CN106814402B  Transient electromagnetic signal Prestack Noise Suppression Methods  Google Patents
Transient electromagnetic signal Prestack Noise Suppression Methods Download PDFInfo
 Publication number
 CN106814402B CN106814402B CN201611201090.6A CN201611201090A CN106814402B CN 106814402 B CN106814402 B CN 106814402B CN 201611201090 A CN201611201090 A CN 201611201090A CN 106814402 B CN106814402 B CN 106814402B
 Authority
 CN
 China
 Prior art keywords
 field
 window
 value
 time
 transient electromagnetic
 Prior art date
Links
 230000001052 transient Effects 0 abstract claims description title 31
 230000001629 suppression Effects 0 abstract claims description title 6
 238000001914 filtration Methods 0 abstract claims description 24
 238000000034 methods Methods 0 abstract claims description 15
 230000000737 periodic Effects 0 abstract claims description 8
 238000003672 processing method Methods 0 abstract claims description 5
 238000003860 storage Methods 0 abstract claims description 4
 230000000875 corresponding Effects 0 claims description 13
 238000005070 sampling Methods 0 claims description 11
 239000007787 solids Substances 0 abstract description 6
 230000001976 improved Effects 0 abstract description 4
 230000002829 reduced Effects 0 abstract description 4
 239000000284 extracts Substances 0 claims 1
 230000000694 effects Effects 0 description 10
 238000005516 engineering processes Methods 0 description 10
 230000002123 temporal effects Effects 0 description 4
 239000011162 core materials Substances 0 description 2
 230000018109 developmental process Effects 0 description 2
 239000007789 gases Substances 0 description 2
 BVKZGUZCCUSVTDUHFFFAOYSAL Carbonate dianion Chemical compound data:image/svg+xml;base64,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 data:image/svg+xml;base64,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 [O]C([O])=O BVKZGUZCCUSVTDUHFFFAOYSAL 0 description 1
 238000004458 analytical methods Methods 0 description 1
 230000003935 attention Effects 0 description 1
 230000015572 biosynthetic process Effects 0 description 1
 238000009826 distribution Methods 0 description 1
 238000005225 electronics Methods 0 description 1
 238000000605 extraction Methods 0 description 1
 238000005755 formation Methods 0 description 1
 238000009499 grossing Methods 0 description 1
 230000001965 increased Effects 0 description 1
 239000010410 layers Substances 0 description 1
 230000000670 limiting Effects 0 description 1
 230000014759 maintenance of location Effects 0 description 1
 230000015654 memory Effects 0 description 1
 238000006011 modification Methods 0 description 1
 230000004048 modification Effects 0 description 1
 239000003921 oil Substances 0 description 1
 238000002203 pretreatment Methods 0 description 1
 238000005365 production Methods 0 description 1
 230000004224 protection Effects 0 description 1
 238000010010 raising Methods 0 description 1
 230000004044 response Effects 0 description 1
Classifications

 G—PHYSICS
 G01—MEASURING; TESTING
 G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS
 G01V3/00—Electric or magnetic prospecting or detecting; Measuring magnetic field characteristics of the earth, e.g. declination, deviation
 G01V3/38—Processing data, e.g. for analysis, for interpretation, for correction
Abstract
Description
Technical field
The present invention relates to geophysical signal processing to explore with analytical technology research field more particularly to a kind of transient electromagnetic Time series signal prestack noiseremoved technology.
Background technique
Long offset transient electromagnetic Array Method (LOATEM) as an emerging technology in electromagnetic prospecting, more and more by To the attention of exploration geophysics circle.Since the nineties in last century, domestic scholars are taken to this method and are led in oil exploration Method test and Interpreting Method in domain are successively successfully carried out in Carbonate Rock Areas, Southern China area and Ordos Basin Method test achieves some encouraging as a result, further developing and laying a good foundation for this method.
Although LOATEM method has solid theoretical basis, although it is excellent in terms of the factors such as resolution ratio, signaltonoise ratio In domain electromagnetic exploitation method, but in terms of data processing and interpretation, there are still some very distinct issues.Firstly, since should Method record is time series, and sampled point is intensive (0.1 to 0.5 millisecond of sampling interval), and the period is long (general 8 to 64 seconds), battle array Data volume is huge when column acquisition, and processing is quite heavy with inverting task, and scene cannot generally provide electric section in time, thus shadow The evaluation to the acquisition quality of data and exploration effects is rung.
With the quickening of process of industrialization, though electromagnetic prospecting, in remote mountain area, cultural barriers still remain, and Intolerable stage, needless to say developed regions are arrived.Although LOATEM has, transmission power is big, depth of exploration is deep, anti The features such as interference performance is strong, convenient, flexible, but still there are many measuring point due to being influenced by cultural barriers, signaltonoise ratio is very low, number According to often without method interpretation.Quality is explained in order to improve letter, and " denoising " is the prelimi nary work of electromagnetic method data processing.Currently, There are many method of denoising.At, big, the live superposition of Single time series record data volume high for LOATEM method time sampling density The features such as reason is difficult, the invention proposes a kind of prestack noiseremoved technologies of transient electromagnetic time series signal.
Transient electromagnetic signal has early signal amplitude high, and decaying is fast, and advanced stage signal is weak, and decaying is slow, and dynamic range is big etc. Feature.Conventional FFT method is because it is mainly from the frequency district characteristic analysis of all the period of time, it is difficult to window when to transient electromagnetic difference Time series signal in noise preferably eliminate.Three vertex degrees of some scholars' research and development approach nonlinear smoothing It makes an uproar and the signal denoising technology based on wavelet threshold, achieves certain effect.Transient electromagnetic method improves the side of signaltonoise ratio at present Method mainly by be superimposed and increase transmitting ource electric current come.Since the raising of signaltonoise ratio and the index of stacking fold are directly proportional again, It is thus limited to improve signaltonoise ratio by stacking fold.Improve emission current is influenced by instrument power, device and ground connection, also can only It is a limited means.Carry out signal identification and Study on Extraction Method, under the premise of superposition is with supply current is increased, exploitation New noiseremoved technology is the important channel that current transient electromagnetic method improves data quality.
Summary of the invention
The technical problem to be solved in the present invention is that high, the singlepoint time for prior art LOATEM method time sampling density The defect of big, the live superposition processing difficulty of sequential recording data volume etc., providing one kind may make the processing of LOATEM method data fast Fast effectively signaltonoise ratio is obviously improved, and data distortion is obviously reduced, and data quality significantly improves, and calculates for apparent resistivity curve, high Solid foundation has been established in the fine inverting differentiated.
The technical solution adopted by the present invention to solve the technical problems is:
A kind of Prestack Noise Suppression Methods of the time series signal of transient electromagnetic exploration are provided, comprising the following steps:
S1, storage transient electromagnetic time series；
S2, the filter of inverted order Temporal Recursive is carried out to each periodic attenuation curve of the field component time series of Transient electromagnetic measure Wave；
S3, windowing process is carried out to the field component attenuation curve in period each after filtering, obtains the equally spaced field value of logarithm Attenuation curve；
S4, the statistical stacking processing method for taking given threshold, to multiple in the value attenuation curve of logarithm equally spaced field Same a period of time window field value of observation cycle carries out flying spot and is superimposed.
In method of the present invention, in step S1, the signal after shutdown is only extracted, and by the pass twice in each period The attenuation curve of secondary field having no progeny is average again after being added, and forms an attenuation curve, carries out K period observation in a measuring point When, then it is average after being added for the field value of the same delay time of K periodic attenuation curve.
When carrying out windowing process in method of the present invention, in step S3, the time after shutdown is divided into N number of logarithm etc. Window when interval, the interval width in logdomain is Δ, and the center time point of window is denoted as t when each_{1},t_{2}......t_{N}, corresponding to Field value be calculated as EMF (t respectively_{1}),EMF(t_{2}),.......EMF(t_{N}), N is natural number；Window is divided into 4 parts when will be each, successively It is denoted as t_{i+Δ/4},t_{i+Δ/2},t_{i+3Δ/4},t_{i+Δ}；In order to calculate the field component value EMF (t at center time point_{i}), by when window width expand Greatly to 3 Δ/2, window has the overlapping of Δ/2 when guaranteeing each；
K sampled point when falling into window is subjected to least square fitting；If y is sampling time ts and parameterFunction, (y_{i},ts_{i}) it is K to observation, wherein y_{i}It is observation field values, ts_{i}When being corresponding sampling Between, i=1 ..., K, K, M are natural number；Seek parameterMake objective functionReach minimum, wherein σ_{i}For observation field values y_{i}The standard deviation of error；Window center point when then Corresponding field value isWherein i=1 ..., N, N are natural number.
In method of the present invention, when observation field values points are less than 3 in window at that time, intermediate value or spline interpolation side are taken Field value corresponding to window center point when method is sought.
In method of the present invention, set in step S4 transient electromagnetic field component attenuation curve when window number as N, observation week Issue is M, and the field value in jth of period after adding window is denoted as EMF^{j}(t_{i}), wherein i=1 ..., N；J=1 ..., M, M, N is natural number；And set threshold value as ε, then M period superposition processing is as follows:
1. calculating different cycles with the average value of a period of time window center field of points value:
2. given threshold ε is rejectedField value, wherein j=1 ..., M；
3. window center field of points value when being recalculated to remaining field value by formula in 1..
Beneficial effects of the present invention: the invention enables the processing of LOATEM method data quickly and effectively, and signaltonoise ratio is obviously improved, Data distortion is obviously reduced, and data quality significantly improves, and calculates for apparent resistivity curve, and heavily fortified point has been established in highresolution fine inverting Real basis.
Detailed description of the invention
Present invention will be further explained below with reference to the attached drawings and examples, in attached drawing:
Fig. 1 is Z plane inpolar and O point distribution situation, wherein+it is pole, O is zero point；
Fig. 2 is the amplitudefrequency characteristic of trapper；
Fig. 3 is that 50Hz standard signal sequence is compared with inverted order filter effect；
Fig. 4 is actual measurement dieaway time sequence (Hz noise is serious)；
Fig. 5 is inverted order and positive sequence timerecursive filtering Contrast on effect；
Fig. 6 is the overlapping windowing process of transient electromagnetic field data；
Fig. 7 (a) is that transient electromagnetic magnetic data prestack filters apparent resistivity curve figure；
Fig. 7 (b) is that transient electromagnetic magnetic data poststack filters apparent resistivity curve figure.
Specific embodiment
In order to make the objectives, technical solutions, and advantages of the present invention clearer, with reference to the accompanying drawings and embodiments, right The present invention is further elaborated.It should be appreciated that described herein, specific examples are only used to explain the present invention, not For limiting the present invention.
Prestack processing Method And Principle and process
Due to electric dipole source transient electromagnetic exploitation method have depth of exploration it is big, it is adaptable, multi components observe when pair High resistant thin layer has higher resolution capability, is widely used in oilgas exploration.Currently, record magnanimity electric dipole source transient electromagnetic Depth measurement multi components time series is possibly realized, and is provided solid data for the signaltonoise ratio that Enhanced time sequence data gives processing and is protected Card.
1. transient electromagnetic time series stores journey
Usually power supply wave mode is bipolar square wave, duty ratio 50%, Zhou Wei T.It is only mentioned in the signal that collection terminal receives Signal after taking shutdown.The attenuation curve of secondary field after shutdown twice in each period adds up average production, forms one and declines Subtract curve (note: every secondary extinction curve is added with attenuation of the first kind curve divided by 2 multiplied by negative sign).In this way, a measuring point into It is that the field value of the same delay time of K periodic attenuation curve is added up and is averaged when the observation of K period of row.For folded Pretreatment, it is not necessary to which retention time sequence data, only constantly superposition, retains last stack result.
2. prestack processing method and technology
(1) prestack inverted order timerecursive filtering
Basic principle
If the time series recorded in YUTEM sampling time section is { x_{i}, i=1,2 ..., N, filtering factor h_{i}, then filter Time series { y after wave_{i}Are as follows:
In actual operation, filtering factor h_{j}Finite term can only be taken, this just unavoidably generates error, misses to reduce Difference, h_{j}It takes into finite term but number wants quite a lot of, maxitem N is 4800 in YUTEM mode, and such operation, which is got up, both to be occupied Memory is big, and Fei Jishi, and recursive filtering can well solve this problem.
The thought of recursive filtering is: thinking output valve y_{i}Between it is associated with each other, therefore calculate y_{i}When, it to utilize pervious Calculated result y_{i1},y_{i2}....According to this thought, conventional recursive filtering formula are as follows:
y_{i}=a_{0}x_{i}+a_{1}x_{i1}+…+a_{n}x_{in}(b_{1}y_{i1}+b_{2}y_{i2}+…+b_{m}y_{im}) (2)
Wherein m, n are natural number, a_{j}、b_{j}For recursive filtering parameter.
Z plane method designs filter
What LOATEM measuring signal suffered from is the interference of certain fixed frequencies, thus the design of filter should be directed to The trapper of some fixed frequency.Temporal Recursive trapper is designed with Z plane method with can be convenient.Socalled Z plane method design Filter exactly selects 0 point appropriate and pole on Z plane, designs according to the requirement to filter to amplitude and phase Filter.According to the basic theories of transform, the transform of the amplitudefrequency characteristic of temporal recursive filter can be written as
Z plane method is that have to fall into amplitudefrequency function corresponding to the point using 0 point on Unit Circle and the pole near it Wave property (as depicted in figs. 1 and 2) quickly determines W (z), i.e., inf_{r},f_{r}Amplitude should be 0 at point, and except this two o'clock Amplitude is 1 (normalized value).Meet abovementioned two condition, then requires amplitude function that must contain and pole, and 0 simultaneously at 0 point Point and pole should be very close to.For this purpose, (3) formula can be written as
W (z) is quickly determined below by Z plane method.
The plural form of zero point and pole in Z plane can be written as
z_{z}=cos Ω_{r}±sinΩ_{r}=R_{z}±iI_{z} (5)
z_{p}=r_{p} cosΩ_{p}±r_{p} sinΩ_{p}=R_{p}±iI_{p} (6)
Wherein r_{p}For filter factor, Ω_{r}For the angular frequency for wanting trap, mathematic(al) representation are as follows:
Ω_{r}=± 180 ° of f_{r}/f_{N}
f_{r}For the frequency for wanting trap, f_{N}For nyquist frequency, had according to sampling thheorem:
f_{N}=1/ (2 Δ t)
(5), (6) formula are substituted into (4) formula and by simple mathematical operation, then the frequency response of trapper can be written as
G is by nyquist frequency ω_{N}When=π, enables W (z)=1 and obtains, it may be assumed that
According to the property of recursion filter, we write out the mathematic(al) representation of temporal recursive filter in which can be convenient:
? Sequence filtering
In practical filtering, first time series data is fallen to arrange, then carry out abovementioned filtering processing, then fall again Row obtains last filter result, this ensure that the filter quality of time series front end data, because of the transient signal after shutdown It is most important with inverting to data processing.By taking the filtering of 50Hz sine wave as an example, as shown in Figure 3.Black dotted line is 50Hz sine wave letter Number, for red line to use (8) formula filter effect under the conditions of positive sequence, blue line is inverted order filter effect.As can be seen that the leading portion of positive sequence filtering There is one section of resistance decaying, signal obviously distorts, and effect is poor, and the leading portion data filtering effect for arranging filtering is very ideal.It is based on Inverted order filtering technique, we handle real data, and as shown in figs. 4 and 5, inverted order filters when making leading portion treatment effect Between sequence signaltonoise ratio greatly improve, and ensure that the authenticity of signal.
(2) it is overlapped windowing process
After field component attenuation curve in each period is filtered, it is desirable that window curve when taking logdomain, formation pair Number field at equal intervals when window data.In order to guarantee that curve is continuous, smooth, we take overlapping windowing process.Its Method And Principle such as Fig. 6 It is shown.
Window when time after shutdown is divided into N number of logarithm at equal intervals, the interval width in logdomain are Δ, center time point It is denoted as t_{1},t_{2}......t_{N}, corresponding to field value be calculated as EMF (t respectively_{1}),EMF(t_{2}),..EMF(t_{N}).Window point when will be each At 4 parts, it is successively denoted as t_{i+Δ/4},t_{i+Δ/2},t_{i+3Δ/4},t_{i+Δ}.In order to calculate the field component value EMF (t at center time point_{i}), it will When window width be expanded to 3 Δ/2, window has the overlapping of Δ/2 when guaranteeing each in this way, as shown in Figure 3.
K sampled point when falling into window is subjected to least square fitting.If y is sampling time ts and parameterFunction, (y_{i},ts_{i}), (i=1 ..., K) is K to observation, wherein y_{i}It is observation field values, ts_{i}It is the corresponding sampling time.Seek parameterMake objective function
Reach minimum, wherein σ_{i}For observation field values y_{i}The standard deviation of error.Field value corresponding to window center point is when then
When observation field values points are less than 3 in window at that time, window center point when intermediate value or Spline Interpolation Method can be taken to seek Corresponding field value.
(3) the statistical stacking processing of given threshold
If the when window number of transient electromagnetic field component attenuation curve is N, observation cycle number is M, in jth of period after adding window Field value be denoted as EMF^{j}(t_{i}), (i=1 ..., N；J=1 ..., M), and set threshold value as ε, then M period superposition processing By as follows:
1. calculating different cycles with the average value of a period of time window center field of points value:
2. given threshold ε is rejectedField value.
3. to remaining field value by 1. formula recalculates when window center field of points value.
Fig. 7 (a) and 7 (b) is singlepoint prestack and poststack data processing apparent resistivity curve.As can be seen that the view electricity of prestack filtering Resistance rate curve quality significantly improves.
Large power long offset distance transient electromagnetic array exploitation method of the invention is a kind of for deep prospecting, especially oily Electromagnetic exploration method in gas exploration.It, can with the fast development of electronics and computer technology and instrument weak signal acquisition technique Simultaneously mass storage multi components multiple tracks observation field values time series data, for this method prestack signal processing provide it is solid Data guarantee.This technology invention first carries out each periodic attenuation curve of the field component time series of Transient electromagnetic measure Inverted order timerecursive filtering preferably eliminates periodic cultural barriers, and has higher fidelity, and core technology is down Sequence filtering.Secondly, carry out obtaining the equally spaced attenuation curve of logarithm to each period field component attenuation curve windowing process, Window center field of points value when window overlapping is solved with nonlinear least square method when core technology is.Finally take multiple weeks of given threshold Phase statistical stacking processing technique carries out flying spot superposition processing to same a period of time window field component value of multiple observation cycles, mentions significantly The high precision of field value estimation.
The invention enables the processing of LOATEM method data quickly and effectively, and signaltonoise ratio is obviously improved, and data distortion is obviously reduced, Data quality significantly improves, and calculates for apparent resistivity curve, and solid foundation has been established in highresolution fine inverting.
It should be understood that for those of ordinary skills, it can be modified or changed according to the above description, And all these modifications and variations should all belong to the protection domain of appended claims of the present invention.
Claims (2)
Priority Applications (1)
Application Number  Priority Date  Filing Date  Title 

CN201611201090.6A CN106814402B (en)  20161222  20161222  Transient electromagnetic signal Prestack Noise Suppression Methods 
Applications Claiming Priority (1)
Application Number  Priority Date  Filing Date  Title 

CN201611201090.6A CN106814402B (en)  20161222  20161222  Transient electromagnetic signal Prestack Noise Suppression Methods 
Publications (2)
Publication Number  Publication Date 

CN106814402A CN106814402A (en)  20170609 
CN106814402B true CN106814402B (en)  20190823 
Family
ID=59109601
Family Applications (1)
Application Number  Title  Priority Date  Filing Date 

CN201611201090.6A CN106814402B (en)  20161222  20161222  Transient electromagnetic signal Prestack Noise Suppression Methods 
Country Status (1)
Country  Link 

CN (1)  CN106814402B (en) 
Families Citing this family (1)
Publication number  Priority date  Publication date  Assignee  Title 

CN109343134A (en) *  20181127  20190215  中煤科工集团西安研究院有限公司  A kind of Transient Electromagnetic Method in Mine data analysis interpretation method and system 
Citations (4)
Publication number  Priority date  Publication date  Assignee  Title 

CN101825665A (en) *  20100528  20100908  桂林电子科技大学  Method for detecting stochastic resonance transient electromagnetic weak signals 
CN101882964A (en) *  20100612  20101110  桂林电子科技大学  Denoising method of transient electromagnetic detecting echo signal 
CN104793253A (en) *  20150422  20150722  吉林大学  Airborne electromagnetic data denoising method based on mathematical morphology 
CN105629317A (en) *  20160408  20160601  中国矿业大学(北京)  Magnetotelluric noise suppressing method based on intersite transfer function 
Family Cites Families (1)
Publication number  Priority date  Publication date  Assignee  Title 

US10139517B2 (en) *  20141219  20181127  Baker Huges, A Ge Company Llc  Hybrid image of earth formation based on transient electromagnetc measurements 

2016
 20161222 CN CN201611201090.6A patent/CN106814402B/en active IP Right Grant
Patent Citations (4)
Publication number  Priority date  Publication date  Assignee  Title 

CN101825665A (en) *  20100528  20100908  桂林电子科技大学  Method for detecting stochastic resonance transient electromagnetic weak signals 
CN101882964A (en) *  20100612  20101110  桂林电子科技大学  Denoising method of transient electromagnetic detecting echo signal 
CN104793253A (en) *  20150422  20150722  吉林大学  Airborne electromagnetic data denoising method based on mathematical morphology 
CN105629317A (en) *  20160408  20160601  中国矿业大学(北京)  Magnetotelluric noise suppressing method based on intersite transfer function 
NonPatent Citations (1)
Title 

瞬变电磁信号降噪算法;罗倩 等;《计算机仿真》;20130630;第30卷(第6期);第210213页 
Also Published As
Publication number  Publication date 

CN106814402A (en)  20170609 
Similar Documents
Publication  Publication Date  Title 

Yuan et al.  Lithospheric and upper mantle structure of southern Tibet from a seismological passive source experiment  
Chen et al.  Seismic attribute technology for reservoir forecasting and monitoring  
TorresVerdin et al.  Principles of spatial surface electric field filtering in magnetotellurics: Electromagnetic array profiling (EMAP)  
US4544892A (en)  Signal processing apparatus for frequency domain geophysical electromagnetic surveying system  
Kodera et al.  A new method for the numerical analysis of nonstationary signals  
US7502690B2 (en)  System and method for using timedistance characteristics in acquisition, processing, and imaging of tCSEM data  
AU2013257511B2 (en)  Systems and methods for removal of swell noise in marine electromagnetic surveys  
Taner et al.  Application of amplitude, frequency, and other attributes to stratigraphic and hydrocarbon determination: Section 2. Application of seismic reflection configuration to stratigraphic interpretation  
Cantwell et al.  Preliminary report on crustal magnetotelluric measurements  
Cummer  Lightning and ionospheric remote sensing using VLF/ELF radio atmospherics  
Wang  Multichannel matching pursuit for seismic trace decomposition  
James  Whistler‐mode hiss at low and medium frequencies in the dayside‐cusp ionosphere  
Effersø et al.  Inversion of band‐limited TEM responses  
CN1552035A (en)  Method of imaging subsurface formations using a virtual source array  
Archambeau et al.  Data processing techniques for the detection and interpretation of teleseismic signals  
BRPI0618185B1 (en)  hydrocarbon reservoir mapping method and equipment for use in performing the method  
Beamish  Characteristics of nearsurface electrokinetic coupling  
AU2008242961A1 (en)  Methods of hydrocarbon detection using spectral energy analysis  
Hoover et al.  Some results from audiomagnetotelluric investigations in geothermal areas  
AU2005330955B2 (en)  True amplitude transient electromagnetic system response measurement  
Wang et al.  Timefrequency analysis of seismic data using synchrosqueezing transform  
CN102004264B (en)  A kind of earthquakecapturing data quality quantitative test and evaluation method  
US4633447A (en)  Response waveform characterization of geophysical data  
Zhang et al.  Angle gathers from reverse time migration  
CN103376464B (en)  A kind of inversion method for stratigraphic quality factor 
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 