CN103760599B - A kind of miniature fault detection method and fault detection device - Google Patents
A kind of miniature fault detection method and fault detection device Download PDFInfo
- Publication number
- CN103760599B CN103760599B CN201410001435.8A CN201410001435A CN103760599B CN 103760599 B CN103760599 B CN 103760599B CN 201410001435 A CN201410001435 A CN 201410001435A CN 103760599 B CN103760599 B CN 103760599B
- Authority
- CN
- China
- Prior art keywords
- prediction
- data
- seismic
- module
- amplitude
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 43
- 238000001914 filtration Methods 0.000 claims abstract description 97
- 238000012545 processing Methods 0.000 claims abstract description 67
- 238000000034 method Methods 0.000 claims abstract description 41
- 238000004458 analytical method Methods 0.000 claims abstract description 20
- 238000005070 sampling Methods 0.000 claims description 19
- 230000001131 transforming effect Effects 0.000 claims description 12
- 238000004364 calculation method Methods 0.000 claims description 8
- 230000001427 coherent effect Effects 0.000 description 9
- 238000010586 diagram Methods 0.000 description 9
- 238000012937 correction Methods 0.000 description 7
- 239000000523 sample Substances 0.000 description 5
- 230000000694 effects Effects 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 230000005284 excitation Effects 0.000 description 3
- 238000013508 migration Methods 0.000 description 3
- 230000005012 migration Effects 0.000 description 3
- 230000035772 mutation Effects 0.000 description 3
- 230000003068 static effect Effects 0.000 description 3
- 238000011161 development Methods 0.000 description 2
- 238000010521 absorption reaction Methods 0.000 description 1
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 238000009792 diffusion process Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000000717 retained effect Effects 0.000 description 1
Landscapes
- Geophysics And Detection Of Objects (AREA)
Abstract
The present invention provides detection method and the fault detection device of a kind of micro-small fault.Described miniature fault detection method includes following process step: the seismic data recording of collection is carried out high data fidelity process；Before prediction after processing high data fidelity, geological data is predicted filtering in territory, frequency space；The amplitude of front to geological data after the prediction after the predictive filtering of territory, frequency space and prediction geological data is adjusted to same amplitude rank；Subtract each other corresponding with the amplitude of the geological data after the prediction of scaling for geological data before prediction；Geological data before prediction is done slice analysis with the corresponding predictive filtering residual error data subtracting each other acquisition of amplitude of geological data after prediction.Utilize the present invention, the feature of micro-small fault in geological data can be identified, and then identify micro-small fault.
Description
Technical Field
The invention relates to the field of seismic data interpretation in seismic exploration, in particular to a micro fault detection method and a fault detection device.
Background
In order to survey and develop more oil and gas resources in the stratum, more and more concealed and more finely structured oil and gas fields need to be found. Oil and gas resources are distributed in the fault. The stratum of earth is broken when it is stressed to a certain strength and has obvious relative movement along the broken surface. The fault is not only the boundary of the oil and gas reservoir, but also the channel for migration and aggregation of oil and gas resources, so that the identification of the fault is very important work in the oil and gas exploration and development.
The first step of fault identification is to obtain high fidelity seismic data, including the processes of acquiring density of the seismic data in the field and high fidelity data processing of the acquired seismic data indoors, etc. The first step of increasing seismic data acquisition density in the field mainly refers to increasing the excitation times of seismic sources of different shot points and increasing the distribution density of detection points for receiving seismic waves, and aims to improve the acquisition density of seismic data from the aspect of acquisition and acquire and record the reflection signals of underground faults to the maximum extent. And secondly, performing high-fidelity data processing on the acquired seismic data record. High fidelity data processing refers to a series of processing procedures for improving the signal-to-noise ratio of seismic data, improving the resolution ratio of the seismic data and maintaining the amplitude of the seismic data in seismic data processing.
Secondly, analyzing the high-fidelity seismic data obtained in the above step, and accurately identifying the fault in seismic data interpretation as far as possible through analyzing information such as amplitude, phase, waveform and the like of the seismic data.
The main method for detecting faults in high-fidelity seismic data at present is coherent body fault identification technology. The method is characterized in that wave features among seismic channels change in zones with faults, stratum lithology mutation and special geologic bodies, so that correlation between local channels changes suddenly, the correlation value is small, then correlation values of all sample points and sample points around the sample points are obtained to form a coherent data body, and the geologic phenomenon of the faults is revealed by horizontally slicing the coherent data body. During seismic data acquisition, the seismic data recorded at each receiver point that receives seismic information is called a seismic trace, or trace. The implementation process of the coherent body fault identification technology mainly comprises three steps, wherein the first step is to perform migration processing on the obtained high-fidelity three-dimensional seismic data. The migration processing can improve the resolution ratio of the seismic section, recover wavelet waveform and amplitude characteristics and facilitate seismic data interpretation. And secondly, in the three-dimensional data volume obtained through deviation, the coherent value of each sampling point and the data of the surrounding sampling points is obtained to form a coherent data volume. And the third step is to carry out slice analysis on the coherent data volume formed in the second step, and carry out structural and lithological interpretation.
Among the above-mentioned faults, a minor fault is a fault which has a small fault distance (usually not more than 10 meters) and is difficult to identify even if the signal-to-noise ratio is high. The coherent body fault identification technology is used for identifying the fault through slice analysis by identifying the characteristics of amplitude mutation, waveform mutation and the like of the seismic data caused by the fault in the seismic data. And the characteristics of the waveform, the amplitude, the phase and the like at two sides of the micro fault in the seismic data are not obviously changed. Even if a micro fault exists in the seismic data record, the change of the coherent value of the seismic wave characteristic at the micro fault is not large, the characteristic of the micro fault cannot be found, the micro fault cannot be identified, and the exploration and the development of an oil and gas reservoir where the micro fault is located are influenced.
Disclosure of Invention
The invention aims to provide a micro fault detection method and a fault detection device, wherein the predictive filtered seismic data is subjected to scaling and pre-predictive seismic data comparison, and the amplitude of the pre-predictive seismic data is correspondingly subtracted from the amplitude of the post-predictive seismic data sample point data subjected to frequency space domain predictive filtering to obtain predictive filtering residual data.
The invention provides a micro fault detection method, which comprises the following processing steps:
s1: carrying out high-fidelity data processing on the acquired seismic data record;
s2: performing predictive filtering on the pre-prediction seismic data after the high-fidelity data processing in a frequency space domain, wherein the process of performing predictive filtering in the frequency space domain comprises the following processing steps:
s201: transforming the pre-prediction seismic data from a time domain to a frequency space domain through Fourier transform to form space domain vector data;
s202: on a single frequency point of the space domain vector data, a predictor is obtained by a wiener filtering method;
s203: on the single-point frequency point of the space domain vector data, multiplying the seismic traces with a total length of one predictor before and after the predicted seismic trace by the corresponding predictors one by one, adding the multiplication results to be the single-point frequency of the predicted seismic trace, completing the prediction of all the single-point frequencies of the frequency domain, and forming the predicted data of the frequency space domain;
s204: transforming the predicted data of the frequency space domain from the frequency domain to the time domain by inverse fourier transform;
s3: adjusting the amplitudes of the post-prediction seismic data and the pre-prediction seismic data after the frequency space domain filtering to the same amplitude level to obtain post-prediction scaled data, wherein the process of adjusting the amplitudes of the post-prediction seismic data and the pre-prediction seismic data after the frequency space domain filtering to the same amplitude level comprises the following processing steps:
multiplying each seismic trace in the predicted seismic data after the frequency space domain prediction filtering by a coefficient corresponding to the seismic trace;
the coefficient is the ratio of the average value of the root-mean-square amplitude values of the seismic channels in the seismic data before prediction to the average value of the root-mean-square amplitude values of the seismic channels corresponding to the seismic data after prediction;
s4: correspondingly subtracting the amplitudes of sampling points of the pre-prediction seismic data and the post-prediction scaling data;
s5: and carrying out slice analysis on prediction filtering residual data obtained by correspondingly subtracting the amplitudes of the sampling points of the pre-prediction seismic data and the post-prediction scaling data.
In the above-described minor fault detection method, a preferred embodiment further includes setting a threshold value for prediction filtering residual data obtained by subtracting amplitude correspondence between the pre-prediction seismic data and the post-prediction seismic data in S4, and replacing data having an amplitude absolute value smaller than the threshold value in the prediction filtering residual data with zero.
In the above-described minor fault detection method, a preferable embodiment of the method is that, in S3, the process of adjusting the amplitudes of the post-prediction seismic data and the pre-prediction seismic data after the frequency-space domain prediction filtering to the same amplitude level includes the following processing steps:
multiplying each seismic trace in the pre-prediction seismic data by a coefficient corresponding to the seismic trace;
the coefficient is the ratio of the average value of the root mean square amplitude values of the seismic traces in the post-prediction seismic data to the average value of the root mean square amplitude values of the pre-prediction seismic data and the corresponding seismic traces.
The invention provides a fault detection device by utilizing the method for detecting the micro fault, and the fault detection device comprises a high-fidelity data processing module, a frequency space domain filtering module, an amplitude adjusting module, an amplitude subtracting module and a slice analysis module; wherein,
the high-fidelity data processing module is used for improving the signal-to-noise ratio and the resolution of the earthquake data;
the frequency space domain filtering module is used for carrying out predictive filtering on the pre-prediction seismic data obtained by the high-fidelity data processing module in a frequency space domain;
the amplitude adjusting module is used for adjusting the amplitude of the pre-prediction seismic data obtained by the high-fidelity data processing module and the amplitude of the post-prediction seismic data obtained by the frequency space domain filtering module to the same amplitude level;
the amplitude subtraction module is used for subtracting the amplitude of the seismic channel corresponding to the post-prediction zooming data obtained by the amplitude adjustment module and the pre-prediction seismic data obtained by the high-fidelity data processing module;
the frequency space domain filtering module comprises a Fourier transform module, a wiener filtering module, a prediction data module and an inverse Fourier transform module; wherein,
the Fourier transform module is used for transforming the pre-prediction seismic data obtained by the high-fidelity data module from a time domain to a frequency space domain through Fourier transform to form space domain vector data;
the wiener filtering module is used for solving a prediction operator by using a wiener filtering method on a single frequency point of the space domain vector data;
the prediction data module is used for multiplying the seismic traces with a total length of one predictor in front of and behind the predicted seismic trace one by one with the corresponding predictors on the single-point frequency point of the space domain vector data, adding the multiplication results to be the single-point frequency of the predicted seismic trace, completing the prediction of all the single-point frequencies of the frequency domain, and forming the prediction data of the frequency space domain;
an inverse Fourier transform module for transforming the prediction data of the frequency space domain obtained by the prediction data module from the frequency domain to the time domain by inverse Fourier transform;
the amplitude adjusting module comprises a coefficient calculating module and a data scaling module; wherein,
the coefficient calculation module is used for calculating the ratio of the average value of the root-mean-square amplitude values of the seismic channels in the seismic data before prediction to the average value of the root-mean-square amplitude values of the seismic channels corresponding to the seismic data after prediction;
and the data scaling module is used for multiplying the amplitude of the sampling point of each seismic channel in the predicted seismic data by the coefficient corresponding to the seismic channel calculated by the coefficient calculating module.
In the above-mentioned minor fault detection apparatus, it is preferable that the amplitude subtraction module further includes a threshold processing module for filtering interference information in the prediction filtering residual data, and the threshold processing module includes a threshold setting module and a threshold processing module; wherein,
the threshold value setting module is used for setting a threshold value for the prediction filtering residual data;
and the threshold value processing module is used for replacing the amplitude of which the absolute value is smaller than the threshold value in the prediction filtering residual data with zero.
The above-mentioned minor fault detection device, in a preferred embodiment, may be that the coefficient calculation module is configured to calculate a ratio between an average value of root-mean-square amplitude values of seismic traces in the post-prediction seismic data and an average value of root-mean-square amplitude values of seismic traces corresponding to the pre-prediction seismic data;
and the data scaling module is used for multiplying the amplitude of the sampling point of each seismic channel in the predicted pre-seismic data by the coefficient corresponding to the seismic channel calculated by the coefficient calculating module.
The method comprises the steps of firstly carrying out filtering prediction on seismic data in a frequency space domain to obtain predicted seismic data, then subtracting the predicted seismic data from the predicted seismic data to obtain predicted residual data, and identifying micro faults from slices of the predicted residual data. The method is based on the principle that linear homophase axes in a frequency space domain can be predicted, the prediction effect is not ideal in the places where the linear homophase axes in the frequency space domain have the micro faults, when subtraction is carried out on seismic data before and after prediction, the places where the micro faults exist have obvious non-zero residual values, and in the places where the micro faults do not exist, the prediction effect is good, and the theoretical value of the prediction residual is zero. The method of the invention relates the identification of the micro fault with the prediction residual value of the frequency space domain seismic data, embodies the characteristics of the micro fault in the prediction residual data, and then identifies the micro fault by combining the slice analysis of the prediction residual data with the seismic interpretation data and the local actual geological condition.
Drawings
FIG. 1 is a flow chart of a micro-fault detection method provided in embodiment 1 of the present invention;
FIG. 2 is a diagram of a segment of pre-prediction seismic data records in example 1 of the present invention;
FIG. 3 is a graph of a predicted seismic data record of FIG. 2 after prediction filtering in the frequency-space domain in accordance with example 1 of the present invention;
FIG. 4 is a graph of prediction filtered residual data obtained by subtracting the values of FIG. 3 from FIG. 2 in embodiment 1 of the present invention;
FIG. 5 is a graph of the prediction filtered residual data of FIG. 4 after threshold processing according to embodiment 1 of the present invention;
fig. 6 is a schematic block configuration diagram of a fault detection apparatus provided in embodiment 2 of the present invention;
fig. 7 is a schematic block diagram of a frequency-space domain filtering module in the fault detection apparatus according to embodiment 2 of the present invention;
fig. 8 is a schematic structural diagram of an amplitude adjustment module in the fault detection apparatus according to embodiment 2 of the present invention;
fig. 9 is a schematic block diagram of a threshold processing module in the fault detection apparatus according to embodiment 2 of the present invention.
Detailed Description
In order to make those skilled in the art better understand the technical solutions in the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. It is to be understood that the embodiments described are only some of the embodiments in the present application, and not all of them. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Embodiment 1 is a method for detecting a micro fault in the context of high-fidelity seismic data, and fig. 1 is a flow chart of the micro fault detection method provided in embodiment 1 of the invention. As shown in fig. 1, an implementation process of the micro fault detection method includes the following processing steps:
s1: and carrying out high-fidelity data processing on the acquired seismic data records.
And increasing the acquisition density of the seismic data in the field to obtain seismic data records, performing high-fidelity data processing on the acquired seismic data records, and obtaining pre-prediction seismic data records after the high-fidelity data processing.
The high fidelity data processing refers to a series of processing processes for improving the signal-to-noise ratio and the resolution of seismic data. In the actual field-collected seismic data recording, due to the influence of factors such as excitation and receiving factors (different seismic source excitation depths, different shot-geophone distances and the like), stratum layered structures, various interference waves and the like, the signal-to-noise ratio and the resolution of the original seismic data recording are low, and the method is not suitable for subsequent seismic data processing and interpretation. The offset distance is the distance from the demodulator probe to the shot point. The geophone point is the point at which the geophone receiving the data is located during seismic data acquisition. The shot point is the point at which the seismic source that excited the seismic waves at the time of seismic data acquisition is located. The high fidelity data processing is a series of static correction, dynamic correction, deconvolution, amplitude maintenance and other processing processes for improving the signal-to-noise ratio and the resolution of seismic data in order to eliminate interference and errors caused by various non-geological factors as much as possible during seismic data processing. For example, the static correction is a movement amount determined by the static correction amount, and the amplitude discrete value of the seismic trace is moved as a whole, thereby eliminating the influence of surface anomalies such as a ground valley, a surface weathered layer, and a low-speed zone. The dynamic correction is to calculate the dynamic correction value through a dynamic correction formula to eliminate the normal time difference generated when the seismic waves reach different detection points. The seismic signals received by the detector can be regarded as convolution of the reflection coefficient sequence and the seismic wavelets, the resolution can be reduced due to the existence of the wavelets, the deconvolution is to compress the wavelets of the received seismic signals, and finally only the reflection coefficient is reserved on a seismic trace, so that the seismic data resolution is improved. The energy of the seismic waves is influenced by absorption and attenuation in the process of propagation, so that the energy difference of shallow, middle and deep layers is large, and the amplitude is kept by using the technologies of spherical diffusion compensation, residual amplitude compensation and the like to eliminate the energy difference caused by space and time factors in the seismic data record as much as possible.
And performing high-fidelity data processing on the seismic data records acquired in the field to obtain the seismic data records before prediction after the high-fidelity data processing.
S2: and performing predictive filtering on the pre-prediction seismic data after the high-fidelity data processing in a frequency space domain.
The filtering of the seismic data in the frequency space domain is a process for suppressing random noise in the seismic data record, and the filtering method can refer to chinese patent with application number 200410102646.7.
The connection line of the amplitude extreme values (commonly called wave crest or wave trough) with the same vibration phase on each seismic data record is called the homophase axis. In the time space domain, the frequency of the in-phase axis is the same or similar, and when transforming to the frequency space domain, the in-phase axis is approximated as a straight line. When several points on the straight line of the in-phase axis of the frequency space domain are known, the values of other points on the straight line can be predicted by some specific algorithm, and the characteristic that the linear in-phase axis of the seismic data in the frequency space domain can be predicted is called.
The method for suppressing random noise in the seismic records with low signal-to-noise ratio is based on the principle that the linear in-phase axis of the seismic data in the frequency space domain can be predicted. The method for performing predictive filtering on the pre-prediction seismic data in the frequency space domain mainly comprises the following steps of:
s201: transforming the pre-prediction seismic data from a time domain to a frequency space domain through Fourier transform to form space domain vector data with amplitude and direction in the seismic data;
s202: on a single frequency point of the space domain vector data, a predictor is obtained by a wiener filtering method;
s203: on the single-point frequency point of the space domain vector data, multiplying the seismic traces with a total length of one predictor before and after the predicted seismic trace by the corresponding predictors one by one, adding the multiplication results to be the single-point frequency of the predicted seismic trace, completing the prediction of all the single-point frequencies of the frequency domain, and forming the predicted data of the frequency space domain;
s204: and transforming the predicted data of the frequency space domain from the frequency domain to the time domain through inverse Fourier transform to form predicted seismic data after frequency space domain filtering.
And filtering the pre-prediction seismic data subjected to high-fidelity data processing in a frequency space domain to form post-prediction seismic data subjected to frequency space domain filtering.
S3: and adjusting the amplitudes of the post-prediction seismic data and the pre-prediction seismic data after the frequency space domain filtering to the same amplitude level to obtain post-prediction scaling data.
Multiplying each seismic trace in the frequency space domain filtered predicted seismic data of S2 by a coefficient corresponding to the seismic trace, and scaling the root mean square amplitude of each seismic trace sampling point to be consistent with the root mean square amplitude of the original seismic trace sampling point corresponding to the original seismic trace before prediction by the coefficient to obtain frequency space domain predicted filtered scaled data. The processing in the process aims to adjust the amplitude of the seismic data before and after filtering to an energy level, and when the average values of the root mean square amplitudes of the seismic traces corresponding to the pre-prediction seismic data and the post-prediction seismic data to be processed are consistent, the amplitudes of the pre-prediction seismic data and the post-prediction seismic data can be considered to be in the same amplitude level.
Each seismic trace of the post-prediction seismic data has a coefficient corresponding to the post-prediction seismic data, and the coefficient is a ratio of an average value of root-mean-square amplitude values of seismic traces in the pre-prediction seismic data to an average value of root-mean-square amplitude values of the post-prediction seismic data and the corresponding seismic trace. If the amplitude difference value of the seismic channel corresponding to the pre-prediction seismic data and the post-prediction seismic data after the frequency space domain filtering is not changed greatly, that is, the amplitude change error is within the range acceptable by seismic data interpreters, the value of the coefficient is considered to be 1 at this moment, and the post-prediction seismic data can not be zoomed any more practically.
The amplitudes of the pre-prediction seismic data and the post-prediction seismic data are adjusted to the same amplitude level, or each seismic trace in the pre-prediction seismic data may be multiplied by a coefficient corresponding to the seismic trace, where the coefficient is a ratio of an average value of root-mean-square amplitude values of seismic traces in the post-prediction seismic data to an average value of root-mean-square amplitude values of seismic traces corresponding to the pre-prediction seismic data.
And adjusting the amplitude level of the post-prediction seismic data subjected to frequency space domain prediction filtering to be consistent with the amplitude level of the pre-prediction seismic data, and obtaining post-prediction scaling data.
S4: and correspondingly subtracting the amplitudes of the sampling points of the pre-prediction seismic data and the post-prediction scaling data.
And correspondingly subtracting the amplitude of the sampling points of the pre-prediction seismic data and the obtained post-prediction scaling data to obtain prediction filtering residual data. Based on the principle that the frequency space domain linear same-phase axis can be predicted, the prediction effect is not ideal in the place where the frequency space domain linear same-phase axis has the micro fault. When the seismic data before and after prediction are subtracted, the prediction effect is good in places without micro faults, the theoretical value of the prediction residual is zero, and obvious non-zero residual values can appear in places with micro faults. The method of the invention relates the identification of the characteristics of the micro fault to the prediction residual value of the frequency space domain seismic data.
And the amplitude of the sampling points of the pre-prediction seismic data and the post-prediction scaled data is correspondingly subtracted, wherein the amplitude of the sampling points of the pre-prediction seismic data and the amplitude of the sampling points of the post-prediction scaled data can be subtracted by the pre-prediction seismic data or the pre-prediction seismic data.
A threshold value can be set for the prediction filtering residual data, and data with an amplitude absolute value smaller than the threshold value in the prediction filtering residual data is replaced by zero to form prediction filtering residual data after the threshold value is filtered, so that the characteristics of the micro fault in the prediction filtering residual data are more obvious. The threshold value is an empirical value, and a person skilled in the art can select a numerical value which can better embody the characteristics of the micro fault after a plurality of experiments.
And correspondingly subtracting the amplitudes of the sampling points of the pre-prediction seismic data and the post-prediction scaling data to obtain prediction filtering residual data.
S5: and carrying out slice analysis on prediction filtering residual data obtained by correspondingly subtracting the amplitudes of sampling points of the pre-prediction seismic data and the post-prediction zoom data.
And carrying out slice analysis on the prediction filtering residual data so as to identify a micro fault from the slice. A slice is a two-dimensional volume of data having a geophysical significance extracted in a plane or curved surface along one direction of a three-dimensional seismic data volume. A common slice analysis method is horizontal slicing. Horizontal slicing is an iso-planogram, which is a reflection of geological information of different horizons of the subsurface in the same time plane. The features of the fault on the horizontal slice are mainly that the amplitude is regularly distributed, and the features of the prediction filtering residual data caused by the micro fault on the horizontal slice are mainly a section of homophase axis which regularly appears. Micro faults in the seismic data are identified by slice analysis of the prediction filtered residual data.
And carrying out slice analysis on the prediction filtering residual data, finding the characteristics embodied by the micro fault from the slice, and identifying the micro fault.
FIGS. 2 through 5 are processes for identifying micro-faults in high fidelity seismic data recordings using the method of the present invention. Firstly, seismic data records collected in the field are subjected to high fidelity data processing to form pre-prediction seismic data, as shown in fig. 2. As shown in FIG. 2, the horizontal axis represents the seismic trace number of the seismic data record, as shown in the sections 1980 to 2080 of FIG. 2. The numbering of each seismic trace is called the seismic trace number. The vertical axis is the recording time of the seismic data in milliseconds, such as 600 milliseconds to 1500 milliseconds in fig. 2. As shown in FIG. 2, the pre-prediction seismic data includes two distinct in-phase axes, one a horizontal in-phase axis at 1000 millimeters, and the other a tilted in-phase axis from 1020ms to 1500 ms. Two micro-faults are arranged on a horizontal same-phase axis of the pre-prediction seismic data record, two micro-faults are arranged on an inclined same-phase axis, and the four micro-faults are difficult to visually identify from the graph 2 and difficult to identify even if further analysis is carried out. Next, the pre-prediction seismic data in fig. 2 is subjected to prediction filtering processing in the frequency space domain by the filtering method described in S2 in the method of the present invention, and post-prediction seismic data is obtained, as shown in fig. 3. And thirdly, adjusting the amplitudes of the post-prediction seismic data and the pre-prediction seismic data after the frequency space domain filtering to the same amplitude level. Comparing the data in fig. 3 and fig. 2, it can be seen that the amplitude of the post-prediction seismic data in fig. 3 does not change significantly as a whole, and is more balanced than the amplitude of each corresponding seismic trace in the pre-prediction seismic data, so that the post-prediction data amplitude scaling coefficient in fig. 3 in this embodiment 1 is 1. The post-prediction seismic data of FIG. 3 is then subtracted from the pre-prediction seismic data of FIG. 2 to form prediction filtered residual data, as shown in FIG. 4. At A, B, C, D, it can be seen in fig. 4 that there is significant prediction filtered residual data, while there is much prediction filtered residual interference data in the periphery, which may affect the interpreter's discrimination of the micro-faults. Therefore, in the embodiment 1, 6 threshold values are set for fig. 4, the magnitudes are 50, 100, 200, 300, 400, and 500, and the data of fig. 4 are filtered by using the 6 threshold values respectively. The results of the 6 threshold filtered prediction filtered residual data were analyzed by comparison, wherein the characteristics of the micro-faults in the prediction filtered residual data are clearly shown when the threshold value of 400 was chosen. Therefore, 400 is selected as the threshold value in the present embodiment 1. In fig. 4, the amplitudes with absolute amplitude values greater than the threshold 400 are retained, and the amplitude values less than the threshold 400 are replaced with zeros to form the prediction filtered residual data filtered by the threshold 400, as shown in fig. 5. The characteristics of the micro-faults at a1, B1, C1 and D1 can be clearly identified from fig. 5. And finally, carrying out slice analysis on the prediction filtering residual data in the graph 5 so as to identify the micro fault, thereby achieving the purpose of detecting the micro fault by using the method.
Embodiment 2 of the present invention is a fault detection apparatus using the method for detecting a micro fault of the present invention. Fig. 6 is a schematic block diagram of a fault detection apparatus according to embodiment 2 of the present invention. As shown in fig. 6, the fault detection device includes a high fidelity data processing module 1, a frequency space domain filtering module 2, an amplitude adjusting module 3, an amplitude subtracting module 4 and a slice analysis module 5. Wherein,
the high-fidelity data processing module 1 can be used for processing the acquired seismic data to improve the signal-to-noise ratio and resolution of the seismic data and acquiring pre-prediction seismic data;
the frequency space domain filtering module 2 can be used for filtering the pre-prediction seismic data obtained by the high-fidelity data processing module 1 in a frequency space domain to obtain post-prediction seismic data;
the amplitude adjusting module 3 can be used for adjusting the amplitude of the pre-prediction seismic data obtained by the high-fidelity data processing module 1 and the amplitude of the post-prediction seismic data obtained by the frequency space domain filtering module 2 to the same amplitude level to obtain post-prediction zoom data;
the amplitude subtraction module 4 is used for subtracting the amplitude of the seismic channel corresponding to the post-prediction zoom data obtained by the amplitude adjustment module 3 and the pre-prediction seismic data obtained by the high-fidelity data processing module to obtain prediction filtering residual data;
the slice analysis module 5 may be configured to perform slice analysis on the prediction filtering residual data obtained by the amplitude subtraction module.
The frequency-space domain filtering module 2 includes a fourier transform module 201, a wiener filtering module 202, a prediction data module 203, and an inverse fourier transform module 204. Fig. 7 is a schematic structural diagram of the frequency-space domain filtering module 2 in the fault detection apparatus provided in embodiment 2 of the present invention. Wherein,
the Fourier transform module 201 can be used for transforming the pre-prediction seismic data obtained by the high fidelity data module 1 from a time domain to a frequency space domain through Fourier transform to form space domain vector data;
the wiener filtering module 202 may be configured to obtain a predictor by using a wiener filtering method at a single frequency point of the spatial domain vector data;
the prediction data module 203 is configured to multiply seismic traces with a total length of one predictor before and after a predicted seismic trace and corresponding predictors one by one at a single-point frequency point of the spatial domain vector data, and add the multiplication results to serve as a single-point frequency of the predicted seismic trace, so as to complete prediction of all single-point frequencies in a frequency domain, and form prediction data in a frequency spatial domain;
an inverse fourier transform module 204 may be configured to transform the prediction data in the frequency space domain obtained by the prediction data module 203 from the frequency domain to the time domain by an inverse fourier transform.
The amplitude adjustment module 3 includes a coefficient calculation module 301 and a data scaling module 302. Fig. 8 is a schematic structural diagram of the amplitude adjustment module 3 in the fault detection apparatus provided in embodiment 2 of the present invention. Wherein,
the coefficient calculation module 301 may be configured to calculate a ratio of the root mean square amplitude of the seismic traces corresponding to the pre-prediction seismic data and the post-prediction seismic data;
the data scaling module 302 may be configured to multiply the amplitude of each seismic trace in the predicted seismic data by the coefficient corresponding to the seismic trace calculated by the coefficient calculation module 301 to obtain the predicted scaled data.
The amplitude subtraction module 4 may further include a threshold processing module 401, configured to filter interference information in the prediction filtering residual data obtained by the amplitude subtraction module. Fig. 9 is a schematic block diagram of the threshold processing module 401 in the fault detection apparatus according to embodiment 2 of the present invention. As shown in fig. 9, the threshold processing module 401 includes a threshold setting module 4011 and a threshold processing module 4012. Wherein,
a threshold setting module 4011, configured to set a threshold for the prediction filtering residual data obtained by the amplitude subtraction module 4;
the threshold processing module 4012 may be configured to replace, by zero, data whose absolute value of amplitude is smaller than the threshold in the prediction filtering residual data obtained by the amplitude subtraction module 4.
According to the fault detection device provided by the invention, the filtering of the frequency space domain of the seismic data, the amplitude scaling, the threshold value filtering and the like are realized through each functional module to obtain accurate prediction filtering residual data, then the characteristics of the fault are reflected in the prediction filtering residual data, and the fault can be quickly and efficiently detected.
In the micro-fault detection method, the amplitude of the pre-prediction seismic data and the amplitude of the post-prediction scaling data are subtracted correspondingly, and the pre-prediction seismic data may be subtracted from the pre-prediction seismic data or subtracted from the post-prediction seismic data. The invention is not limited in this regard.
In the fault detection apparatus, the amplitude of the pre-prediction seismic data and the amplitude of the post-prediction scaling data in the amplitude subtraction module 4 are subtracted correspondingly, and the pre-prediction seismic data may be subtracted from the pre-prediction seismic data, or the post-prediction seismic data may be subtracted from the pre-prediction seismic data. The invention is not limited in this regard.
According to the detection method and the fault detection device for detecting the micro fault by using the frequency space domain seismic data prediction filtering residual error data, the prediction filtering residual error data is obtained through calculation by a specific processing method, and then slice analysis is carried out on the prediction filtering residual error data.
Claims (6)
1. A micro fault detection method is characterized by comprising the following processing steps:
s1: carrying out high-fidelity data processing on the acquired seismic data record;
s2: performing predictive filtering on the pre-prediction seismic data after the high-fidelity data processing in a frequency space domain, wherein the process of performing predictive filtering in the frequency space domain comprises the following processing steps:
s201: transforming the pre-prediction seismic data from a time domain to a frequency space domain through Fourier transform to form space domain vector data;
s202: on a single frequency point of the space domain vector data, a predictor is obtained by a wiener filtering method;
s203: on the single-point frequency point of the space domain vector data, multiplying the seismic traces with a total length of one predictor before and after the predicted seismic trace by the corresponding predictors one by one, adding the multiplication results to be the single-point frequency of the predicted seismic trace, completing the prediction of all the single-point frequencies of the frequency domain, and forming the predicted data of the frequency space domain;
s204: transforming the predicted data of the frequency space domain from the frequency domain to the time domain by inverse fourier transform;
s3: adjusting the amplitudes of the post-prediction seismic data and the pre-prediction seismic data after the frequency space domain filtering to the same amplitude level to obtain post-prediction scaling data; the process of adjusting the amplitudes of the post-prediction seismic data and the pre-prediction seismic data after the frequency space domain prediction filtering to the same amplitude level comprises the following processing steps:
multiplying each seismic trace in the predicted seismic data after the frequency space domain prediction filtering by a coefficient corresponding to the seismic trace;
the coefficient is the ratio of the average value of the root-mean-square amplitude values of the seismic channels in the seismic data before prediction to the average value of the root-mean-square amplitude values of the seismic channels corresponding to the seismic data after prediction;
s4: correspondingly subtracting the amplitudes of sampling points of the pre-prediction seismic data and the post-prediction scaling data;
s5: and carrying out slice analysis on prediction filtering residual data obtained by correspondingly subtracting the amplitudes of the sampling points of the pre-prediction seismic data and the post-prediction scaling data.
2. The minor fault detection method according to claim 1, further comprising setting a threshold value for prediction filtered residual data obtained by subtracting amplitude correspondence of the pre-prediction seismic data and the post-prediction seismic data in S4, and replacing data having an absolute value of amplitude smaller than the threshold value in the prediction filtered residual data with zero.
3. The minor fault detection method according to claim 1, wherein the step of adjusting the amplitudes of the post-prediction seismic data after the frequency-space domain prediction filtering and the pre-prediction seismic data to the same amplitude level in S3 includes the steps of:
multiplying each seismic trace in the pre-prediction seismic data by a coefficient corresponding to the seismic trace;
the coefficient is the ratio of the average value of the root mean square amplitude values of the seismic traces in the post-prediction seismic data to the average value of the root mean square amplitude values of the pre-prediction seismic data and the corresponding seismic traces.
4. The fault detection device is characterized by comprising a high-fidelity data processing module, a frequency space domain filtering module, an amplitude adjusting module, an amplitude subtracting module and a slice analyzing module; wherein,
the high-fidelity data processing module is used for improving the signal-to-noise ratio and the resolution of the earthquake data;
the frequency space domain filtering module is used for carrying out predictive filtering on the pre-prediction seismic data obtained by the high-fidelity data processing module in a frequency space domain;
the amplitude adjusting module is used for adjusting the amplitude of the pre-prediction seismic data obtained by the high-fidelity data processing module and the amplitude of the post-prediction seismic data obtained by the frequency space domain filtering module to the same amplitude level;
the amplitude subtraction module is used for subtracting the amplitude of the seismic trace corresponding to the post-prediction zooming data obtained by the amplitude adjustment module and the pre-prediction seismic data obtained by the high-fidelity data processing module;
the slice analysis module is used for carrying out slice analysis on the prediction filtering residual error data obtained by the amplitude subtraction module;
the frequency space domain filtering module comprises a Fourier transform module, a wiener filtering module, a prediction data module and an inverse Fourier transform module; wherein,
the Fourier transform module is used for transforming the pre-prediction seismic data obtained by the high-fidelity data module from a time domain to a frequency space domain through Fourier transform to form space domain vector data;
the wiener filtering module is used for solving a prediction operator by using a wiener filtering method on a single frequency point of the space domain vector data;
the prediction data module is used for multiplying seismic traces with a total length of one predictor in front of and behind the predicted seismic trace and the corresponding predictors one by one on a single-point frequency point of the space domain vector data, adding multiplication results to be used as single-point frequencies of the predicted seismic trace, completing prediction of all single-point frequencies of a frequency domain, and forming prediction data of the frequency space domain;
the inverse Fourier transform module is used for transforming the predicted data of the frequency space domain obtained by the predicted data module from the frequency domain to the time domain through inverse Fourier transform;
the amplitude adjusting module comprises a coefficient calculating module and a data scaling module; wherein,
the coefficient calculation module is used for calculating the ratio of the average value of the root-mean-square amplitude values of the seismic channels in the pre-prediction seismic data to the average value of the root-mean-square amplitude values of the post-prediction seismic data and the corresponding seismic channels;
and the data scaling module is used for multiplying the amplitude of the sampling point of each seismic channel in the predicted seismic data by the coefficient corresponding to the seismic channel calculated by the coefficient calculating module.
5. The apparatus of claim 4, wherein the amplitude subtraction module further comprises a threshold processing module for filtering interference information in the prediction filtered residual data obtained by the amplitude subtraction module, and the threshold processing module comprises a threshold setting module and a threshold processing module; wherein,
the threshold value setting module is used for setting a threshold value for the prediction filtering residual error data obtained by the amplitude subtraction module;
and the threshold value processing module is used for replacing the data of which the absolute value of the amplitude is smaller than the threshold value in the prediction filtering residual data obtained by the amplitude subtraction module with zero.
6. A fault detection device according to claim 4, wherein:
the coefficient calculation module is used for calculating the ratio of the average value of the root-mean-square amplitude values of the seismic channels in the post-prediction seismic data to the average value of the root-mean-square amplitude values of the seismic channels corresponding to the pre-prediction seismic data;
and the data scaling module is used for multiplying the amplitude of the sampling point of each seismic channel in the predicted pre-seismic data by the coefficient corresponding to the seismic channel calculated by the coefficient calculating module.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410001435.8A CN103760599B (en) | 2014-01-02 | 2014-01-02 | A kind of miniature fault detection method and fault detection device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410001435.8A CN103760599B (en) | 2014-01-02 | 2014-01-02 | A kind of miniature fault detection method and fault detection device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103760599A CN103760599A (en) | 2014-04-30 |
| CN103760599B true CN103760599B (en) | 2016-08-31 |
Family
ID=50527864
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410001435.8A Active CN103760599B (en) | 2014-01-02 | 2014-01-02 | A kind of miniature fault detection method and fault detection device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103760599B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111239809B (en) * | 2018-11-29 | 2022-08-02 | 中国石油天然气集团有限公司 | Dominant reservoir determination method and system based on attribute fusion principle |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102033242A (en) * | 2010-10-22 | 2011-04-27 | 中国石油化工股份有限公司 | Deep inclined fractured reservoir earthquake amplitude prediction method |
| CN103076623A (en) * | 2011-10-25 | 2013-05-01 | 中国石油化工股份有限公司 | Crack detection method based on prestack coherence |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7561491B2 (en) * | 2005-03-04 | 2009-07-14 | Robinson John M | Radon transformations for removal of noise from seismic data |
-
2014
- 2014-01-02 CN CN201410001435.8A patent/CN103760599B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102033242A (en) * | 2010-10-22 | 2011-04-27 | 中国石油化工股份有限公司 | Deep inclined fractured reservoir earthquake amplitude prediction method |
| CN103076623A (en) * | 2011-10-25 | 2013-05-01 | 中国石油化工股份有限公司 | Crack detection method based on prestack coherence |
Non-Patent Citations (1)
| Title |
|---|
| 《小断层识别技术研究及应用》;王彦君 等;《勘探地球物理进展》;20070430;第30卷(第2期);135-139 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103760599A (en) | 2014-04-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11880011B2 (en) | Surface wave prediction and removal from seismic data | |
| Halliday et al. | Interferometric ground-roll removal: Attenuation of scattered surface waves in single-sensor data | |
| US20130194893A1 (en) | Method and apparatus for processing seismic data | |
| CA2750982C (en) | Method of detecting or monitoring a subsurface hydrocarbon reservoir-sized structure | |
| US9448315B2 (en) | Device and method for denoising ocean bottom data | |
| Al-Hagan et al. | Iterative supervirtual refraction interferometry | |
| Takougang et al. | Characterization of small faults and fractures in a carbonate reservoir using waveform inversion, reverse time migration, and seismic attributes | |
| Economou et al. | Spectral balancing GPR data using time-variant bandwidth in the t-f domain | |
| AU2013201072B2 (en) | Method and apparatus for automated noise removal from seismic data | |
| Zhang et al. | Retrieval of shallow S-wave profiles from seismic reflection surveying and traffic-induced noise | |
| CN104597485B (en) | A kind of miniature fault detection method and fault detection device | |
| US9201154B2 (en) | Device and method for denoising ocean bottom data | |
| Ikeda et al. | Advanced surface-wave analysis for 3D ocean bottom cable data to detect localized heterogeneity in shallow geological formation of a CO2 storage site | |
| Zhang et al. | Separating the scattered wavefield from teleseismic P using curvelets on the long beach array data set | |
| CN104570115B (en) | A kind of surface wave attenuation method and device | |
| CN103760599B (en) | A kind of miniature fault detection method and fault detection device | |
| EP3232234B1 (en) | Method and apparatus for estimating surface wave coda using time-reversal experiments | |
| Mukherjee et al. | S-transform based processing of noisy surface wave record for recovering high-resolution spectrum | |
| Dinh et al. | Processing and quality-control strategies for consistent time-lapse seismic attributes: A case history on an internal blowout using vintage data | |
| CN112526611A (en) | Method and device for extracting surface seismic wave quality factor | |
| CN112799132B (en) | Micro-local linear noise suppression method and device | |
| Nakata et al. | Usage of pre-stack vibroseis data before sweep correlation for noise removal and near-surface structural analysis. | |
| Tylor-Jones et al. | Processing Essentials | |
| Souza et al. | Seismic velocity field estimation using sonic well logs decomposed by Singular Spectrum Analysis (SSA) and Exponentially Weighted Moving Average (EWMA)-A case of study in Recôncavo Basin, Brazil | |
| Panea et al. | 2D versus pseudo-3D seismic data processing in the AVO Analysis-A study case |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |