横纵向组合的卡尔曼地震资料迭代滤波方法;陈刚 等;《石油地球物理勘探》;20210630;第56卷(第3期);第468-474页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113900140A (en) | 2022-01-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Nazari Siahsar et al. | Data-driven multitask sparse dictionary learning for noise attenuation of 3D seismic data | |
| US10436924B2 (en) | Denoising seismic data | |
| CN111596366B (en) | Wave impedance inversion method based on seismic signal optimization processing | |
| CN111025386B (en) | Vertical and horizontal wave separation method without separation false image | |
| EA032186B1 (en) | Seismic adaptive focusing | |
| CN107179550B (en) | A kind of seismic signal zero phase deconvolution method of data-driven | |
| Zhong‐Lai et al. | A SECOND‐ORDER SYNCHROSQUEEZING S‐TRANSFORM AND ITS APPLICATION IN SEISMIC SPECTRAL DECOMPOSITION | |
| CN116520419B (en) | Hot fluid crack channel identification method | |
| CN110687597B (en) | Wave impedance inversion method based on joint dictionary | |
| CN113900140B (en) | Seismic data optimization method and device based on space-time combination | |
| CN116068644B (en) | Method for improving resolution and noise reduction of seismic data by using generation countermeasure network | |
| AU2015309050B2 (en) | Joint estimation of electromagnetic earth responses and ambient noise | |
| CN108919345B (en) | Submarine cable land detection noise attenuation method | |
| CN114779332A (en) | Seismic data deposition background removing method and device and electronic equipment | |
| CN111239805B (en) | Block constraint time-lapse seismic difference inversion method and system based on reflectivity method | |
| Xu et al. | Automatic source localization and attenuation of seismic interference noise using density-based clustering method | |
| US9784869B2 (en) | Noise models by selection of transform coefficients | |
| CN112558147B (en) | Polarization analysis method and system for microseism data in well | |
| CN113391348B (en) | Common reflection point gather construction method and device for prestack inversion | |
| Sun et al. | Deep learning-based Vz-noise attenuation for OBS data | |
| CN111352158A (en) | Seismic signal enhancement method and device | |
| CN110398773B (en) | Recovery and reconstruction method for partial missing seismic data | |
| CN114114422B (en) | Prestack seismic data noise elimination method based on directional multi-scale decomposition | |
| CN114624765B (en) | Phase domain seismic data processing and reconstructing method and device and storable medium | |
| CN116774293B (en) | Method, system, electronic equipment and medium for automatically picking up same phase shaft |
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 |