CN-121995506-A - Ocean controllable source electromagnetic signal denoising method

CN121995506ACN 121995506 ACN121995506 ACN 121995506ACN-121995506-A

Abstract

The invention discloses a denoising method for ocean controllable source electromagnetic signals, which relates to the technical field of ocean geophysical exploration and signal processing, and comprises the following steps: and reading an original ocean controllable source electromagnetic CSEM time sequence signal, and performing self-adaptive framing treatment to obtain the current frame length and frame movement. The invention adopts the variable step size minimum mean square self-adaptive filtering algorithm, the step size factor of the variable step size minimum mean square self-adaptive filtering algorithm can be dynamically adjusted according to noise components, the variable step size mechanism based on a noise power function is introduced, the contradiction between the convergence speed and the steady-state precision of the traditional fixed step size algorithm is effectively solved, the frame length is dynamically adjusted by tracking the noise change in real time through the self-adaptive framing mechanism, the tracking capability of the algorithm on the time-varying noise is further improved, and the noise removal robustness and the noise removal adaptability under the complex marine environment are remarkably enhanced.

Inventors

LI ZELIN
CHEN HUI
LI QIANG
WANG XIANXIANG
LI HAILIN
LI JINSONG

Assignees

东华理工大学

Dates

Publication Date: 20260508
Application Date: 20251219

Claims (10)

1. The marine controllable source electromagnetic signal denoising method is characterized by comprising the following steps of: s1, reading an original ocean controllable source electromagnetic CSEM time sequence signal, and performing self-adaptive framing treatment to obtain a current frame length and frame movement; s2, based on the current frame length and the signal section divided in the frame shift, taking the signal section as a reference input signal of the adaptive filter after unit delay; S3, denoising each frame of signal by adopting a variable-step-length minimum mean square VSS-LM adaptive filtering algorithm, wherein a step-length factor is dynamically adjusted according to noise components, and an error signal after denoising of each frame is output; And S4, overlapping and reconstructing the error signals processed by all frames according to frame shift to obtain a complete denoised time sequence signal.
2. The method for denoising electromagnetic signals of marine controllable source according to claim 1, wherein in the first step, the adaptive framing process comprises the steps of: s11, calculating noise characterization parameters by adopting a sliding window, wherein the noise characterization parameters comprise local variance and short-time signal-to-noise ratio; s12, triggering frame length adjustment when the variation of the noise characterization parameter in a plurality of continuous sampling points exceeds a preset threshold; S13, dynamically determining the length of the current frame within the preset minimum frame length and maximum frame length range according to the current noise characterization parameters.
3. The method for denoising electromagnetic signals of marine controllable source according to claim 2, wherein in step S13, the method for dynamically determining the length of the current frame is as follows: If the noise is changed drastically, the minimum frame length is adopted; when the noise is stable, the maximum frame length is adopted; When the noise is moderately changed, the frame length is dynamically determined between the minimum frame length and the maximum frame length through linear interpolation; The minimum frame length is not less than twice the order of the adaptive filter, the maximum frame length is not more than eight times the order of the adaptive filter, and the frame shift is always half of the current frame length.
4. The method for denoising the electromagnetic signal of the ocean controllable source according to claim 1, wherein in the step S3, denoising each frame signal by adopting a variable step-size minimum mean square VSS-LM adaptive filtering algorithm comprises the following steps: s31, at each sampling moment, calculating an output signal of the adaptive filter by using the weight coefficient of the adaptive filter updated at the moment; s32, subtracting the original ocean controllable source electromagnetic time sequence signal from the output signal of the adaptive filter to obtain an error signal; S33, performing zero-phase FIR low-pass filtering on the error signal, and separating an effective signal component and a noise component; s34, updating a step factor based on a power function form of a noise component, and updating a weight coefficient of the adaptive filter by using the step factor for calculating the next sampling moment; and S35, calibrating the cut-off frequency of the low-pass filter through STFT every 10 frames.
5. The method for denoising an electromagnetic signal of marine controllable source according to claim 4, wherein in S33, zero-phase FIR low-pass filtering is designed as follows: the cut-off frequency of the zero-phase FIR low-pass filter needs to dynamically track the lowest frequency of the ocean controllable source electromagnetic effective signal and is set to be half of the lowest frequency; The order of the low-pass filter is set to be half of the order of the adaptive filter; the filtering process is performed by combining the forward direction and the backward direction.
6. The method for denoising electromagnetic signals of marine controllable source according to claim 4, wherein in S35, the low-pass filter cut-off frequency is calibrated by STFT every 10 frames, comprising the following steps: S351, carrying out short-time Fourier transform on the error signal, wherein the length of a used time window is consistent with the frame length of the current self-adaptive framing, and calculating the power spectrum density of the signal; s352, extracting peak frequency from the power spectrum density, determining the lowest frequency of the effective signal based on the peak frequency, and updating the cut-off frequency of the low-pass filter according to the lowest frequency; S353, the calibration of the cut-off frequency is performed in a fixed unit of every 10 frames, and the update is performed once every predetermined number of data frames are processed.
7. The method of denoising an electromagnetic signal of marine controllable source of claim 1, wherein the adaptive filter is a transversal finite impulse response filter.
8. The method for denoising the electromagnetic signal of the ocean controllable source according to claim 1, wherein in the step S4, the error signals after all frame processing are overlapped and reconstructed according to frame shift to obtain a complete denoised time sequence signal, specifically: When the frame length is switched from a long frame to a short frame, overlapping a plurality of sampling points at the tail of the previous long frame with equivalent sampling points at the head of the current short frame, and carrying out weighted average on signal values of overlapping parts, wherein the weights of the signal values are in linear transition from the previous frame to the current frame; When the frame length is switched from a short frame to a long frame, the overlapping and weighted average processing is carried out by adopting the mode, and finally the complete denoised time sequence signal is obtained.
9. The method for denoising an electromagnetic signal of marine controllable source according to claim 2, wherein the number of consecutive plurality of sampling points in S12 is equal to half the order of the adaptive filter.
10. The method for denoising electromagnetic signals of marine controllable source according to claim 4, wherein in S34, a minimum threshold constraint is set for the step size factor, and it is ensured that the step size factor is not lower than the minimum threshold constraint.

Description

Ocean controllable source electromagnetic signal denoising method Technical Field The invention relates to the technical field of marine geophysical exploration and signal processing, in particular to a marine controllable source electromagnetic signal denoising method. Background Ocean Controlled Source Electromagnetic (CSEM) is one of the key geophysical techniques for detecting subsea reservoirs. By towing an artificial electromagnetic source and transmitting low frequency electromagnetic signals to the sea floor, and then recording the electric and magnetic field components in a receiver array deployed on the sea floor. Due to complex marine environments (such as sea wave motion, sea vortex, thermal noise of instruments, ship interference and the like), the received signals are inevitably mixed with random noise with high intensity and complex characteristics, and the noise seriously floods the primary field response signals, so that the extraction of effective signals becomes extremely difficult. In the prior art, the traditional frequency domain filtering (such as band-pass filtering) is simple and easy to implement, but requires that the frequency bands of the effective signals and the noise are completely separated, however, the CSEM effective signals and the ocean environment noise have serious frequency band overlapping, a large amount of low-frequency effective signals are lost and Gibbs phenomenon is introduced by using the method, the wavelet threshold denoising method has good time-frequency localization characteristics, but the performance of the wavelet threshold denoising method is highly dependent on the selection of wavelet basis functions, decomposition layers and threshold functions, the adaptivity is lacked, the signal processing effects on different work areas and different noise levels are unstable, and pseudo Gibbs phenomenon is easily generated near signal singular points, and the method based on Principal Component Analysis (PCA) or Singular Value Decomposition (SVD) is suitable for removing certain coherent noise, but has limited denoising effect on the ocean environment noise with non-stability and strong randomness, and has large calculated amount and is unfavorable for real-time processing. Therefore, a denoising method for ocean controllable source electromagnetic signals is provided to solve the problems. Disclosure of Invention The invention mainly aims to provide a denoising method for ocean controllable source electromagnetic signals, which aims to solve the problems in the background. In order to achieve the purpose, the technical scheme adopted by the invention is that the ocean controllable source electromagnetic signal denoising method comprises the following steps: s1, reading an original ocean controllable source electromagnetic CSEM time sequence signal, and performing self-adaptive framing treatment to obtain a current frame length and frame movement; s2, based on the current frame length and the signal section divided in the frame shift, taking the signal section as a reference input signal of the adaptive filter after unit delay; S3, denoising each frame of signal by adopting a variable-step-length minimum mean square VSS-LM adaptive filtering algorithm, wherein a step-length factor is dynamically adjusted according to noise components, and an error signal after denoising of each frame is output; And S4, overlapping and reconstructing the error signals processed by all frames according to frame shift to obtain a complete denoised time sequence signal. Preferably, in the first step, the adaptive framing process includes the following steps: s11, calculating noise characterization parameters by adopting a sliding window, wherein the noise characterization parameters comprise local variance and short-time signal-to-noise ratio; s12, triggering frame length adjustment when the variation of the noise characterization parameter in a plurality of continuous sampling points exceeds a preset threshold; S13, dynamically determining the length of the current frame within the preset minimum frame length and maximum frame length range according to the current noise characterization parameters. Preferably, in the step S13, the method for dynamically determining the length of the current frame is as follows: If the noise is changed drastically, the minimum frame length is adopted; when the noise is stable, the maximum frame length is adopted; When the noise is moderately changed, the frame length is dynamically determined between the minimum frame length and the maximum frame length through linear interpolation; The minimum frame length is not less than twice the order of the adaptive filter, the maximum frame length is not more than eight times the order of the adaptive filter, and the frame shift is always half of the current frame length. Preferably, in the step S3, the denoising processing for each frame signal by adopting a variable step-size minimum mean square VSS-LM ad