A method for inverting energy spectrum of high-frequency magnetotelluric instrument

Abstract

This application, applicable to the field of geological surveying technology, provides a method for inverting the energy spectrum of a high-frequency magnetotelluric instrument. This method uses the energy spectrum obtained by transforming and processing the time series signals in high-frequency magnetotelluric data as the data and model objective function in the inversion objective function, thereby obtaining the large-scale underground conductivity structure through inversion. This method avoids modifications to existing observation methods and equipment, does not incur additional costs, and is easy to operate.

Description

Technical Field

[0001] The present application belongs to the field of geological exploration technology, and in particular relates to an energy spectrum inversion method for a high-frequency magnetotelluric instrument. Background Art

[0002] In existing electromagnetic surveys, high-frequency magnetotelluric instruments have a detection frequency range of 10Hz–1000Hz. Compared to the conventional magnetotelluric frequency range of 0.0001Hz–1000Hz, these instruments measure only high-frequency electromagnetic fields and lack low-frequency information. Because low-frequency electromagnetic field information is closely related to large-scale subsurface structures, the lack of this low-frequency information can easily lead to local minima in the inversion of magnetotelluric data, making it impossible to accurately determine the subsurface conductivity structure. This presents a shortcoming in existing technologies. Summary of the Invention

[0003] The purpose of this application is to not change the existing observation methods and detection instruments, but only try to extract low-frequency signals from the received high-frequency signals to meet the actual needs of large-scale inversion of large-scale underground conductivity structures.

[0004] The present application provides an energy spectrum inversion method for a high-frequency magnetotelluric instrument, the method comprising the following steps:

[0005] a1. Observe the magnetotelluric field using a high-frequency magnetotelluric instrument to obtain time series signals;

[0006] a2. Perform a Fourier transform on the time series signal to obtain a signal spectrum with frequency as the abscissa and amplitude as the ordinate;

[0007] a3. Perform convolution operation on the full frequency band of the signal spectrum to obtain an energy spectrum;

[0008] a4. Select the largest value in the energy spectrum for inversion to extract large-scale underground conductivity information.

[0009] The present application also provides an energy spectrum inversion method for a high-frequency magnetotelluric instrument, the method comprising the following steps:

[0010] a1. Observe the magnetotelluric field using a high-frequency magnetotelluric instrument to obtain time series signals;

[0011] a2. Perform a Fourier transform on the time series signal to obtain a signal spectrum with frequency as the abscissa and amplitude as the ordinate;

[0012] a3 'of the full frequency band of the signal spectrum Hilbert transform to obtain an energy spectrum;

[0013] a4. Select the largest value in the energy spectrum for inversion to extract large-scale underground conductivity information.

[0014] This application uses the energy spectrum obtained by transforming time series signals from high-frequency magnetotelluric data as the data and model objective function in the inversion objective function, thereby obtaining the large-scale underground conductivity structure through inversion. This avoids changes to existing observation methods and equipment, does not increase additional costs, and is easy to operate. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 is a schematic diagram of the signal spectrum obtained by Fourier transforming the time series signal in this application;

[0016] Figure 2 yes Figure 1 The low-frequency energy spectrum extracted after convolution of the signal spectrum shown is shown;

[0017] Figure 3 This is the main flow chart of the energy spectrum inversion method of the high-frequency magnetotelluric instrument of this application. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical solutions and advantages of this application more clearly understood, the present application is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0019] The following describes the specific implementation of this application in detail with reference to specific embodiments:

[0020] Example:

[0021] Figure 3 The following is a flowchart of the energy spectrum inversion method for the high-frequency magnetotelluric instrument provided in the first embodiment of the present application. For ease of illustration, only the part related to the embodiment of the present application is shown, which is described in detail as follows:

[0022] A method for inverting an energy spectrum of a high-frequency magnetotelluric instrument, the method comprising the following steps:

[0023] a1. Observe the magnetotelluric field using a high-frequency magnetotelluric instrument to obtain time series signals.

[0024] In the specific implementation process, the time series signal contains both large-scale and small-scale electrical conductivity structure information of the underground.

[0025] a2. Perform Fourier transform on the time series signal to obtain the signal spectrum with frequency as the horizontal axis and amplitude as the vertical axis.

[0026] Specifically, the signal spectrum obtained by Fourier transform in step a2 only reveals information about the small-scale conductivity structure, but not the large-scale conductivity structure. Inversion interpretation can be performed based on this signal spectrum, but because inversion is prone to falling into local minima, it cannot actually obtain an accurate understanding of the small-scale subsurface conductivity structure. Therefore, this application requires further processing of this signal spectrum.

[0027] a3. Perform convolution operation on the full frequency band of the signal spectrum to obtain the energy spectrum;

[0028] In specific implementation, the frequency spectrum in step a2 used for convolution operation has a smaller error as the frequency range used is larger. Therefore, the present application preferably performs convolution processing on the entire frequency band of the signal spectrum obtained in step a2.

[0029] Specifically, the energy spectrum obtained in step a3 can display the low-frequency information part of the magnetotelluric signal, thereby resolving the large-scale conductivity structure information in the electromagnetic signal; based on this energy spectrum, inversion can be performed to obtain the large-scale conductivity structure of the underground.

[0030] a4. Select the largest value in the energy spectrum for inversion to extract large-scale underground conductivity information.

[0031] In a preferred embodiment, the energy spectrum of step a4 and the signal spectrum of the aforementioned step a2 jointly display the complete (large-scale and small-scale) information of the underground conductivity structure. Inversion interpretation based on these two types of information helps to overcome the problem of local minima in inversion, and is conducive to accurately extracting the complete underground conductivity structure.

[0032] In specific implementation, a simple time series signal can be used to simulate the time series signal observed by magnetotelluric. The frequency information is: f1 = 20 Hz, f2 = 25 Hz. The expression used to simulate the time series signal of magnetotelluric is:

[0033] s(t)=2sin(2πf1t)+2cos(2πf2t); (3)

[0034] Among them, f1 and f2 are the frequencies of the two peaks in the spectrum corresponding to the horizontal axis; t is the observation time; the spectrum of the signal is obtained by Fourier transform of the time series signal, as shown in Figure 1 shown.

[0035] At the same time, the energy in the magnetotelluric signal is defined as the square of the time series signal, that is:

[0036] p(t)=s1(t)*s1(t);

[0037] Where p(t) is the energy of the time series signal.

[0038] Performing a convolution operation on the spectrum of a time series signal can obtain an energy spectrum containing low-frequency signals.

[0039] Specifically, such as Figure 2 As shown, observe the attached Figure 2 The processed energy spectrum shown in the figure shows that the frequencies corresponding to the energy spectrum peaks are in high-frequency (40-50 Hz) and low-frequency (0-5 Hz) regions. The frequency range corresponding to the low-frequency region is far lower than the frequency of the source time series signal (20 Hz and 25 Hz). The low-frequency signal appearing in this energy spectrum has a frequency far lower than the frequency of the time series signal, but its amplitude is much higher than the intensity of the high-frequency portion. The signal-to-noise ratio in this low-frequency region is very high. Because the low-frequency signal contains large-scale underground information, inversion based on the low-frequency information in the spectrum energy can obtain large-scale underground structural information, providing support for conventional high-frequency magnetotelluric inversion and avoiding the problem of local minima.

[0040] Furthermore, the observation time in step a1 is set to be greater than or equal to 10 minutes and less than or equal to 20 minutes.

[0041] In practice, when Fourier transforming a time-domain signal, the longer the time-domain signal used, the higher the quality of the resulting spectrum signal. However, longer time-domain signals require longer observation times, which is uneconomical in production. Therefore, setting the observation time to 10 to 20 minutes achieves a balance between high signal quality and economical observation (shorter observation time).

[0042] Furthermore, in step a3, the energy spectrum is specifically calculated as follows:

[0043]

[0044] Where ω is the frequency of energy, and the range of ω is 0Hz-60Hz; η is the frequency of the signal; P(ω) is the energy spectrum; S1(η) is the signal spectrum.

[0045] Specifically, in formula (1), the range of η as the frequency of the signal is positive and negative infinity, but in actual operation, such a large range is not possible. Therefore, the value of η can be selected according to the conventional convolution integral discretization operation.

[0046] In other embodiments, step a3' is used instead of step a3.

[0047] In step a3', Hilbert transform is used to process the signal spectrum of the entire frequency band into an energy spectrum.

[0048] The specific calculation method of Hilbert energy spectrum is:

[0049]

[0050] Where ω is the energy frequency, which ranges from 0 Hz to 60 Hz; η is the signal frequency; H(ω) is the Hilbert energy spectrum; and S1(η) is the signal spectrum. Using the Hilbert transform to process the signal spectrum requires less computation. Compared to using convolution, it reduces processing device resource usage and increases computational speed.

[0051] Specifically, in formula (2), the range of η as the frequency of the signal is positive and negative infinity, but in actual operation, such a large range is not possible. Therefore, the value of η can be selected according to the conventional convolution integral discretization operation.

[0052] Furthermore, in step a3, the energy spectrum includes a plurality of different frequency ranges selected from the full-band signal spectrum for calculation to obtain a spectrum corresponding to the selected frequency.

[0053] Specifically, in formula 1, ω is the frequency of energy. Each time the calculation of formula (1) is completed, a value ( Figure 2 The spectrum of the energy in the Figure 2 As shown, ω is selected in the range of 0-60Hz, and ω is selected in multiple different values. For each value, the calculation of formula (1) is repeated once, and the following can be obtained: Figure 2 The energy spectrum in .

[0054] like Figure 2 In a study, it was found that when the frequency approaches 0 Hz, the energy spectrum reaches its maximum value, demonstrating strong resistance to interference. Therefore, the 0 Hz energy spectrum is selected for inversion to obtain large-scale subsurface conductivity information. In practice, different time series spectra will vary depending on the high-frequency magnetotelluric signals measured. The resulting energy spectrum will also vary. Generally speaking, after calculating the energy spectrum, selecting one or more signal spectra with the largest spectral values ​​as inversion data can improve the accuracy of extracting large-scale subsurface conductivity information.

[0055] The embodiment of the present application is based on the improvement of low-frequency signal extraction from high-frequency magnetotelluric field signals measured by traditional observation methods and existing high-frequency magnetotelluric instruments. By Fourier transforming the time series signal and further processing the signal spectrum into an energy spectrum, the low-frequency information contained in the energy signal is obtained. There is no need to observe the low-frequency information during observation, nor is there any need to improve the observation instrument. The original observation method and equipment are retained without adding additional workload and working time. At the same time, a low-frequency signal that can obtain large-scale structural information is provided for inversion, providing support for the inversion of conventional high-frequency magnetotelluric methods and avoiding the problem of local minima.

[0056] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent replacements and improvements made within the spirit and principles of the present application should be included in the scope of protection of the present application.

Claims

1. A method for inverting the energy spectrum of a high-frequency magnetotelluric instrument, characterized in that: The method comprises the following steps: a1. Observe the magnetotelluric field using a high-frequency magnetotelluric instrument to obtain time series signals; a2. Perform a Fourier transform on the time series signal to obtain a signal spectrum with frequency as the abscissa and amplitude as the ordinate; a3. Perform convolution operation on the full frequency band of the signal spectrum to obtain an energy spectrum; a4. Select the largest median value of the energy spectrum for inversion to extract large-scale underground conductivity information; The energy spectrum obtained in step a3 can display the low-frequency signal portion of the magnetotelluric signal. Since the low-frequency signal contains large-scale information about the underground, the large-scale conductivity structure information in the electromagnetic signal is parsed. Based on this energy spectrum, inversion is performed to obtain the large-scale conductivity structure of the underground.

2. The method according to claim 1, wherein The observation time in step a1 is set to be greater than or equal to 10 minutes and less than or equal to 20 minutes.

3. The method according to claim 1, wherein In step a3, the energy spectrum is specifically calculated as follows: Wherein, ω is the frequency of energy, and the selection range of ω is: 0 Hz-60 Hz; η is the frequency of the signal; P(ω) is the energy spectrum; S1(η) is the signal spectrum.

4. The method according to claim 3, wherein In step a3, the energy spectrum includes spectrums corresponding to multiple frequencies; the frequencies are selected based on the range of the full frequency band.

5. A method for inverting the energy spectrum of a high-frequency magnetotelluric instrument, characterized in that: The method comprises the following steps: a1. Observe the magnetotelluric field using a high-frequency magnetotelluric instrument to obtain time series signals; a2. Perform a Fourier transform on the time series signal to obtain a signal spectrum with frequency as the abscissa and amplitude as the ordinate; a3 'of the full frequency band of the signal spectrum Hilbert transform to obtain an energy spectrum; a4. Select the largest value in the energy spectrum for inversion to extract large-scale underground conductivity information.

Images