Sea wave induction electromagnetic noise suppression method and device

Abstract

The invention relates to the technical field of ocean geophysical exploration, and provides a sea wave induction electromagnetic noise suppression method and device, and the method comprises the steps: obtaining an optimal structural element; obtaining a multi-scale opening-closing operation result and a multi-scale closing-opening operation result; performing weighted linear combination on the multi-scale generalized form opening-closing operation and the multi-scale generalized form closing-opening operation, updating a weighting coefficient through iterative calculation, and obtaining ocean magnetotelluric data after the whole section of noise is suppressed; and evaluating the noise suppression effect. According to the method, multi-scale adaptive generalized morphological filtering is improved, the filtering precision of the marine magnetotelluric data is remarkably improved, and the quality of the magnetotelluric data after noise suppression is remarkably improved.

Description

Technical Field

[0001] This application relates to the technical field of marine geophysical exploration, and particularly to a method and device for suppressing ocean wave-induced electromagnetic noise. Background Art

[0002] Marine Magnetotelluric (MMT) is a geophysical method that uses the natural electromagnetic field as the field source, measures the mutually orthogonal electric and magnetic fields through an ocean-bottom electromagnetic acquisition station (OBEM), and studies the electrical properties of the earth. Due to the complex marine environment, the seawater is constantly in motion. The moving seawater will cut the electromagnetic field to generate a secondary electromagnetic field, and the electromagnetic receiving instruments on the seabed are also easily washed by the seawater and vibrate, which seriously interferes with marine magnetotelluric exploration and reduces the signal-to-noise ratio of marine magnetotelluric data. Among them, the induced electromagnetic field generated by ocean wave motion is one of the main noises affecting the quality of marine electromagnetic data. The electromagnetic field intensity generated by it is relatively large, and the influence on the electromagnetic field is mainly concentrated around 0.1 Hz, resulting in a decrease in the signal-to-noise ratio of magnetotelluric data in this frequency band and serious distortion of apparent resistivity and phase curves. Therefore, when processing marine magnetotelluric data, it is necessary to extract the characteristics of ocean wave-induced electromagnetic fields.

[0003] At present, there are relatively few studies on suppressing ocean wave-induced noise. Literature 1 (Wei Wenbo et al., Experimental study on marine magnetotelluric sounding in the southern Yellow Sea [J]. Chinese Journal of Geophysics) appropriately corrects marine electromagnetic noise by analyzing the characteristics of ocean wave-induced electromagnetic fields, but has a high requirement for the calculation accuracy of ocean wave-induced electromagnetic fields; Literature 2 (Yu Caixia et al., Application of Hilbert-Huang transform in processing marine magnetotelluric sounding data [J]. Progress in Geophysics) applies empirical mode decomposition to the processing of marine magnetotelluric data, effectively suppressing the electromagnetic noise generated by seawater movement, but there is a "flying point" phenomenon; In recent years, the wavelet analysis-based denoising method for marine electromagnetic data can suppress ocean wave electromagnetic noise to a certain extent, but this method is highly dependent on the selection of wavelet basis functions. The above methods all have some limitations or deficiencies in the extraction of ocean wave-induced noise. Therefore, it is of great significance to find a method that can effectively separate magnetotelluric signals from ocean wave-induced electromagnetic noise. Summary of the Invention

[0004] Aiming at the problems existing in the prior art, this application provides a method and device for suppressing ocean wave-induced electromagnetic noise to solve the problem that the prior art cannot effectively separate magnetotelluric signals from ocean wave-induced electromagnetic noise.

[0005] In a first aspect, this application provides a method for suppressing ocean wave-induced electromagnetic noise, the method comprising: Step S1: Determine the type and size of the structural element and obtain the optimal structural element; Step S2: Perform multi-scale operations on the entire original marine magnetotelluric data using the optimal structural element to obtain the multi-scale opening-closing operation result and the multi-scale closing-opening operation result; Step S3: Perform a weighted linear combination of the multi-scale generalized morphological opening-closing operation and the multi-scale generalized morphological closing-opening operation, and use the gradient method to iteratively calculate the weighted coefficients of the multi-scale generalized morphological opening-closing operation and the weighted coefficients of the multi-scale morphological closing-opening operation to update the weighted coefficients, and obtain the marine magnetotelluric data after suppressing the noise of the entire segment; Step S4: Introduce the signal-to-noise ratio SNR , the root mean square error RMSE, and the normalized cross-correlation coefficient NCC Three indicators are used to evaluate the noise suppression effect; Step S5: Perform Robust impedance estimation on the marine magnetotelluric data after suppressing the noise of the entire segment, calculate the apparent resistivity and phase, and evaluate the apparent resistivity and phase data.

[0006] Furthermore, the determining the type and size of the structural element and obtaining the optimal structural element includes: According to the marine magnetotelluric data and the characteristics of the sea wave-induced electromagnetic noise, set the type of the optimal structural element as a triangular structural element; For the determination of the structural element size, it is achieved through the following steps: S11: Perform preliminary filtering to determine the structural element size range; S12: Input the entire ideal marine magnetotelluric data, perform multi-scale generalized morphological filtering to traverse all structural elements within the structural element size range, and obtain the marine magnetotelluric data after multi-scale generalized morphological filtering of the entire segment and the marine magnetotelluric data after multi-scale generalized morphological filtering of the frequency band greatly affected by the sea wave noise; According to the entire ideal marine magnetotelluric data, the marine magnetotelluric data after multi-scale generalized morphological filtering of the entire segment, the ideal marine magnetotelluric data of the frequency band greatly affected by the sea wave noise, and the marine magnetotelluric data after multi-scale generalized morphological filtering of the frequency band greatly affected by the sea wave noise, set the evaluation index, and obtain the optimal structural element size under the optimal evaluation index.

[0007] Furthermore, the evaluation index is: , where C is the evaluation index, is the normalized cross-correlation coefficient of the time series of the entire ideal marine magnetotelluric data and the marine magnetotelluric data after multi-scale generalized morphological filtering of the entire segment, is the normalized cross-correlation coefficient between the amplitude spectra of the ideal marine magnetotelluric data in the frequency band significantly affected by ocean wave noise and the marine magnetotelluric data after multi-scale generalized morphological filtering in the frequency band significantly affected by ocean wave noise: , wherein, represents the entire segment of ideal marine magnetotelluric data, represents the entire segment of marine magnetotelluric data after multi-scale generalized morphological filtering, is the sampling point of the time series data, represents the length of the time series data, represents the length of the frequency band affected by ocean waves in the amplitude spectrum.

[0008] Furthermore, the step S2 includes: The generalized morphological filter includes a generalized morphological opening-closing filter and a generalized morphological closing-opening filter, and the generalized morphological opening-closing filter and the generalized morphological closing-opening filter are: , wherein the input signal is the entire segment of original marine magnetotelluric data; is the sampling point of the time series data, , represents the length of the time series data, and represent structural elements of different sizes; the symbol represents the opening operation, and the symbol represents the closing operation; GOC represents the generalized morphological opening-closing filter, and GCO represents the generalized morphological closing-opening filter, ; Based on the generalized morphological filter, multi-scale generalized morphological operations are obtained by performing multiple erosions and multiple dilations on the structural element. The multi-scale generalized morphological opening operation and the multi-scale generalized morphological closing operation are: , wherein, is the entire segment of original marine magnetotelluric data; and represent structural elements of different sizes, the symbol represents the dilation operation, and the symbol represents the erosion operation, s represents the scale of the multi-scale generalized morphological operation, is the sampling point of the time series data, represents the length of the time series data; Based on the multi-scale generalized morphological operation, the multi-scale generalized morphological opening-closing operation is obtained and the multi-scale generalized morphological closing-opening operation : , wherein, is the entire original marine magnetotelluric data; and represent structural elements of different sizes, s represents the scale of the multi-scale generalized morphological operation, is the sampling point of the time series data, represents the length of the time series data.

[0009] Furthermore, the weighted linear combination of the multi-scale generalized morphological opening-closing operation and the multi-scale generalized morphological closing-opening operation includes: Performing a weighted linear combination of the multi-scale generalized morphological opening-closing operation and the multi-scale generalized morphological closing-opening operation to obtain the marine magnetotelluric data after suppressing the entire noise is: , wherein, is the weighted coefficient of the multi-scale generalized morphological opening-closing operation , is the weighted coefficient of the multi-scale morphological closing-opening operation , is the sampling point of the time series data; Assume , , then the above formula can be expressed as: , wherein, is the weighted coefficient, , is the sampling point of the time series data.

[0010] Furthermore, the iterative calculation of the weighted coefficients of the multi-scale generalized morphological opening-closing operation and the weighted coefficients of the multi-scale morphological closing-opening operation by using the gradient method to update the weighted coefficients to obtain the marine magnetotelluric data after suppressing the entire noise includes: Step S31: Performing the multi-scale generalized morphological opening-closing operation on the input signal to obtain , performing the multi-scale closing-opening operation to obtain , then the output of the multi-scale generalized morphological filter after the k-th iteration is: , wherein, is the weighted coefficient of the k-th iteration, , is a sampling point of time series data, , indicating the length of the time series data; Step S32: Obtain a single error sample after the k-th iteration : , wherein, is the output of the multi-scale generalized morphological filter after the (k - 1)-th iteration, is the output of the multi-scale generalized morphological filter after the k-th iteration, is a sampling point of time series data, , indicating the length of the time series data, then the gradient is: , , wherein, is the square of the single error sample at the k-th iteration with respect to the gradient of the weighting coefficient , is the square of the single error sample at the k-th iteration with respect to the gradient of the weighting coefficient ; Step S33: Calculate the direction vector : , wherein, is the first component of the direction vector at the k-th iteration, is the second component of the direction vector at the k-th iteration, is the -th component of the direction vector at the k-th iteration, is the -th component of the direction vector at the (k - 1)-th iteration, is the square of the single error sample at the k-th iteration with respect to the gradient of the weighting coefficient , is the square of the single error sample at the (k - 1)-th iteration with respect to the gradient of the weighting coefficient ; Step S34: Calculate the weighting coefficient: , wherein, is the weighting coefficient at the (k + 1)-th iteration, is the weighting coefficient for the k-th iteration, is the direction vector for the k-th iteration, , is the step size parameter.

[0011] Further, the step S3 further includes: The multi-scale adaptive generalized morphological filtering is an iterative filtering process. In the iterative filtering process, the initial input signal is the entire segment of the original marine magnetotelluric data. After each iterative filtering, data reconstruction is performed on the filtered marine magnetotelluric data. The data reconstruction is to perform band-pass filtering on the filtered marine magnetotelluric data, and only retain the marine magnetotelluric data in the frequency band that is more affected by the sea wave noise in the filtered marine magnetotelluric data, which is denoted as the intermediate marine magnetotelluric data in the frequency band more affected by the sea wave noise. The marine magnetotelluric data in the remaining frequency bands is the same as the original marine magnetotelluric data. The intermediate marine magnetotelluric data in the frequency band more affected by the sea wave noise is reconstructed with the marine magnetotelluric data in the remaining frequency bands to obtain the reconstructed marine magnetotelluric data after each iterative filtering; then the reconstructed marine magnetotelluric data is used as the input signal to repeat the iterative filtering process; when the iterative stop condition is satisfied, the marine magnetotelluric data after noise suppression in the frequency band more affected by the sea wave noise and the entire segment of the marine magnetotelluric data after noise suppression are obtained.

[0012] Further, the signal-to-noise ratio SNR , the root mean square error RMSE and the normalized cross-correlation coefficient NCC are respectively: , wherein, represents the entire segment of the original marine magnetotelluric data, represents the entire segment of the marine magnetotelluric data after noise suppression, is the time series data sampling point, represents the length of the time series data.

[0013] In a second aspect, the present application provides a device for suppressing sea wave-induced electromagnetic noise, and the device includes: Optimal structure element determination module: Determine the type and size of the structure element to obtain the optimal structure element; Multi-scale generalized morphological operation module: Perform multi-scale operations on the entire segment of the original marine magnetotelluric data using the optimal structure element to obtain the multi-scale opening-closing operation result and the multi-scale closing-opening operation result; Noise suppression module: Perform a weighted linear combination of multi-scale generalized morphological opening-closing operation and multi-scale generalized morphological closing-opening operation, and use the gradient method to iteratively calculate the weighted coefficients of the multi-scale generalized morphological opening-closing operation and the weighted coefficients of the multi-scale morphological closing-opening operation to update the weighted coefficients, so as to obtain the marine magnetotelluric data after suppressing the whole section of noise; The first noise suppression effect evaluation module: Introduce the signal-to-noise ratio SNR , the root mean square error RMSE and the normalized cross-correlation coefficient NCC to evaluate the noise suppression effect with three indicators; The second noise suppression effect evaluation module: Perform Robust impedance estimation on the marine magnetotelluric data after suppressing the whole section of noise, calculate the apparent resistivity and phase, and evaluate the apparent resistivity and phase data.

[0014] In a third aspect, an electronic device according to the present application includes: A processor; A memory; And a computer program, where the computer program is stored in the memory, and the computer program includes instructions that, when executed by the processor, cause the electronic device to execute the method described in the first aspect.

[0015] Based on the above invention content, compared with the prior art, the present application improves the multi-scale adaptive generalized morphological filtering. By setting evaluation indicators and improving the method for obtaining the structural element size, the obtained structural element is more suitable for filtering marine magnetotelluric data; by performing a weighted linear combination of the multi-scale generalized morphological opening-closing operation and the multi-scale generalized morphological closing-opening operation, the filtering accuracy is improved, and the gradient method is used to iteratively calculate the weighted coefficients of the multi-scale generalized morphological opening-closing operation and the weighted coefficients of the multi-scale generalized morphological closing-opening operation to update the weighted coefficients, further improving the accuracy of the multi-scale adaptive generalized morphological filter; after each iterative filtering, the filtered data is reconstructed as the input data for the next iteration, further improving the filtering accuracy; applying the parameter experience of the simulated data to the measured data to test the actual application effect of the method. In summary, using the method of the present application significantly improves the filtering accuracy of marine magnetotelluric data, significantly improves the noise suppression effect, and significantly improves the quality of the magnetotelluric data after noise suppression. Description of the Drawings

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings required for the embodiments. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0017] Figure 1 It is a schematic flow chart of a method for suppressing ocean wave-induced electromagnetic noise provided by an embodiment of the present application; Figure 2 It is a one-dimensional horizontally layered geoelectric model for simulating the E y component and the H x component of marine magnetotelluric data provided by an embodiment of the present application; Figure 3a It is the E y component of the simulated marine magnetotelluric data provided by an embodiment of the present application; Figure 3b It is the H x component of the simulated marine magnetotelluric data provided by an embodiment of the present application; Figure 4a It is the time series of the X component of the ocean wave-induced magnetic field provided by an embodiment of the present application; Figure 4b It is the amplitude spectrum of the X component of the ocean wave-induced magnetic field provided by an embodiment of the present application; Figure 5a It is the time series of the H x component of the marine magnetotelluric data before and after adding noise. The red line represents the noise-free data, and the blue line represents the data with added noise; Figure 5b It is the amplitude spectrum of the H x component of the marine magnetotelluric data before and after adding noise provided by an embodiment of the present application; Figure 6a It is the comparison of the time series of the H x component of the simulated marine magnetotelluric data before and after noise suppression provided by an embodiment of the present application; Figure 6b It is the detailed comparison of the time series of the H x component of the simulated marine magnetotelluric data before and after noise suppression provided by an embodiment of the present application; Figure 6c It is the comparison of the power spectra of the H x component of the simulated marine magnetotelluric data before and after noise suppression provided by an embodiment of the present application; Figure 7a It is the comparison of the overall time series of the simulated ocean wave-induced magnetic noise and the X component of the ocean wave-induced magnetic noise extracted by filtering provided by an embodiment of the present application; Figure 7b It is the comparison of the local time series of the simulated ocean wave-induced magnetic noise and the X component of the ocean wave-induced magnetic noise extracted by filtering provided by an embodiment of the present application; Figure 8Comparison of apparent resistivity curves and phase curves estimated from the original data (blue), noise data (black), data denoised by morphological filtering (MMF) (red), and EMD method (green) provided in the embodiments of the present application; Figure 9a is the measured marine magnetotelluric data E x component time series; Figure 9b is the measured marine magnetotelluric data E x component amplitude spectrum; Figure 9c is the measured marine magnetotelluric data E y component time series; Figure 9d is the measured marine magnetotelluric data E y component amplitude spectrum; Figure 9e is the measured marine magnetotelluric data H x component time series; Figure 9f is the measured marine magnetotelluric data H x component amplitude spectrum; Figure 9g is the measured marine magnetotelluric data H y component time series; Figure 9h is the measured marine magnetotelluric data H y component amplitude spectrum; Figure 10a is the measured horizontal magnetic field H x time series before and after data denoising; Figure 10b is the measured horizontal magnetic field H x local time series before and after data denoising; Figure 10c is the measured horizontal magnetic field H x power spectral density before and after data denoising; Figure 10d is the measured horizontal magnetic field H y time series before and after data denoising; Figure 10e is the measured horizontal magnetic field H y time series before and after data denoising; Figure 10fis the measured horizontal magnetic field H provided by the embodiments of the present application y Power spectral density before and after data denoising; Figure 11a is the measured horizontal magnetic field H provided by the embodiments of the present application x Time-frequency diagram before data noise suppression; Figure 11b is the measured horizontal magnetic field H provided by the embodiments of the present application x Time-frequency diagram after data noise suppression; Figure 11c is the measured horizontal magnetic field H provided by the embodiments of the present application y Time-frequency diagram before data noise suppression; Figure 11d is the measured horizontal magnetic field H provided by the embodiments of the present application y Time-frequency diagram after data noise suppression; Figure 12a are the apparent resistivity and phase curves in the XY direction before and after data noise suppression of the measured data provided by the embodiments of the present application; Figure 12b are the apparent resistivity and phase curves in the YX direction before and after data noise suppression of the measured data provided by the embodiments of the present application; Figure 13 is the structural block diagram of a sea wave induced electromagnetic noise suppression device provided by the embodiments of the present application; Figure 14 is the structural schematic diagram of an electronic device provided by the embodiments of the present application. Detailed implementation manners

[0018] For a better understanding of the technical solutions of the present invention, the embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0019] It should be clear that the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without any creative work fall within the scope of protection of the present invention.

[0020] The terms used in the embodiments of the present application are only for the purpose of describing specific embodiments, and are not intended to limit the present application. The singular forms of "a", "the" and "said" used in the embodiments of the present application and the appended claims are also intended to include the plural forms, unless the context clearly indicates otherwise.

[0021] It should be understood that the term "and / or" used herein is merely a description of the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B may represent three situations: A exists alone, A and B exist simultaneously, and B exists alone. Additionally, the character " / " in this text generally indicates that the associated objects before and after are in an "or" relationship.

[0022] Referring to Figure 1 , an embodiment of the present application provides a method for suppressing sea wave-induced electromagnetic noise. Specifically, when implemented, the following steps may be included.

[0023] Step S1: Determine the type and size of the structural element and obtain the optimal structural element.

[0024] Based on multi-scale adaptive generalized morphological filtering, the present application suppresses sea wave-induced electromagnetic noise. During the filtering process, the selection of the type and size of the structural element has a greater impact on the suppression result of sea wave-induced electromagnetic noise.

[0025] According to the marine magnetotelluric data and the characteristics of sea wave-induced electromagnetic noise, the triangular structural element better fits the characteristics of marine magnetotelluric data. Therefore, the type of the optimal structural element is set to the triangular structural element.

[0026] For the determination of the size of the structural element, it is achieved through the following steps: S11: Perform preliminary filtering to determine the range of the structural element size.

[0027] S12: Input the entire ideal marine magnetotelluric data, perform multi-scale generalized morphological filtering to traverse all structural elements within the range of the structural element size, and obtain the marine magnetotelluric data after multi-scale generalized morphological filtering for the entire section and the marine magnetotelluric data after multi-scale generalized morphological filtering for the frequency band greatly affected by sea wave noise.

[0028] The multi-scale generalized morphological filtering is an iterative filtering process. In the iterative filtering process, the initial input signal is the entire ideal marine magnetotelluric data. After each iterative filtering, data reconstruction is performed on the filtered marine magnetotelluric data. The data reconstruction is to perform band-pass filtering on the marine magnetotelluric data after multi-scale generalized morphological filtering, and only retain the marine magnetotelluric data in the frequency band that is greatly affected by ocean wave noise in the filtered marine magnetotelluric data, which is denoted as the intermediate marine magnetotelluric data in the frequency band greatly affected by ocean wave noise. The marine magnetotelluric data in the remaining frequency bands is the same as the ideal marine magnetotelluric data. The intermediate marine magnetotelluric data in the frequency band greatly affected by ocean wave noise and the marine magnetotelluric data in the remaining frequency bands are reconstructed to obtain the reconstructed marine magnetotelluric data after each filtering. Then, the reconstructed marine magnetotelluric data is used as the input signal to repeat the iterative filtering process. When the iterative stop condition is met, the marine magnetotelluric data after multi-scale generalized morphological filtering in the frequency band greatly affected by ocean wave noise and the entire marine magnetotelluric data after multi-scale generalized morphological filtering are obtained.

[0029] According to the entire ideal marine magnetotelluric data, the entire marine magnetotelluric data after multi-scale generalized morphological filtering, the ideal marine magnetotelluric data in the frequency band greatly affected by ocean wave noise, and the marine magnetotelluric data after multi-scale generalized morphological filtering in the frequency band greatly affected by ocean wave noise, evaluation indexes are set, and the optimal structural element size under the optimal evaluation index is obtained.

[0030] The evaluation indexes are as follows: , where C is the evaluation index, is the normalized cross-correlation coefficient of the time series of the entire ideal marine magnetotelluric data and the entire marine magnetotelluric data after multi-scale generalized morphological filtering, is the normalized cross-correlation coefficient of the amplitude spectra of the ideal marine magnetotelluric data in the frequency band greatly affected by ocean wave noise and the marine magnetotelluric data after multi-scale generalized morphological filtering in the frequency band greatly affected by ocean wave noise: , where, represents the entire ideal marine magnetotelluric data, represents the entire marine magnetotelluric data after multi-scale generalized morphological filtering, is the sampling point of the time series data, represents the length of the time series data, represents the length of the frequency band affected by ocean waves in the amplitude spectrum.

[0031] When C reaches the maximum, the structural element size is the optimal structural element size.

[0032] Step S2: Perform multi-scale operations on the entire segment of the original marine magnetotelluric data using the optimal structural element to obtain the multi-scale opening-closing operation result and the multi-scale closing-opening operation result.

[0033] The generalized morphological filter includes a generalized morphological opening-closing filter and a generalized morphological closing-opening filter, and the generalized morphological opening-closing filter and the generalized morphological closing-opening filter are: , where the input signal is the entire segment of the original marine magnetotelluric data; is the sampling point of the time series data, , represents the length of the time series data, and represent structural elements of different sizes; the symbol represents the opening operation, and the symbol represents the closing operation; GOC represents the generalized morphological opening-closing filter, and GCO represents the generalized morphological closing-opening filter, .

[0034] Based on the generalized morphological filter, multi-scale generalized morphological operations are obtained by performing multiple erosions and multiple dilations on the structural element. The multi-scale generalized morphological opening operation and the multi-scale generalized morphological closing operation are: , where is the entire segment of the original marine magnetotelluric data; and represent structural elements of different sizes, the symbol represents the dilation operation, and the symbol represents the erosion operation, s represents the scale of the multi-scale generalized morphological operation, is the sampling point of the time series data, represents the length of the time series data.

[0035] According to the multi-scale generalized morphological operation, the multi-scale generalized morphological opening-closing operation and the multi-scale generalized morphological closing-opening operation are obtained: , where is the entire segment of the original marine magnetotelluric data; and represent structural elements of different sizes, s represents the scale of the multi-scale generalized morphological operation, is the sampling point of the time series data represents the length of the time series data

[0036] Step S3: Perform a weighted linear combination of the multi-scale generalized morphological opening-closing operation and the multi-scale generalized morphological closing-opening operation, and use the gradient method to iteratively calculate the weighted coefficients of the multi-scale generalized morphological opening-closing operation and the weighted coefficients of the multi-scale morphological closing-opening operation to update the weighted coefficients, so as to obtain the marine magnetotelluric data after suppressing the whole section of noise

[0037] Perform a weighted linear combination of the multi-scale generalized morphological opening-closing operation and the multi-scale generalized morphological closing-opening operation to obtain the marine magnetotelluric data after suppressing the whole section of noise is as follows , where is the weighted coefficient of the multi-scale generalized morphological opening-closing operation is the weighted coefficient is the multi-scale morphological closing-opening operation is the weighted coefficient is the sampling point of the time series data

[0038] Assume , then the above formula can be expressed as , where is the weighted coefficient , is the sampling point of the time series data

[0039] In order to make the filtering effect optimal, use the gradient method to iteratively calculate the weighted coefficients of the multi-scale generalized morphological opening-closing operation and the weighted coefficients of the multi-scale morphological closing-opening operation to obtain the optimal weighted coefficients

[0040] Let the actual input data be the whole section of the original marine magnetotelluric data is as follows , where is the whole section of the ideal marine magnetotelluric data is the whole section of the simulated ocean wave noise is the sampling point of the time series data represents the length of the time series data

[0041] The output signal is the marine magnetotelluric data after suppressing the whole section of noise , is the time series of the whole section of the ideal marine magnetotelluric data The difference between the marine magnetotelluric data after suppressing the whole - segment noise, that is is .

[0042] Then the mean - square error of the marine magnetotelluric data after suppressing the whole - segment noise is: , where is the mean - square error of the output signal, is the ideal marine magnetotelluric data for the whole segment, is the marine magnetotelluric data after suppressing the whole - segment noise, is the ideal marine magnetotelluric data for the whole segment and the difference between the marine magnetotelluric data after suppressing the whole - segment noise , is the weighting coefficient, , is the sampling point of the time - series data, , represents the length of the time - series data.

[0043] The gradient method is used to gradually correct the value of the weighting coefficient, so that the marine magnetotelluric data after suppressing the whole - segment noise constantly approaches the ideal marine magnetotelluric data for the whole segment under the condition of the minimum mean - square error.

[0044] To simplify the calculation, take the square of a single error sample as the estimate of the mean - square error , the gradient of the weighting coefficient is: , where is the gradient of with respect to the weighting coefficient is the gradient of with respect to the weighting coefficient, n is the sampling point of the time - series data, is the sampling point of the time - series data, , represents the length of the time - series data, , .

[0045] The iterative calculation process of multi - scale adaptive generalized morphological filtering based on the gradient method is as follows: Step S31: Perform multi - scale generalized morphological opening - closing operation on the input signal to obtain , and perform multi - scale closing - opening operation to obtain , the output of the multi-scale generalized morphological filter after the k-th iteration is: , where, is the weighting coefficient of the k-th iteration, , is the sampling point of the time series data, , represents the length of the time series data.

[0046] Step S32: Obtain a single error sample after the k-th iteration : , where, is the output of the multi-scale generalized morphological filter after the (k - 1)-th iteration, is the output of the multi-scale generalized morphological filter after the k-th iteration, is the sampling point of the time series data, , represents the length of the time series data, then the gradient is: , where, is the square of the single error sample at the k-th iteration with respect to the gradient of the weighting coefficient , is the square of the single error sample at the k-th iteration with respect to the gradient of the weighting coefficient .

[0047] Step S33: Calculate the direction vector : , where, is the first component of the direction vector at the k-th iteration, is the second component of the direction vector at the k-th iteration, is the -th component of the direction vector at the k-th iteration, is the -th component of the direction vector at the (k - 1)-th iteration, is the square of the single error sample at the k-th iteration with respect to the gradient of the weighting coefficient , is the square of the single error sample at the (k - 1)-th iteration with respect to the gradient of the weighting coefficient .

[0048] Step S34: Calculate the weighting coefficient: , where is the weighting coefficient of the (k + 1)-th iteration, is the weighting coefficient of the k-th iteration, is the direction vector of the k-th iteration, , is the step size parameter.

[0049] The multi-scale adaptive generalized morphological filtering is an iterative filtering process. In the iterative filtering process, the initial input signal is the entire segment of the original marine magnetotelluric data. After each iterative filtering, data reconstruction is performed on the filtered marine magnetotelluric data. The data reconstruction is to perform band-pass filtering on the filtered marine magnetotelluric data, and only retain the marine magnetotelluric data in the frequency band that is more affected by the sea wave noise in the filtered marine magnetotelluric data, which is denoted as the intermediate marine magnetotelluric data in the frequency band more affected by the sea wave noise. The marine magnetotelluric data in the remaining frequency bands is the same as the original marine magnetotelluric data. The intermediate marine magnetotelluric data in the frequency band more affected by the sea wave noise and the marine magnetotelluric data in the remaining frequency bands are reconstructed to obtain the reconstructed marine magnetotelluric data after each iterative filtering. Then, the reconstructed marine magnetotelluric data is used as the input signal to repeat the iterative filtering process. When the iterative stop condition is met, the marine magnetotelluric data with noise suppression in the frequency band more affected by the sea wave noise and the entire segment of the marine magnetotelluric data with noise suppression are obtained.

[0050] Step S4: Introduce the signal-to-noise ratio SNR , the root mean square error RMSE, and the normalized cross-correlation coefficient NCC to evaluate the noise suppression effect with three indicators: , where represents the entire segment of the original marine magnetotelluric data, represents the entire segment of the marine magnetotelluric data with noise suppression, is the sampling point of the time series data, represents the length of the time series data.

[0051] The signal-to-noise ratio SNR and the normalized cross-correlation coefficient NCC the larger ， the smaller the root mean square error RMSE, indicating that the noise suppression effect is better.

[0052] Step S5: Perform Robust impedance estimation on the entire segment of the marine magnetotelluric data with noise suppression, calculate the apparent resistivity and phase, and evaluate the apparent resistivity and phase data.

[0053] The technical solution of the present application will be further described below in conjunction with specific examples.

[0054] Example 1: In this example, simulated data is first set for experiments. In order to obtain marine magnetotelluric simulated data, a one-dimensional geoelectric model as shown in Figure 2 is designed, and the marine magnetotelluric data is simulated using the structural system function method. Assuming the sampling frequency is 10 Hz and the sampling duration is 30 min, the time series of the E y component of the marine magnetotelluric field is simulated as shown in Figure 3a , and the time series of the H x component of the marine magnetotelluric field is simulated as shown in Figure 3b .

[0055] The simulation of the sea wave induced magnetic field starts from Maxwell's equations, uses the direct differentiation method to derive the differential equation of the sea wave motion induced magnetic field, and combines the continuity conditions of the magnetic field on the sea surface and the sea bottom surface to derive the expression of the sea wave motion induced magnetic field. Assuming the geomagnetic field is 50000 nT, the wind speed is 9 m / s, the seawater depth is 50 m, and the sampling rate is 10 Hz. Figure 4a To simulate the 30-min time series of the sea wave induced magnetic field, Figure 4b is the amplitude spectrum of the simulated sea wave induced magnetic field.

[0056] From the amplitude spectrum of Figure 4b , it can be seen that the energy of the simulated sea wave induced magnetic noise is mainly concentrated in the frequency band of 0.08 - 0.22 Hz. The simulated sea wave induced magnetic noise is added to the time series data of the H Figure 3b component of the simulated marine magnetotelluric field as shown in x , and the time series and amplitude spectrum of the simulated marine magnetotelluric data containing the sea wave induced magnetic noise are obtained, as shown in Figure 5a and 5b respectively. It can be seen from Figure 5b that a strong energy band appears near 0.1 Hz in the amplitude spectrum of the noisy data, that is, the sea wave electromagnetic interference.

[0057] Using the method of the present invention to suppress the noise of the synthesized magnetotelluric data containing the sea wave induced magnetic noise, through testing, the type of the structural element is determined to be triangular, and the size of the structural element is determined according to step S1 to obtain the optimal structural element. At the same time, through testing, the scale and the initial weighting coefficient are determined. Since too large a scale has little improvement on the result but increases the calculation cost, the scale should not be selected too large. The finally selected filtering parameters are shown in Table 1. The overall comparison of the time series and the detailed comparison of the time series of the H x component of the simulated marine magnetotelluric before and after noise suppression are respectively as shown in Figure 6a and6b As shown, the power spectrum comparison between the filtering result and the pure data is as follows Figure 6c As shown, the overall time series comparison between the simulated sea wave induced magnetic noise and the X component of the sea wave induced magnetic noise extracted by filtering is as follows Figure 7a As shown, the local time series comparison is as follows Figure 7b . From Figure 6a , 6b it can be seen that the error between the marine magnetotelluric data after noise suppression and the original marine magnetotelluric data is small. From Figure 6c it can be seen that near 0.1 Hz, the power spectral density anomaly envelope caused by the sea wave induced magnetic noise can be well suppressed. From Figure 7a - Figure 7b it can be seen that the extracted sea wave induced magnetic noise basically coincides with the simulated sea wave induced magnetic noise.

[0058] Table 1: , The changes in the signal-to-noise ratio SNR, root mean square error RMSE, and normalized cross-correlation coefficient NCC of the H x component of the simulated marine magnetotelluric data before and after noise suppression are shown in Table 2. The normalized cross-correlation coefficient between the extracted sea waves and the added sea wave induced magnetic noise also reaches 0.9832, indicating that the quality of the magnetotelluric data has been significantly improved after noise suppression.

[0059] Table 2: , To further verify the improvement of the marine magnetotelluric data after noise suppression, Robust impedance estimation is performed on the simulated marine magnetotelluric data, the simulated marine magnetotelluric data containing sea wave induced magnetic noise, and the marine magnetotelluric data after noise suppression obtained by the method of the present invention. The apparent resistivity and phase are calculated and compared with the calculation results of the empirical mode decomposition (EMD) method, as Figure 8 shown. It can be seen that before noise suppression, due to the influence of the sea wave induced magnetic noise, the magnetotelluric apparent resistivity and phase curves are severely distorted in the period range of 5 - 10 s. The EMD method can restore the distortion of the apparent resistivity curve to a certain extent, while after the multi-scale adaptive generalized morphological filtering denoising process, the magnitude of the apparent resistivity can be better restored, and the apparent resistivity curve and the phase curve are more continuous and reasonable.

[0060] Example 2: The method of the present invention is used to process the measured marine magnetotelluric data of a certain site in the South Yellow Sea to further verify the actual application effect of the method. In order to minimize the electromagnetic noise interference caused by tidal movement, the data during the slack tide period is selected for noise suppression. The time series and amplitude spectra of the two horizontal components of the electric field selected are respectively as Figure 9a~9dAs shown, the time series and amplitude spectra of the two horizontal magnetic field components are respectively as Figure 9e~9h shown. The sampling rate is 500 Hz and the sampling duration is 100 min. It can be seen from the amplitude spectrum that the horizontal electric field component is less affected by sea waves and can be assumed to be a noise-free signal; the horizontal magnetic field component is more affected by seawater movement, and noise interference appears in the amplitude spectrum within the frequency range of 0.08 - 0.3 Hz. Therefore, this application only performs noise suppression on the magnetic field component of the measured data.

[0061] Using the method of the present invention to Figure 9e and 9g process and analyze the measured data shown, the filtering parameters used for the H x component of the measured marine magnetotelluric data and the H y component of the measured marine magnetotelluric data are shown in Table 3. The overall time series comparison and local time series comparison of the H x component of the measured marine magnetotelluric data before and after noise suppression are respectively as Figure 10a~10b shown, and the power spectral density comparison is as Figure 10c shown. The overall time series comparison and local time series comparison of the H y component of the measured marine magnetotelluric data before and after noise suppression are respectively as Figure 10d and 10e shown, and the power spectral density comparison is as Figure 10f shown. It can be seen from the power spectral density that the abnormally strong energy of the H x component of the measured marine magnetotelluric data and the H y component within the range of 0.08 - 0.3 Hz is suppressed, and the energy of the power spectral density returns to the normal level. Figure 11a is the time-frequency diagram of the H x component of the measured marine magnetotelluric data before noise suppression, Figure 11b is the time-frequency diagram of the H x component of the measured marine magnetotelluric data after noise suppression, Figure 11c is the time-frequency diagram of the H y component of the measured marine magnetotelluric data before noise suppression, Figure 11d is the time-frequency diagram of the H y component of the measured marine magnetotelluric data after noise suppression. Similarly, it can be seen that the abnormally strong energy of the H y component of the measured marine magnetotelluric data after noise suppression and the H y component within the black frame is effectively suppressed.

[0062] Table 3: , Using the Robust impedance estimation method to estimate the impedance of the measured data before and after noise suppression, and calculate the apparent resistivity and phase. The results in the XY direction are as Figure 12a shown, and the results in the YX direction are as Figure 12b shown. It can be seen that under the influence of the sea wave induced magnetic field, large distortions occur in both the apparent resistivity curve and the phase curve between the periods of 3 to 10 s. The apparent resistivity curve and the phase curve after noise suppression are significantly improved and become more continuous and reasonable.

[0063] This application improves the multi-scale adaptive generalized morphological filtering. By setting evaluation indicators, the method of obtaining the structural element size is improved, making the obtained structural element more suitable for filtering marine magnetotelluric data; through the weighted linear combination of multi-scale generalized morphological opening-closing operation and multi-scale generalized morphological closing-opening operation, the filtering accuracy is improved, and the weighted coefficients of the multi-scale generalized morphological opening-closing operation and the weighted coefficients of the multi-scale generalized morphological closing-opening operation are iteratively calculated using the gradient method to update the weighted coefficients, further improving the accuracy of the multi-scale adaptive generalized morphological filter; after each iterative filtering, the filtered data is reconstructed as the input data for the next iteration, further improving the filtering accuracy; applying the parameter experience of the simulated data selection to the measured data to test the actual application effect of this method. In summary, using the method of this application significantly improves the filtering accuracy of marine magnetotelluric data, the noise suppression effect is significantly improved, and the quality of the magnetotelluric data after noise suppression is significantly improved.

[0064] Corresponding to the above embodiments, this application also provides a device for suppressing sea wave induced electromagnetic noise.

[0065] See Figure 13 , which is the structural block diagram of a device for suppressing sea wave induced electromagnetic noise provided by the embodiment of this application. As Figure 13 shown, it mainly includes the following modules.

[0066] Optimal structural element determination module 1301: Determine the type and size of the structural element to obtain the optimal structural element; Multi-scale generalized morphological operation module 1302: Perform multi-scale operations on the entire segment of the original marine magnetotelluric data using the optimal structural element to obtain the multi-scale opening-closing operation result and the multi-scale closing-opening operation result; Noise suppression module 1303: Perform a weighted linear combination of the multi-scale generalized morphological opening-closing operation and the multi-scale generalized morphological closing-opening operation, and use the gradient method to iteratively calculate the weighted coefficients of the multi-scale generalized morphological opening-closing operation and the weighted coefficients of the multi-scale morphological closing-opening operation to update the weighted coefficients, and obtain the entire segment of the marine magnetotelluric data after noise suppression; First noise suppression effect evaluation module 1304: Introduce the signal-to-noise ratioSNR , the root mean square error RMSE and the normalized cross - correlation coefficient NCC evaluate the noise suppression effect with these three indicators; The second noise suppression effect evaluation module 1305: Perform Robust impedance estimation on the ocean magnetotelluric data after suppressing the whole - section noise, calculate apparent resistivity and phase, and evaluate the apparent resistivity and phase data.

[0067] It should be noted that the specific content involved in the embodiments of this application can refer to the description of the above - mentioned method embodiments. For the sake of brevity, it will not be repeated here.

[0068] Corresponding to the above - mentioned embodiments, the embodiments of this application also provide an electronic device.

[0069] See Figure 14 , which is a schematic structural diagram of an electronic device provided by the embodiments of this application. As Figure 14 shown, the electronic device 1400 may include: a processor 1401, a memory 1402, and a communication unit 1403. These components communicate through one or more buses. Those skilled in the art can understand that the structure of the electronic device shown in the figure does not constitute a limitation on the embodiments of this application. It can be a bus - shaped structure, a star - shaped structure, and may also include more or fewer components than shown in the figure, or combine certain components, or have different component arrangements.

[0070] Among them, the communication unit 1403 is used to establish a communication channel so that the electronic device can communicate with other devices.

[0071] The processor 1401 is the control center of the electronic device. It uses various interfaces and lines to connect all parts of the electronic device, and by running or executing software programs and / or modules stored in the memory 1402, as well as calling data stored in the memory, it executes various functions of the electronic device and / or processes data. The processor can be composed of an integrated circuit (IC). For example, it can be composed of a single - packaged IC, or composed of multiple packaged ICs with the same or different functions connected together. For example, the processor 1401 may only include a central processing unit (CPU). In the embodiments of this application, the CPU can be a single - operation core or include multiple operation cores.

[0072] A memory 1402 for storing execution instructions of a processor 1401. The memory 1402 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disc.

[0073] When the execution instructions in the memory 1402 are executed by the processor 1401, the electronic device 1400 is enabled to execute some or all of the steps in the above method embodiments.

[0074] Corresponding to the above embodiments, an embodiment of the present application further provides a computer-readable storage medium. The computer-readable storage medium can store a program. When the program runs, it can control the device where the computer-readable storage medium is located to execute some or all of the steps in the above method embodiments. Specifically, the computer-readable storage medium can be a magnetic disk, an optical disc, a read-only memory (ROM), a random access memory (RAM), etc.

[0075] Corresponding to the above embodiments, an embodiment of the present application further provides a computer program product. The computer program product includes executable instructions. When the executable instructions are executed on a computer, the computer is enabled to execute some or all of the steps in the above method embodiments.

[0076] In the embodiments of the present application, "at least one" means one or more, and "a plurality" means two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships can exist. For example, A and / or B can represent the cases of A existing alone, A and B existing simultaneously, and B existing alone. Where A and B can be singular or plural. The character " / " generally represents an "or" relationship between the associated objects before and after. "At least one of the following" and its similar expressions refer to any combination of these items, including any combination of single items or plural items. For example, at least one of a, b, and c can represent: a, b, c, a - b, a - c, b - c, or a - b - c, where a, b, and c can be single or multiple.

[0077] Those of ordinary skill in the art can realize that the various units and algorithm steps described in the embodiments disclosed herein can be implemented by electronic hardware, computer software, or a combination of electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. A professional technician can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of this application.

[0078] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the systems, devices, and units described above can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0079] In several embodiments provided in this application, if any function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The foregoing storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs.

[0080] The above is only the specific implementation manner of this application. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed in this application, and all should be covered by the protection scope of this application. The protection scope of this application shall be subject to the protection scope of the claimed rights.

Claims

1. A method for suppressing electromagnetic noise induced by ocean waves, characterized in that, Including: Step S1: Determine the type and size of the structural element and obtain the optimal structural element; Step S2: Perform multi-scale operations on the entire segment of the original marine magnetotelluric data using the optimal structural element to obtain the multi-scale opening-closing operation result and the multi-scale closing-opening operation result; Step S3: Perform a weighted linear combination of the multi-scale generalized morphological opening-closing operation and the multi-scale generalized morphological closing-opening operation, and use the gradient method to iteratively calculate the weighted coefficients of the multi-scale generalized morphological opening-closing operation and the weighted coefficients of the multi-scale morphological closing-opening operation to update the weighted coefficients, and obtain the marine magnetotelluric data after noise suppression for the entire segment; Step S4: Introduce the signal-to-noise ratio SNR , the root mean square error RMSE and the normalized cross-correlation coefficient NCC are used to evaluate the noise suppression effect by three indicators; Step S5: Perform Robust impedance estimation on the marine magnetotelluric data after noise suppression for the entire segment, calculate the apparent resistivity and phase, and evaluate the apparent resistivity and phase data.

2. The method according to claim 1, characterized in that, The determination of the type and size of the structural element and obtaining the optimal structural element includes: According to the marine magnetotelluric data and the characteristics of the sea wave-induced electromagnetic noise, set the type of the optimal structural element as a triangular structural element; For the determination of the structural element size, it is achieved through the following steps: S11: Perform preliminary filtering to determine the structural element size range; S12: Input the entire segment of ideal marine magnetotelluric data, perform multi-scale generalized morphological filtering to traverse all structural elements within the structural element size range, and obtain the marine magnetotelluric data after multi-scale generalized morphological filtering for the entire segment and the marine magnetotelluric data after multi-scale generalized morphological filtering for the frequency band significantly affected by sea wave noise; According to the entire segment of ideal marine magnetotelluric data, the marine magnetotelluric data after multi-scale generalized morphological filtering for the entire segment, the ideal marine magnetotelluric data for the frequency band significantly affected by sea wave noise, and the marine magnetotelluric data after multi-scale generalized morphological filtering for the frequency band significantly affected by sea wave noise, set the evaluation index and obtain the optimal structural element size under the optimal evaluation index.

3. The method according to claim 2, wherein The evaluation index is: ， Among them, C is the evaluation index, is the normalized cross-correlation coefficient between the entire ideal marine magnetotelluric data and the time series of the marine magnetotelluric data after multi-scale generalized morphological filtering for the entire segment, is the normalized cross-correlation coefficient between the ideal marine magnetotelluric data in the frequency band greatly affected by ocean wave noise and the amplitude spectrum of the marine magnetotelluric data after multi-scale generalized morphological filtering in the frequency band greatly affected by ocean wave noise: ， Among them, represents the entire ideal marine magnetotelluric data, represents the marine magnetotelluric data after multi-scale generalized morphological filtering for the entire segment, is the sampling point of the time series data, represents the length of the time series data, represents the length of the frequency band affected by ocean waves in the amplitude spectrum.

4. The method according to claim 1, wherein The said Step S2 includes: The generalized morphological filter includes a generalized morphological opening-closing filter and a generalized morphological closing-opening filter, and the generalized morphological opening-closing filter and the generalized morphological closing-opening filter are: ， Among them, the input signal is the entire original marine magnetotelluric data; is the sampling point of the time series data, , represents the length of the time series data, and represent structural elements of different sizes; the symbol represents the opening operation, and the symbol represents the closing operation; GOC represents the generalized morphological opening-closing filter, and GCO represents the generalized morphological closing-opening filter, ; Based on the generalized morphological filter, multiple erosions and multiple dilations are performed on the structural element to obtain multi-scale generalized morphological operations, including multi-scale generalized morphological opening and multi-scale generalized morphological closing which are defined as: ， Among them, is the entire original marine magnetotelluric data; and represent structural elements of different sizes. The symbol represents the dilation operation, and the symbol represents the erosion operation. s represents the scale of the multi-scale generalized morphological operation. is the sampling point of the time series data, and represents the length of the time series data. Based on multi-scale generalized morphological operations, multi-scale generalized morphological opening-closing operations are obtained and multi-scale generalized morphological closing-opening operations : ， Among them, is the entire original marine magnetotelluric data; and represent structural elements of different sizes, s represents the scale of the multi-scale generalized morphological operation, is the sampling point of the time series data, represents the length of the time series data.

5. The method according to claim 4, wherein The weighted linear combination of the multi-scale generalized morphological opening-closing operation and the multi-scale generalized morphological closing-opening operation includes: Perform a weighted linear combination of the multi-scale generalized morphological opening-closing operation and the multi-scale generalized morphological closing-opening operation to obtain the marine magnetotelluric data after suppressing the entire section of noise. It is: ， Among them, is the weighted coefficient of the multi-scale generalized morphological opening-closing operation , is the weighted coefficient of the multi-scale morphological closing-opening operation , is the sampling point of the time series data; Hypothesis , , then the above formula can be expressed as: ， Among them, is a weighting coefficient, , is a sampling point of time series data.

6. The method according to claim 5, wherein The use of the gradient method to iteratively calculate the weighted coefficients of the multi-scale generalized morphological opening-closing operation and the weighted coefficients of the multi-scale morphological closing-opening operation to update the weighted coefficients and obtain the marine magnetotelluric data after noise suppression for the entire segment includes: Step S31: Perform multi-scale generalized morphological opening-closing operation on the input signal to obtain , perform multi-scale closing-opening operation to obtain , then the output of the multi-scale generalized morphological filter after the k-th iteration is: ， Among them, is the weighting coefficient for the k-th iteration, , is the sampling point of the time series data, , represents the length of the time series data; Step S32: Obtain a single error sample after the k-th iteration : ， Among them, is the output of the multi-scale generalized morphological filter after k - 1 iterations, is the output of the multi-scale generalized morphological filter after k iterations, is the sampling point of the time series data, , represents the length of the time series data, then the gradient is: ， ， Among them, is the square of a single error sample at the k-th iteration for the weighted coefficient gradient, is the square of a single error sample at the k-th iteration for the weighted coefficient gradient; Step S33: Calculate the direction vector : ， Among them, is the first component of the direction vector at the k-th iteration, is the second component of the direction vector at the k-th iteration, is the -th component of the direction vector at the k-th iteration, is the -th component of the direction vector at the (k - 1)-th iteration, is the square of a single error sample at the k-th iteration with respect to the weighting coefficient gradient, is the square of a single error sample at the (k - 1)-th iteration with respect to the weighting coefficient gradient; Step S34: Calculate the weighted coefficients: ， Among them, is the weighting coefficient for the (k + 1)-th iteration, is the weighting coefficient for the k-th iteration, is the direction vector for the k-th iteration, , is the step size parameter.

7. The method according to claim 1, wherein The said Step S3 also includes: The multi-scale adaptive generalized morphological filtering is an iterative filtering process. In the iterative filtering process, the initial input signal is the entire segment of the original marine magnetotelluric data. After each iterative filtering, data reconstruction is performed on the filtered marine magnetotelluric data. The data reconstruction is to perform band-pass filtering on the filtered marine magnetotelluric data, and only retain the marine magnetotelluric data in the frequency band that is more affected by ocean wave noise in the filtered marine magnetotelluric data, which is denoted as the intermediate marine magnetotelluric data in the frequency band more affected by ocean wave noise. The marine magnetotelluric data in the remaining frequency bands is the same as the original marine magnetotelluric data. The intermediate marine magnetotelluric data in the frequency band more affected by ocean wave noise and the marine magnetotelluric data in the remaining frequency bands are subjected to data reconstruction to obtain the reconstructed marine magnetotelluric data after each iterative filtering; then the reconstructed marine magnetotelluric data is used as the input signal to repeat the iterative filtering process; when the iterative stop condition is met, the marine magnetotelluric data with noise suppression in the frequency band more affected by ocean wave noise and the entire segment of marine magnetotelluric data with noise suppression are obtained.

8. The method according to claim 1, wherein The signal-to-noise ratio SNR , the root mean square error RMSE and the normalized cross-correlation coefficient NCC are respectively: ， Among them, represents the entire original marine magnetotelluric data, represents the marine magnetotelluric data after suppressing the entire noise, is the sampling point of the time series data, represents the length of the time series data.

9. An electromagnetic noise suppression device for wave sensing, characterized in that, including: Optimal structure element determination module: Determine the type and size of the structure element to obtain the optimal structure element; Multi-scale generalized morphological operation module: Perform multi-scale operations on the entire segment of the original marine magnetotelluric data using the optimal structure element to obtain the multi-scale opening-closing operation result and the multi-scale closing-opening operation result; Noise suppression module: Perform a weighted linear combination of the multi-scale generalized morphological opening-closing operation and the multi-scale generalized morphological closing-opening operation, and use the gradient method to iteratively calculate and update the weighted coefficients of the multi-scale generalized morphological opening-closing operation and the weighted coefficients of the multi-scale morphological closing-opening operation to obtain the entire segment of marine magnetotelluric data with noise suppression; The first noise suppression effect evaluation module: introduce the signal-to-noise ratio SNR , the root mean square error RMSE and the normalized cross-correlation coefficient NCC to evaluate the noise suppression effect with three indicators; Second noise suppression effect evaluation module: Perform Robust impedance estimation on the entire segment of marine magnetotelluric data with noise suppression, calculate apparent resistivity and phase, and evaluate the apparent resistivity and phase data.

10. An electronic device, characterized in that, including: Processor; Memory; And a computer program, where the computer program is stored in the memory, and the computer program includes instructions that, when executed by the processor, cause the electronic device to execute the method according to any one of claims 1 to 8.

Images