CN104422970A - Amplitude-preserving static correcting method for electromagnetic data - Google Patents

Abstract

The invention belongs to the technology of electromagnetic exploitation data processing. The technology comprises the following steps of collecting electromagnetic data, determining an electric layer with stable depth and resistivity, averaging two branches of apparent resistivity curves of all detecting points on an electromagnetic data detecting line on a corresponding frequency range, and calculating average apparent resistivity data; obtaining the primary correction factors of two branches of curves of each detecting point, correcting the primary correction factors to obtain data of the two branches of apparent resistivity curve, selecting a frequency range again, respectively calculating the average values of apparent resistivity of the two branches of curves of each detecting point within the frequency range before filtration and after filtration, obtaining amplitude-preserving correction factors of the two branches of curves of each detecting point, and calculating to obtain the apparent resistivity data of the two branches of curves after final amplitude-preserving correction. The defect that the configuration of the apparent resistivity curves is changed after the traditional filtration method is used can be overcome, particularly at the places, such as a mountain front and an earth surface lithologic character mutation region, the effect is obviously improved, and the aim of keeping amplitude is fulfilled.

Description

A kind of guarantor's width static correcting method of electromagnetic data

Technical field

The present invention relates to geophysical exploration method, is a kind of electromagnetic survey data, comprises the data processing technique of mt (MT), audio magnetotelluric method (AMT) and continuous profiling mt (CEMP).

Background technology

Static relays in electromagnetic prospecting refer to due to closely show to exist local electrically heterogeneous body time, on resistivity interphase, the heap of polarization charge causes the distortion of electric field, produces an additional electric field be directly proportional to external electric field thus, and irrelevant with frequency.For magnetotelluric data (MT), show on single-point curve, be exactly that log-domain resistivity curve produces translation along the longitudinal axis, phase curve is substantially unaffected.Show on apparent resistivity section diagram, be exactly that upright abrupt change band appears in resistivity value, be commonly called as " vermicelli bar " phenomenon, and corresponding phase place pseudosection map does not just have this phenomenon.Showing on the apparent resistivity planimetric map of single-frequency point, is exactly there is a lot " distortional point " phenomenon.The intensity of static relays can reach two or three orders of magnitude, can cause very big error, and making structure elucidation complicated, therefore needing to carry out static correction before inversion interpretation when inferring the degree of depth.

At present, the method of static correction is a lot, in order to suppress static relays, improves constantly the effect of MT, lot of domestic and international expert has delivered correlative theses, proposes filter method, theoretical calculation, transient electromagnetic correction method, subsurface resistivities correction method, high-frequency electromagnetic function upward continuation method etc.These method major parts are all based on two-dimentional MT static correction, and the principle of various bearing calibration is different with to the requirement of data, are all subject to restriction to a certain extent in actual applications.Because filter method is easy to realize, processing speed is fast, is most widely used in the static corrections processing of MT/CEMP data.But after filtering after method process, the apparent resistivity curve form of a lot of measuring point can be changed, especially in local performances such as mountain front, Lithology sudden change bands obviously, in practice, static correction effect is bad.

Summary of the invention

The present invention is directed to and adopt traditional filter method to carry out the bad defect of effect after electromagnetic data static correction, provide a kind of can overcome apparent resistivity curve form can reformed defect, reach guarantor's width static correcting method of the object electromagnetic data of guarantor's width.

The present invention adopts following steps to realize:

1) according to the magnetotelluric data that work area gathers, determine a degree of depth and the stable electric layers of resistivity, obtain frequency range corresponding on electromagnetic data curve, do on average to two apparent resistivity curves of measuring points all on electromagnetic data survey line in frequency range, calculate Average apparent resistivity data

Described electromagnetic data comprises magnetotelluric data (MT), audio-frequency magnetotelluric magnetic data (AMT), continuous profiling magnetotelluric data (CEMP).

Described on average comprises arithmetic mean and geometric mean.

Described calculating Average apparent resistivity data be:

Wherein, be respectively i-th measuring point jth frequency f _jxY and YX pattern measured visual resistivity value, n and m is selected frequency point number.

2) following formula is utilized to ask for the preliminary corrections factor of each measuring point two curves;

$k_i^{\mathrm{XY}}=\log {\mathrm{\ρ}}_i^a-\log \frac{\underset{j=n}{\overset{m}{\mathrm{\Σ}}}{\mathrm{\ρ}}_i^{\mathrm{XY}}\left(f_j\right)}{m-n+1},$ $k_i^{\mathrm{YX}}=\log {\mathrm{\ρ}}_i^a-\log \frac{\underset{j=n}{\overset{m}{\mathrm{\Σ}}}{\mathrm{\ρ}}_i^{\mathrm{YX}}\left(f_j\right)}{m-n+1}---\left(5\right)$

Wherein, with be respectively the preliminary corrections factor of i-th measuring point XY and YX mode data, the calculating mean value that (4) formula of serving as reasons obtains, the implication of other parameter is identical with (4) formula.

3) carry out preliminary corrections according to following formula, obtain two apparent resistivity curve data after correcting with

${\mathrm{\ρ}}_{\mathrm{si}}^{\mathrm{XY}}=10**(\log {\mathrm{\ρ}}_i^{\mathrm{XY}}+k_i^{\mathrm{XY}}),$ ${\mathrm{\ρ}}_{\mathrm{si}}^{\mathrm{YX}}=10**(\log {\mathrm{\ρ}}_i^{\mathrm{YX}}+k_i^{\mathrm{YX}})---\left(6\right)$

Right again carry out filtering respectively, obtain filtered apparent resistivity curve data with the implication of other parameter and (4), (5) formula are identical.

Described filtering comprises two dimension median filter, two-dimensional space filtering, three-dimensional medium filtering and three dimensions filtering.

4) frequency band is reselected, according to step 1) to 3) calculate before filtering respectively and the arithmetic of each measuring point two curves apparent resistivity in this band limits or geometrical mean after filtering with

${\mathrm{\ρ}}_{\mathrm{si}}^{\mathrm{aXY}}=\frac{\underset{j=\mathrm{nf}1}{\overset{\mathrm{nf}2}{\mathrm{\Σ}}}{\mathrm{\ρ}}_{\mathrm{si}}^{\mathrm{XY}}\left(f_j\right)}{\mathrm{nf}2-\mathrm{nf}1+1},$ ${\mathrm{\ρ}}_{\mathrm{si}}^{\mathrm{aYX}}=\frac{\underset{j=\mathrm{nf}1}{\overset{\mathrm{nf}2}{\mathrm{\Σ}}}{\mathrm{\ρ}}_{\mathrm{si}}^{\mathrm{YX}}\left(f_j\right)}{\mathrm{nf}2-\mathrm{nf}1+1}---\left(7\right)$

${\mathrm{\ρ}}_{\mathrm{fsi}}^{\mathrm{aXY}}=\frac{\underset{j=\mathrm{nf}1}{\overset{\mathrm{nf}2}{\mathrm{\Σ}}}{\mathrm{\ρ}}_{\mathrm{fsi}}^{\mathrm{XY}}\left(f_j\right)}{\mathrm{nf}2-\mathrm{nf}1+1},$ ${\mathrm{\ρ}}_{\mathrm{fsi}}^{\mathrm{aYX}}=\frac{\underset{j=\mathrm{nf}1}{\overset{\mathrm{nf}2}{\mathrm{\Σ}}}{\mathrm{\ρ}}_{\mathrm{fsi}}^{\mathrm{YX}}\left(f_j\right)}{\mathrm{nf}2-\mathrm{nf}1+1}---\left(8\right)$

Wherein, nf1 and nf2 is respectively selected frequency point number, be respectively a si measuring point jth frequency f _jxY and YX pattern measured visual resistivity value, be respectively apparent resistivity arithmetic or the geometrical mean of XY and YX pattern before the filtering of calculating, with be respectively apparent resistivity arithmetic or the geometrical mean of XY and YX pattern after the filtering of calculating.

5) guarantor's width correction factor of each measuring point two curves is asked for according to following formula:

$k_{\mathrm{fi}}^{\mathrm{XY}}=\log {\mathrm{\ρ}}_{\mathrm{si}}^{\mathrm{aXY}}-\log {\mathrm{\ρ}}_{\mathrm{fsi}}^{\mathrm{aXY}},$ $k_{\mathrm{fi}}^{\mathrm{YX}}=\log {\mathrm{\ρ}}_i^{\mathrm{aYX}}-\log {\mathrm{\ρ}}_{\mathrm{fsi}}^{\mathrm{aYX}}---\left(9\right)$

Wherein, with be respectively guarantor's width correction factor of XY and the YX curve of i-th measuring point.

6) obtain final guarantor's width according to following formulae discovery and correct rear two curve apparent resistivity datas with complete static correction.

${\mathrm{\ρ}}_{\mathrm{psi}}^{\mathrm{XY}}=10**(\log {\mathrm{\ρ}}_{\mathrm{si}}^{\mathrm{XY}}+k_{\mathrm{fi}}^{\mathrm{XY}}),$ ${\mathrm{\ρ}}_{\mathrm{psi}}^{\mathrm{YX}}=10**(\log {\mathrm{\ρ}}_{\mathrm{si}}^{\mathrm{YX}}+k_{\mathrm{fi}}^{\mathrm{YX}})---\left(10\right)$

Wherein, with the guarantor's width being exactly final XY and YX curve corrects rear apparent resistivity data.

Compared with background technology, the present invention can to overcome after the traditional filtering method of application apparent resistivity curve form can reformed defect, especially in places such as mountain front, Lithology sudden change bands, obviously effect improved, reaches the object of guarantor's width.

Accompanying drawing explanation

Fig. 1 is then for electromagnetic data protects width static correcting method basic flow sheet.

Embodiment

Be below the specific embodiment of the invention, Fig. 1 is the process flow diagram protecting the enforcement of width static correcting method.

Concrete grammar is:

1) according to the magnetotelluric data that work area gathers, determine a degree of depth and the stable electric layers of resistivity, obtain frequency range corresponding on electromagnetic data curve, as chosen No. 5 to No. 9 frequency band, then n=5, m=9 in formula (4).Do on average to two apparent resistivity curves of measuring points all on electromagnetic data survey line in frequency range, calculate Average apparent resistivity data

Described electromagnetic data comprises magnetotelluric data (MT), audio-frequency magnetotelluric magnetic data (AMT), continuous profiling magnetotelluric data (CEMP).

Described on average comprises arithmetic mean and geometric mean.

Described calculating Average apparent resistivity data be:

Wherein, be respectively i-th measuring point jth frequency f _jxY and YX pattern measured visual resistivity value, n and m is selected frequency point number, is respectively 5 and 9 in the present example.

2) following formula is utilized to ask for the preliminary corrections factor of each measuring point two curves;

$k_i^{\mathrm{XY}}=\log {\mathrm{\ρ}}_i^a-\log \frac{\underset{j=n}{\overset{m}{\mathrm{\Σ}}}{\mathrm{\ρ}}_i^{\mathrm{XY}}\left(f_j\right)}{m-n+1},$ $k_i^{\mathrm{YX}}=\log {\mathrm{\ρ}}_i^a-\log \frac{\underset{j=n}{\overset{m}{\mathrm{\Σ}}}{\mathrm{\ρ}}_i^{\mathrm{YX}}\left(f_j\right)}{m-n+1}---\left(5\right)$

Wherein, with be respectively the preliminary corrections factor of i-th measuring point XY and YX mode data, the calculating mean value that (4) formula of serving as reasons obtains, the implication of other parameter is identical with (4) formula.

3) carry out preliminary corrections according to following formula, obtain two apparent resistivity curve data after correcting with

${\mathrm{\ρ}}_{\mathrm{si}}^{\mathrm{XY}}=10**(\log {\mathrm{\ρ}}_i^{\mathrm{XY}}+k_i^{\mathrm{XY}}),$ ${\mathrm{\ρ}}_{\mathrm{si}}^{\mathrm{YX}}=10**(\log {\mathrm{\ρ}}_i^{\mathrm{YX}}+k_i^{\mathrm{YX}})---\left(6\right)$

Right again carry out filtering respectively, as in the present example, adopt two dimension median filter, medium filtering window selection is (5,3), obtains filtered apparent resistivity curve data with the implication of other parameter and (4), (5) formula are identical.

Described filtering comprises two dimension median filter, two-dimensional space filtering, three-dimensional medium filtering and three dimensions filtering.

4) frequency band is reselected, if No. 7 frequency is to No. 13 frequency band, namely the nf1=7 in formula (7) and formula (8), nf2=13, according to step 1) to 3) calculate before filtering respectively and the arithmetic of each measuring point two curves apparent resistivity in this band limits or geometrical mean after filtering with

${\mathrm{\ρ}}_{\mathrm{si}}^{\mathrm{aXY}}=\frac{\underset{j=\mathrm{nf}1}{\overset{\mathrm{nf}2}{\mathrm{\Σ}}}{\mathrm{\ρ}}_{\mathrm{si}}^{\mathrm{XY}}\left(f_j\right)}{\mathrm{nf}2-\mathrm{nf}1+1},$ ${\mathrm{\ρ}}_{\mathrm{si}}^{\mathrm{aYX}}=\frac{\underset{j=\mathrm{nf}1}{\overset{\mathrm{nf}2}{\mathrm{\Σ}}}{\mathrm{\ρ}}_{\mathrm{si}}^{\mathrm{YX}}\left(f_j\right)}{\mathrm{nf}2-\mathrm{nf}1+1}---\left(7\right)$

${\mathrm{\ρ}}_{\mathrm{fsi}}^{\mathrm{aXY}}=\frac{\underset{j=\mathrm{nf}1}{\overset{\mathrm{nf}2}{\mathrm{\Σ}}}{\mathrm{\ρ}}_{\mathrm{fsi}}^{\mathrm{XY}}\left(f_j\right)}{\mathrm{nf}2-\mathrm{nf}1+1},$ ${\mathrm{\ρ}}_{\mathrm{fsi}}^{\mathrm{aYX}}=\frac{\underset{j=\mathrm{nf}1}{\overset{\mathrm{nf}2}{\mathrm{\Σ}}}{\mathrm{\ρ}}_{\mathrm{fsi}}^{\mathrm{YX}}\left(f_j\right)}{\mathrm{nf}2-\mathrm{nf}1+1}---\left(8\right)$

Wherein, nf1 and nf2 is respectively selected frequency point number, is respectively 7 and 13 in the present example, be respectively a si measuring point jth frequency f _jxY and YX pattern measured visual resistivity value, be respectively apparent resistivity arithmetic or the geometrical mean of XY and YX pattern before the filtering of calculating, with be respectively apparent resistivity arithmetic or the geometrical mean of XY and YX pattern after the filtering of calculating.

5) guarantor's width correction factor of each measuring point two curves is asked for according to (9) with the formula:

$k_{\mathrm{fi}}^{\mathrm{XY}}=\log {\mathrm{\ρ}}_{\mathrm{si}}^{\mathrm{aXY}}-\log {\mathrm{\ρ}}_{\mathrm{fsi}}^{\mathrm{aXY}},$ $k_{\mathrm{fi}}^{\mathrm{YX}}=\log {\mathrm{\ρ}}_i^{\mathrm{aYX}}-\log {\mathrm{\ρ}}_{\mathrm{fsi}}^{\mathrm{aYX}}---\left(9\right)$

Wherein, with be respectively guarantor's width correction factor of XY and the YX curve of i-th measuring point.

6) calculate the final guarantor's width of acquisition according to formula (10) and correct rear two curve apparent resistivity datas with complete and protect width static correction.

${\mathrm{\ρ}}_{\mathrm{psi}}^{\mathrm{XY}}=10**(\log {\mathrm{\ρ}}_{\mathrm{si}}^{\mathrm{XY}}+k_{\mathrm{fi}}^{\mathrm{XY}}),$ ${\mathrm{\ρ}}_{\mathrm{psi}}^{\mathrm{YX}}=10**(\log {\mathrm{\ρ}}_{\mathrm{si}}^{\mathrm{YX}}+k_{\mathrm{fi}}^{\mathrm{YX}})---\left(10\right)$

Wherein, with the guarantor's width being exactly final XY and YX curve corrects rear apparent resistivity data.

Claims (5)

1. guarantor's width static correcting method of electromagnetic data, feature adopts following steps to realize:

1) according to the magnetotelluric data that work area gathers, determine a degree of depth and the stable electric layers of resistivity, obtain frequency range corresponding on electromagnetic data curve, do on average to two apparent resistivity curves of measuring points all on electromagnetic data survey line in frequency range, calculate Average apparent resistivity data

2) following formula is utilized to ask for the preliminary corrections factor of each measuring point two curves;

$k_i^{\mathrm{XY}}=\log {\mathrm{\ρ}}_i^a-\log \frac{\underset{j=n}{\overset{m}{\mathrm{\Σ}}}{\mathrm{\ρ}}_i^{\mathrm{XY}}\left(f_j\right)}{m-n+1},$ $k_i^{\mathrm{YX}}=\log {\mathrm{\ρ}}_i^a-\log \frac{\underset{j=n}{\overset{m}{\mathrm{\Σ}}}{\mathrm{\ρ}}_i^{\mathrm{YX}}\left(f_j\right)}{m-n+1}---\left(5\right)$

Wherein, with be respectively the preliminary corrections factor of i-th measuring point XY and YX mode data, the calculating mean value that (4) formula of serving as reasons obtains, the implication of other parameter is identical with (4) formula;

3) carry out preliminary corrections according to following formula, obtain two apparent resistivity curve data after correcting with

${\mathrm{\ρ}}_{\mathrm{si}}^{\mathrm{XY}}=10**(\log {\mathrm{\ρ}}_i^{\mathrm{XY}}+k_i^{\mathrm{XY}}),$ ${\mathrm{\ρ}}_{\mathrm{si}}^{\mathrm{YX}}=10**(\log {\mathrm{\ρ}}_i^{\mathrm{YX}}+k_i^{\mathrm{YX}})---\left(6\right)$

Right again carry out filtering respectively, obtain filtered apparent resistivity curve data with the implication of other parameter and (4), (5) formula are identical;

4) frequency band is reselected, according to step 1) to 3) calculate before filtering respectively and the arithmetic of each measuring point two curves apparent resistivity in this band limits or geometrical mean after filtering with

${\mathrm{\ρ}}_{\mathrm{si}}^{\mathrm{aXY}}=\frac{\underset{j=\mathrm{nf}1}{\overset{\mathrm{nf}2}{\mathrm{\Σ}}}{\mathrm{\ρ}}_{\mathrm{si}}^{\mathrm{XY}}\left(f_j\right)}{\mathrm{nf}2-\mathrm{nf}1+1},$ ${\mathrm{\ρ}}_{\mathrm{si}}^{\mathrm{aYX}}=\frac{\underset{j=\mathrm{nf}1}{\overset{\mathrm{nf}2}{\mathrm{\Σ}}}{\mathrm{\ρ}}_{\mathrm{si}}^{\mathrm{YX}}\left(f_j\right)}{\mathrm{nf}2-\mathrm{nf}1+1}---\left(7\right)$

${\mathrm{\ρ}}_{\mathrm{fsi}}^{\mathrm{aXY}}=\frac{\underset{j=\mathrm{nf}1}{\overset{\mathrm{nf}2}{\mathrm{\Σ}}}{\mathrm{\ρ}}_{\mathrm{fsi}}^{\mathrm{XY}}\left(f_j\right)}{\mathrm{nf}2-\mathrm{nf}1+1},$ ${\mathrm{\ρ}}_{\mathrm{fsi}}^{\mathrm{aYX}}=\frac{\underset{j=\mathrm{nf}1}{\overset{\mathrm{nf}2}{\mathrm{\Σ}}}{\mathrm{\ρ}}_{\mathrm{fsi}}^{\mathrm{YX}}\left(f_j\right)}{\mathrm{nf}2-\mathrm{nf}1+1}---\left(8\right)$

Wherein, nf1 and nf2 is respectively selected frequency point number, be respectively a si measuring point jth frequency f _jxY and YX pattern measured visual resistivity value, be respectively apparent resistivity arithmetic or the geometrical mean of XY and YX pattern before the filtering of calculating, with be respectively apparent resistivity arithmetic or the geometrical mean of XY and YX pattern after the filtering of calculating;

5) guarantor's width correction factor of each measuring point two curves is asked for according to following formula:

$k_{\mathrm{fi}}^{\mathrm{XY}}=\log {\mathrm{\ρ}}_{\mathrm{si}}^{\mathrm{aXY}}-\log {\mathrm{\ρ}}_{\mathrm{fsi}}^{\mathrm{aXY}},$ $k_{\mathrm{fi}}^{\mathrm{YX}}=\log {\mathrm{\ρ}}_i^{\mathrm{aYX}}-\log {\mathrm{\ρ}}_{\mathrm{fsi}}^{\mathrm{aYX}}---\left(9\right)$

Wherein, with be respectively guarantor's width correction factor of XY and the YX curve of i-th measuring point;

6) obtain final guarantor's width according to following formulae discovery and correct rear two curve apparent resistivity datas with complete static correction;

${\mathrm{\ρ}}_{\mathrm{psi}}^{\mathrm{XY}}=10**(\log {\mathrm{\ρ}}_{\mathrm{si}}^{\mathrm{XY}}+k_{\mathrm{fi}}^{\mathrm{XY}}),$ ${\mathrm{\ρ}}_{\mathrm{psi}}^{\mathrm{YX}}=10**(\log {\mathrm{\ρ}}_{\mathrm{si}}^{\mathrm{YX}}+k_{\mathrm{fi}}^{\mathrm{YX}})---\left(10\right)$

Wherein, with the guarantor's width being exactly final XY and YX curve corrects rear apparent resistivity data.

2. method according to claim 1, feature is that the electromagnetic data described in step 1) comprises magnetotelluric data, audio-frequency magnetotelluric magnetic data, continuous profiling magnetotelluric data.

3. method according to claim 1, feature on average comprises arithmetic mean and geometric mean described in step 1).

4. method according to claim 1, feature is the calculating Average apparent resistivity data described in step 1) be:

Wherein, be respectively i-th measuring point jth frequency f _jxY and YX pattern measured visual resistivity value, n and m is selected frequency point number.

5. method according to claim 1, feature is that the filtering described in step 3) comprises two dimension median filter, two-dimensional space filtering, three-dimensional medium filtering and three dimensions filtering.