Method and system for reducing noise of hydraulic fracturing microseismic signal of coal seam roof

Abstract

The invention provides a coal seam roof hydraulic fracturing micro-seismic signal noise reduction method and system, and the method comprises the steps: decomposing a coal seam roof hydraulic fracturing micro-seismic signal into a plurality of band-limited intrinsic mode components through variational mode decomposition, and carrying out the continuous synchronous compression wavelet transform analysis of each band-limited intrinsic mode component through effectiveness judgment, performing classification denoising processing on the synchronous compression wavelet transform time-frequency diagram according to different denoising strategies; according to instantaneous frequency distribution and local energy distribution of the synchronous compression wavelet transform time-frequency diagram, segmenting different time-frequency regions of the synchronous compression wavelet transform time-frequency diagram, and according to a time-frequency region segmentation result, picking up first arrival time of a micro-seismic waveform of top plate hydraulic fracturing; according to the method, denoising modal components are screened by using first arrival time, and effective modal components are reconstructed, so that signal-noise separation and high-performance noise reduction are performed on hydraulic fracturing micro-seismic signals with low signal-to-noise ratio and unsteady state characteristics, a basis is provided for accurate prevention and control of a mine with frequent occurrence of mine earthquakes, and safe and efficient mining of a coal mine is improved.

Description

Technical Field

[0001] The present invention relates to the field of mine prevention and control technology, and in particular to a method and system for reducing noise of microseismic signals of hydraulic fracturing of a coal seam roof. Background Art

[0002] The key layer of mine earthquake (the main controlling rock layer causing disasters) has the characteristics of large thickness and good integrity. It is difficult to collapse after coal mining and is transformed into an "elastic energy storage body". Since the movement and fracture of the key layer play a decisive role in the occurrence of mine earthquakes, hydraulic fracturing technology is a green technology for anti-bumping, elimination of sudden outbursts and gas extraction in the mining field. Microseismic monitoring of the hydraulic fracturing process of the coal mine roof can realize the tracking of fracturing cracks and effect evaluation. For the microseismic signal of hydraulic fracturing of the coal seam roof, its generation mechanism, propagation path and surrounding environment are complex, and the noise sources mixed in the signal are numerous and unstable, which greatly affects the accuracy of tracking the fracturing cracks and effect evaluation. Traditional noise reduction methods include time domain methods and frequency domain methods. Different methods have corresponding applicability and limitations in different scenarios. For example, the time domain method has high requirements for signal and noise separation and is not suitable for complex noise environments; the frequency domain method can effectively separate spectrally overlapping signals and noise, but spectral leakage or distortion will be introduced during the processing; the wavelet transform is suitable for non-stationary signals and can provide time and frequency information at the same time, but the wavelet basis selection and scale analysis are relatively complex; the adaptive filtering method can automatically adjust the filter parameters according to the characteristics of the signal and noise, but the algorithm is highly complex and requires more computing resources and longer processing time.

[0003] For the hydraulic fracturing microseismic signals of coal seam roof, their generation mechanism, propagation path and complex surrounding environment are mixed with numerous and unstable noise sources. Traditional noise reduction methods are difficult to achieve signal-to-noise separation and high-performance noise reduction for hydraulic fracturing microseismic signals with low signal-to-noise ratio and non-steady-state characteristics. Summary of the invention

[0004] The present invention provides a method and system for reducing noise of hydraulic fracturing microseismic signals of coal seam roof, so as to solve the defect that traditional noise reduction methods are difficult to achieve signal-noise separation and high-performance noise reduction of hydraulic fracturing microseismic signals with low signal-to-noise ratio and non-steady-state characteristics.

[0005] The present invention provides a method for reducing noise of microseismic signals of hydraulic fracturing of coal seam roof, comprising: The microseismic signal of hydraulic fracturing in the coal seam roof is decomposed into multiple band-limited intrinsic mode components by variational mode decomposition. A continuous synchronous compression wavelet transform analysis is performed on each band-limited intrinsic mode component that passes the validity judgment, and the synchronous compression wavelet transform time-frequency diagram of the corresponding band-limited intrinsic mode component is obtained. Based on wavelet threshold denoising, the synchronous compression wavelet transform time-frequency diagram is subjected to classification denoising according to different denoising strategies to obtain denoised modal components; According to the instantaneous frequency distribution and local energy distribution of the synchronous compression wavelet transform time-frequency diagram, different time-frequency regions of the synchronous compression wavelet transform time-frequency diagram are segmented, and the first arrival time of the microseismic waveform of the roof hydraulic fracturing is picked up according to the time-frequency region segmentation result; The denoised modal components are screened using the first arrival time of the microseismic waveform of the roof hydraulic fracturing to obtain effective modal components, and the effective modal components are reconstructed to obtain a reconstructed microseismic signal of the coal seam roof hydraulic fracturing.

[0006] According to the method for reducing noise of microseismic signals of hydraulic fracturing of coal seam roof provided by the present invention, the method decomposes the microseismic signals of hydraulic fracturing of coal seam roof into multiple band-limited eigenmode components by variational mode decomposition, including: When performing variational mode decomposition, the coal seam roof hydraulic fracturing microseismic signal is decomposed into multiple band-limited eigenmode components according to the following formula; In the formula, for The time corresponding to band-limited eigenmode components, for The instantaneous phase corresponding to the moment, for The instantaneous amplitude corresponding to the moment; ; is a positive integer; Sure Moment Band-limited eigenmode components The instantaneous frequency : According to the instantaneous frequency A plurality of band-limited eigenmode components are obtained, and each band-limited eigenmode component contains different proportions of effective components and noise components.

[0007] According to the method for reducing noise of microseismic signals of hydraulic fracturing of coal seam roof provided by the present invention, the effectiveness judgment includes: The microseismic signal of hydraulic fracturing of the coal seam roof is obtained according to the following formula Microseismic signals of hydraulic fracturing in coal seam roof No. Band-limited eigenmode components The waveform correlation coefficient between and variance contribution : In the formula, is the preset similarity threshold, for Moment The microseismic signals of hydraulic fracturing in the coal seam roof are collected by the signal collection channels; is the number of collected microseismic signals of hydraulic fracturing in the coal seam roof; ; is a positive integer; Microseismic signal of hydraulic fracturing in coal seam roof Microseismic signals of hydraulic fracturing in coal seam roof No. Band-limited eigenmode components The waveform correlation coefficient between them; Microseismic signal of hydraulic fracturing in coal seam roof No. Band-limited eigenmode components Microseismic signals of hydraulic fracturing in coal seam roof The variance contribution rate of Microseismic signal of hydraulic fracturing in coal seam roof The total number of band-limited eigenmode components obtained after decomposition; The number of effective band-limited eigenmode components to be determined for adaptive judgment; By waveform correlation coefficient , variance contribution rate and preset similarity threshold , for the microseismic signal of hydraulic fracturing in coal seam roof No. Band-limited eigenmode components The validity of the band-limited eigenmode component is judged. When the band-limited eigenmode component meets the judgment condition, the corresponding band-limited eigenmode component is judged to be a valid component; otherwise, the corresponding band-limited eigenmode component is judged to be an invalid component.

[0008] According to the method for reducing noise of microseismic signals of hydraulic fracturing of coal seam roof provided by the present invention, the method of continuously performing synchronous compression wavelet transform analysis on each band-limited intrinsic mode component that passes the effectiveness judgment, and obtaining the synchronous compression wavelet transform time-frequency diagram of the corresponding band-limited intrinsic mode component, comprises: Compression frequency by sync Concentrate the signal energy to the corresponding frequency, and through the band-limited characteristics, synchronize the compression frequency Select the corresponding frequency interval and extract the frequency component of the response; The corresponding frequency components in the frequency range are reversed through the inverse process of wavelet transform to obtain the corresponding band-limited eigenmode components. The band-limited characteristics are used to perform secondary screening on the modal signals in each frequency range, and the wavelet transform coefficients in the time-frequency domain are inversely transformed back to the time domain. The reconstructed modes are displayed in the time-frequency domain, and the synchronous compressed wavelet transform time-frequency diagram of the band-limited eigenmode components is generated.

[0009] According to the method for denoising microseismic signals of hydraulic fracturing of coal seam roof provided by the present invention, the wavelet threshold denoising is based on which the synchronous compression wavelet transform time-frequency graph is subjected to classified denoising processing according to different denoising strategies, including: After thresholding the frequency domain wavelet transform coefficients, the thresholded value is obtained. , and classify and denoise the frequency bandwidth of the band-limited eigenmode components: for The frequency domain wavelet transform coefficients , and perform corresponding attenuation; for It is determined to be noise and directly reset to zero; for (Right now ), then smoothing is performed using a nonlinear function.

[0010] According to the method for denoising microseismic signals of hydraulic fracturing of coal seam roof provided by the present invention, the wavelet threshold denoising is based on which the synchronous compression wavelet transform time-frequency graph is subjected to classification denoising according to different denoising strategies, and further comprises: Different threshold processing methods are selected through the frequency bandwidth characteristics of the band-limited eigenmode components. For narrowband frequencies, the soft threshold method is used to smooth the wavelet coefficients, retain the main information of the wavelet coefficients, and gradually reduce the impact of noise on the signal to retain more signal features. For broadband frequencies, the difference between signal and noise is significant, and the hard threshold method is used to reduce the signal below the cutoff value. The signal part corresponding to the wavelet coefficient is directly set to zero to remove the noise.

[0011] According to the method for reducing noise of microseismic signals of hydraulic fracturing of coal seam roof provided by the present invention, the segmentation of different time-frequency regions of the synchronous compression wavelet transform time-frequency graph according to the instantaneous frequency distribution and local energy distribution of the synchronous compression wavelet transform time-frequency graph comprises: Determine the minimum weighted within-class variance , to perform time-frequency region segmentation on the synchronous compression wavelet transform time-frequency image after classification denoising, and obtain the shock wave characteristic area and noise area in the SS-CWT time-frequency image of the band-limited eigenmode component; By minimizing the weighted intra-class variance Determine the best segmentation point of seismic wave characteristics and noise to perform time-frequency region segmentation on the synchronous compression wavelet transform time-frequency image after classification denoising, where the seismic wave characteristic area is the amplitude accumulation value ( ) is greater than the preset segmentation threshold and shows consistency in the intra-class variance; while the energy distribution in the noise area is scattered and unconcentrated; the segmentation threshold is dynamically adjusted according to the energy density of different areas.

[0012] According to the method for reducing noise of microseismic signals of hydraulic fracturing of coal seam roof provided by the present invention, the first arrival time of the microseismic waveform of hydraulic fracturing of coal seam roof is picked up according to the time-frequency region segmentation result, including: Restoring the seismic wave characteristic region into a time-amplitude time series; The AIC arrival time picking method is used to determine the peak value of the time-amplitude time series. The time point corresponding to the peak value is the first arrival time of the microseismic waveform.

[0013] According to the method for reducing noise of microseismic signals of hydraulic fracturing of coal seam roof provided by the present invention, the effective modal components are reconstructed to obtain the reconstructed microseismic signals of hydraulic fracturing of coal seam roof, including: Through the first arrival time of the microseismic waveform, the effective modal component is screened as the characteristic area of ​​the seismic wave containing strong signal energy; The seismic wave characteristic regions containing relatively strong signal energy are added together to obtain a reconstructed microseismic signal of hydraulic fracturing of the coal seam roof.

[0014] The present invention also provides a coal seam roof hydraulic fracturing microseismic signal noise reduction system, comprising: A decomposition module is used to decompose the coal seam roof hydraulic fracturing microseismic signal into multiple band-limited intrinsic mode components through variational mode decomposition, and perform continuous synchronous compression wavelet transform analysis on each band-limited intrinsic mode component that passes the validity judgment to obtain the synchronous compression wavelet transform time-frequency diagram of the corresponding band-limited intrinsic mode component; A denoising module, which is used for performing classification denoising on the synchronous compression wavelet transform time-frequency diagram based on wavelet threshold denoising according to different denoising strategies to obtain denoised modal components; A segmentation module is used to segment different time-frequency regions of the synchronous compression wavelet transform time-frequency graph according to the instantaneous frequency distribution and local energy distribution of the synchronous compression wavelet transform time-frequency graph, and pick up the first arrival time of the microseismic waveform of the roof hydraulic fracturing according to the time-frequency region segmentation result; The reconstruction module is used to screen the denoised modal components using the first arrival time of the microseismic waveform of the roof hydraulic fracturing to obtain effective modal components, reconstruct the effective modal components, and obtain the reconstructed microseismic signal of the coal seam roof hydraulic fracturing.

[0015] The present invention provides a method and system for reducing noise of microseismic signals of hydraulic fracturing of coal seam roof. The method and system decompose the microseismic signals of hydraulic fracturing of coal seam roof into multiple band-limited intrinsic mode components through variational mode decomposition, and perform continuous synchronous compression wavelet transform analysis on each band-limited intrinsic mode component that passes the validity judgment to obtain a synchronous compression wavelet transform time-frequency diagram of the corresponding band-limited intrinsic mode component; based on wavelet threshold denoising, the synchronous compression wavelet transform time-frequency diagram is classified and denoised according to different denoising strategies to obtain denoised mode components; according to the instantaneous frequency distribution and local energy distribution of the synchronous compression wavelet transform time-frequency diagram, different time-frequency regions of the synchronous compression wavelet transform time-frequency diagram are segmented, and the time-frequency regions are divided according to the time-frequency regions. The regional segmentation result picks up the first arrival time of the microseismic waveform of the roof hydraulic fracturing; the denoised modal component is screened by using the first arrival time of the microseismic waveform of the roof hydraulic fracturing to obtain the effective modal component, and the effective modal component is reconstructed to obtain the reconstructed microseismic signal of the coal seam roof hydraulic fracturing, thereby realizing signal-to-noise separation and high-performance noise reduction of the hydraulic fracturing microseismic signal with low signal-to-noise ratio and non-steady-state characteristics, providing a basis for the precise prevention and control of mines with frequent mine earthquakes, effectively avoiding the potential threat of sudden mine earthquakes caused by fault-type mine earthquakes in key layers during mining and inducing rock bursts, realizing the precise prevention and control of rock bursts and the support of disaster source management, reducing the cost of rock burst prevention and control, and improving the safe and efficient mining of coal mines. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0017] Figure 1 It is a schematic flow chart of a method for reducing noise of microseismic signals of hydraulic fracturing of a coal seam roof provided by an embodiment of the present invention; Figure 2 It is a logic principle diagram of a method for reducing noise of microseismic signals of hydraulic fracturing of a coal seam roof provided by an embodiment of the present invention; Figure 3 Schematic diagram of variational mode decomposition of microseismic signals of hydraulic fracturing of coal seam roof provided by an embodiment of the present invention; Figure 4 It is a schematic diagram of time-frequency characteristic analysis of microseismic signals of hydraulic fracturing of coal seam roof provided by an embodiment of the present invention; Figure 5 It is a schematic diagram of the arrival time of the microseismic signal of hydraulic fracturing of the coal seam roof provided by an embodiment of the present invention; Figure 6 is a time-frequency schematic diagram of a reconstructed signal provided by an embodiment of the present invention; Figure 7It is a structural schematic diagram of a coal seam roof hydraulic fracturing microseismic signal noise reduction system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0019] Figure 1 A flow chart of a method for reducing noise of microseismic signals of hydraulic fracturing of a coal seam roof provided by an embodiment of the present invention is shown in FIG. Figure 1 As shown, the method for reducing noise of microseismic signals of hydraulic fracturing of coal seam roof provided by the embodiment of the present invention comprises: Step 101, decomposing the coal seam roof hydraulic fracturing microseismic signal into multiple band-limited intrinsic mode components by variational mode decomposition, performing continuous synchronous compression wavelet transform analysis on each band-limited intrinsic mode component that passes the validity judgment, and obtaining a synchronous compression wavelet transform time-frequency diagram of the corresponding band-limited intrinsic mode component; Step 102, based on wavelet threshold denoising, the synchronous compression wavelet transform time-frequency diagram is subjected to classification denoising according to different denoising strategies to obtain denoised modal components; Step 103: segment different time-frequency regions of the synchronous compression wavelet transform time-frequency graph according to the instantaneous frequency distribution and local energy distribution of the synchronous compression wavelet transform time-frequency graph, and pick up the first arrival time of the microseismic waveform of the roof hydraulic fracturing according to the time-frequency region segmentation result; Step 104: screen the denoised modal components using the first arrival time of the microseismic waveform of the roof hydraulic fracturing to obtain effective modal components, reconstruct the effective modal components, and obtain a reconstructed microseismic signal of the coal seam roof hydraulic fracturing.

[0020] The method for denoising microseismic signals of hydraulic fracturing of coal seam roof provided by an embodiment of the present invention decomposes the microseismic signals of hydraulic fracturing of coal seam roof into multiple band-limited intrinsic mode components through variational mode decomposition, performs continuous synchronous compression wavelet transform analysis on each band-limited intrinsic mode component that passes the validity judgment, and obtains a synchronous compression wavelet transform time-frequency diagram of the corresponding band-limited intrinsic mode component; based on wavelet threshold denoising, the synchronous compression wavelet transform time-frequency diagram is classified and denoised according to different denoising strategies to obtain denoised modal components; according to the instantaneous frequency distribution and local energy distribution of the synchronous compression wavelet transform time-frequency diagram, different time-frequency regions of the synchronous compression wavelet transform time-frequency diagram are segmented, and the time-frequency regions are divided according to the time-frequency regions. The regional segmentation result picks up the first arrival time of the microseismic waveform of the roof hydraulic fracturing; the denoised modal component is screened by using the first arrival time of the microseismic waveform of the roof hydraulic fracturing to obtain the effective modal component, and the effective modal component is reconstructed to obtain the reconstructed microseismic signal of the coal seam roof hydraulic fracturing, thereby realizing signal-to-noise separation and high-performance noise reduction of the hydraulic fracturing microseismic signal with low signal-to-noise ratio and non-steady-state characteristics, providing a basis for the precise prevention and control of mines with frequent mine earthquakes, effectively avoiding the potential threat of sudden mine earthquakes caused by fault-type mine earthquakes in key layers during mining and inducing rock bursts, realizing the precise prevention and control of rock bursts and the support of disaster source management, reducing the cost of rock burst prevention and control, and improving the safe and efficient mining of coal mines.

[0021] Based on any of the above embodiments, Figure 2 As shown, the method for reducing noise of microseismic signals of hydraulic fracturing of coal seam roof includes: Step S201, decomposing the coal seam roof hydraulic fracturing microseismic signal into multiple band-limited intrinsic mode components through variational mode decomposition, and adaptively judging the validity of the multiple band-limited intrinsic mode components based on the IMF component judgment criterion.

[0022] In the embodiment of the present invention, Figure 3 As shown in the figure, the complex microseismic signal is adaptively decomposed into a set of band-limited intrinsic mode function (IMF) components with different center frequencies through variational mode decomposition (VMD), so as to effectively show the multi-scale structure of the microseismic signal and provide a stable and controllable technical basis for the subsequent denoising processing.

[0023] Through VMD decomposition, each IFM component obtained has a certain bandwidth. Therefore, the signal is analyzed from the frequency domain so that the signal characteristics can be analyzed from different frequency bands. When performing VMD decomposition, according to the formula: The microseismic signal of hydraulic fracturing in the coal seam roof is decomposed into multiple band-limited eigenmode components. for The time corresponding to band-limited eigenmode components, for The instantaneous phase corresponding to the moment, for The instantaneous amplitude corresponding to the moment; ; Is a positive integer.

[0024] According to the formula: Sure Moment Band-limited eigenmode components The instantaneous frequency , that is, the band-limited eigenmode component The center frequency of

[0025] Multiple IMF components are obtained by VMD decomposition. Each IMF component contains different proportions of effective components and noise components. In the embodiment of the present invention, the noise component (including the IMF component with a high noise ratio) is removed and the waveform correlation coefficient is used. and variance contribution Optimize the IMF components obtained by decomposition. Specifically, according to the formula: Adaptively judge the validity of the band-limited eigenmode components. is the preset similarity threshold, . for Moment The microseismic signals of hydraulic fracturing in the coal seam roof are collected by the signal collection channels; is the number of collected microseismic signals of hydraulic fracturing in the coal seam roof; ; is a positive integer; Microseismic signal of hydraulic fracturing in coal seam roof Microseismic signals of hydraulic fracturing in coal seam roof No. Band-limited eigenmode components The waveform correlation coefficient between them; Microseismic signal of hydraulic fracturing in coal seam roof No. Band-limited eigenmode components Microseismic signals of hydraulic fracturing in coal seam roof The variance contribution rate of Microseismic signal of hydraulic fracturing in coal seam roof The total number of band-limited eigenmode components obtained after decomposition; The number of effective band-limited eigenmode components to be determined for adaptive judgment; By waveform correlation coefficient , variance contribution rate and preset similarity threshold , for the microseismic signal of hydraulic fracturing in coal seam roof No. Band-limited eigenmode components The validity of the band-limited eigenmode component is judged. When the band-limited eigenmode component meets the judgment condition, the band-limited eigenmode component is a valid component; otherwise, it is an invalid component. The reconstruction of the microseismic signal of hydraulic fracturing of the coal seam roof can be achieved through the effective band-limited eigenmode component.

[0026] Step S202: Perform continuous synchronous compressed wavelet transform analysis on each band-limited eigenmode component that passes the validity judgment to obtain the SS-CWT time-frequency diagram of the corresponding band-limited eigenmode component.

[0027] Specifically, according to the formula: The band-limited eigenmode components that have passed the validity judgment are subjected to continuous synchronous compression wavelet transform. For Microseismic signals of hydraulic fracturing in coal seam roof collected at all times The translation factor and frequency Performing continuous wavelet transform on the synchronous compressed frequency domain; By using the wavelet basis function right Microseismic signals of hydraulic fracturing in coal seam roof collected at all times According to the telescopic scale and translation factor Perform continuous wavelet transform in the frequency domain; Respectively Microseismic signals of hydraulic fracturing in coal seam roof collected at all times Perform the scaling and translation factors of continuous wavelet transform; Microseismic signal of hydraulic fracturing in coal seam roof The Fourier transform of is the wavelet basis function The Fourier transform of To follow the telescopic scale and translation factor Microseismic signals of hydraulic fracturing in coal seam roof The instantaneous frequency obtained by continuous compression wavelet transform; For Microseismic signals of hydraulic fracturing in coal seam roof collected at all times In telescopic scale and translation factor Time-frequency representation after continuous wavelet transform; is the Dirac function; is an imaginary unit, representing the microseismic signal of hydraulic fracturing in the coal seam roof Complex number representation in wavelet transforms.

[0028] Here, For Microseismic signals of hydraulic fracturing in coal seam roof collected at all times The translation factor and frequency The synchronous compression frequency domain of continuous wavelet transform is performed on the coal seam roof hydraulic fracturing microseismic signal Through the Dirac function To achieve frequency compression, the results of wavelet transform are concentrated on the corresponding frequency, thereby converting the wavelet coefficients Medium and frequency The corresponding components are accumulated to obtain a clearer synchronous compression frequency domain representation. Extraction of band-limited eigenmode components and generation of time-frequency diagrams are performed.

[0029] First, extract the frequency components of synchronous compression, that is, by synchronously compressing the frequency Concentrate the signal energy to the corresponding frequency, and through the band-limited characteristics, synchronize the compression frequency Select the corresponding frequency interval in and extract the frequency components of the response. In each It represents the time-frequency energy distribution of the signal at this frequency. Through synchronous compression operation, the signal energy is concentrated near a specific instantaneous frequency, producing a sharper frequency. To extract the band-limited eigenmode component, it is necessary to extract the band-limited eigenmode component in a specific frequency range. The required frequency components are selected through a bandpass filter.

[0030] Then, reconstruct the band-limited eigenmode components, that is, the frequency range The corresponding frequency components are reversed through the inverse process of wavelet transform to obtain the corresponding band-limited eigenmode components. Specifically, according to the extracted frequency range The required frequency components within the The band-limited eigenmode components are reconstructed by the inverse process of wavelet transform, that is, the frequency range is extracted. In this process, the band-limited characteristics are used to reconstruct the modal signal in each frequency range, and the wavelet transform coefficients in the time-frequency domain are inversely transformed back to the time domain.

[0031] Next, generate the SS-CWT time-frequency diagram, that is, display the reconstructed mode in the time-frequency domain, and generate the SS-CWT time-frequency diagram of the band-limited eigenmode component. Specifically, by selecting the band-limited eigenmode decomposed in different time periods, generate the corresponding SS-CWT time-frequency diagram. The obtained modal components retain the main characteristics of the signal, while the noise is effectively suppressed. These time-frequency diagrams have high time-frequency resolution and can more clearly show the changes of the signal at different times and frequencies, thereby revealing the dynamic characteristics of the band-limited eigenmodes.

[0032] Step S203, analyzing the time-frequency characteristics of the SS-CWT time-frequency diagram of the band-limited intrinsic mode component to obtain the instantaneous frequency distribution and local energy distribution of the SS-CWT time-frequency diagram of the band-limited intrinsic mode component; and based on wavelet threshold denoising, classifying and denoising the SS-CWT time-frequency diagram of the corresponding band-limited intrinsic mode component through different denoising strategies.

[0033] Specifically, according to the determined wavelet basis function, the time-frequency characteristics of the SS-CWT time-frequency diagram of the band-limited intrinsic mode component are subjected to wavelet decomposition. The extraction of the time-frequency characteristics of the SS-CWT time-frequency diagram of the band-limited intrinsic mode component includes the extraction of the instantaneous frequency distribution and the local energy distribution. In the process of extracting the instantaneous frequency distribution, firstly, the signal is decomposed into time and frequency by CWT to obtain Microseismic signals of hydraulic fracturing in coal seam roof collected at all times In telescopic scale and translation factor Time-frequency representation after continuous wavelet transform , that is, the wavelet transform coefficient Then, through synchronous compression technology, the energy is concentrated to a specific instantaneous frequency. In order to extract the instantaneous frequency distribution of the signal, we can find the moment (translation factor) on the time-frequency plane. The main frequency where the energy is concentrated can be obtained as the signal at time (translation factor) The instantaneous frequency of: In the SS-CWT time-frequency diagram, the instantaneous frequency is displayed as a time on the video plane (translation factor) The maximum position of is obtained by calculating the translation factor at each moment By extracting the instantaneous frequency, we can get the curve of instantaneous frequency changing with time, that is, the instantaneous frequency distribution.

[0034] In the embodiment of the present invention, the local energy distribution refers to the energy density of the signal at a specific time and frequency, reflecting the energy distribution of the signal in the time-frequency domain. In the process of obtaining the local energy distribution, first, the local energy calculation is performed. Specifically, for each moment (translation factor) and frequency , the value in the SS-CWT time-frequency diagram reflects the energy distribution of the time-frequency point, and the energy It is also related to the square of the wavelet transform coefficient, that is: The above formula characterizes the signal at time (translation factor) and frequency The energy density on .

[0035] Then, the local energy distribution is extracted. Specifically, in the time-frequency diagram, directly from Extract the energy distribution from . For the moment (translation factor) , you can view different frequencies The energy distribution on the time-frequency diagram can be used to obtain the local frequency energy distribution. Therefore, the energy of the signal can be observed over time and frequency through the cumulative energy distribution on the time-frequency diagram, effectively revealing the main energy concentration frequency band of the signal.

[0036] Thus, the combination of instantaneous frequency distribution and local energy distribution effectively describes the characteristics of the signal. The instantaneous frequency distribution shows how the frequency of the signal changes over time. It is determined by tracking the frequency position of the maximum energy in the time-frequency diagram, which helps to identify the dynamic frequency characteristics of the signal, especially the frequency changes of non-stationary signals. The local energy distribution shows the energy concentration of the signal at a specific time and frequency; by analyzing the energy density in the time-frequency diagram, it helps to identify the main energy concentration frequency band of the signal and the change of energy over time. The combination of instantaneous frequency distribution and local energy distribution helps to better identify and analyze the main components of complex non-stationary signals.

[0037] After extracting the time-frequency features of the SS-CWT time-frequency diagram of the band-limited intrinsic mode component, the time-frequency features are decomposed by wavelet basis functions, and the SS-CWT time-frequency diagram is converted into wavelet transform coefficients in the frequency domain, that is, the SS-CWT time-frequency diagram of the band-limited intrinsic mode component is converted into wavelet transform coefficients in the frequency domain. Among them, the time-frequency features are decomposed by wavelet basis functions to obtain the wavelet transform coefficients in the time-frequency domain. ; By wavelet transform coefficients Processing (integration, averaging, compression, etc.) is performed to obtain the frequency domain wavelet change coefficient of the signal.

[0038] Specifically, first, the signal is decomposed into time and frequency by wavelet basis function to obtain wavelet transform coefficients ; Then, find the energy value corresponding to the main frequency at different times from the SS-CWT time-frequency diagram, and remove the time dependence by accumulation or integration; then, Accumulate or weighted average to obtain the wavelet transform coefficients in pure frequency domain, that is, the time-frequency coefficients (wavelet transform coefficients) The time dependence of the signal is removed, the frequency characteristics of the signal are retained, and converted into frequency domain coefficients; finally, through time-frequency operations such as synchronous compression, the resolution and accuracy of the frequency domain characteristics are improved to make the frequency information clearer.

[0039] After converting the SS-CWT time-frequency diagram of the band-limited intrinsic mode component into the frequency-domain wavelet transform coefficients, the frequency-domain wavelet transform coefficients corresponding to the band-limited intrinsic mode component are thresholded based on the determined threshold function, and according to the frequency bandwidth of the band-limited intrinsic mode component, the soft threshold or hard threshold is selected to classify and denoise the SS-CWT time-frequency diagram of the band-limited intrinsic mode component.

[0040] Specifically, based on the threshold function: The frequency domain wavelet transform coefficients corresponding to the band-limited eigenmode components are thresholded. is the frequency domain wavelet transform coefficient obtained by transforming the SS-CWT time-frequency diagram of the band-limited eigenmode component; is a preset threshold limit for frequency domain wavelet transform coefficients; , is the cutoff value of the wavelet transform coefficient in the frequency domain; is a piecewise function (symbolic function); is the thresholded value of the frequency domain wavelet transform coefficient corresponding to the band-limited eigenmode component; The scaling scale of continuous wavelet transform for hydraulic fracturing microseismic signals in coal seam roof is ; Represents the method (soft thresholding or hard thresholding) or process used for thresholding.

[0041] Here, the cutoff value is used to limit and reduce the small value of the frequency domain wavelet transform coefficient, which is located at the boundary between the noise and signal characteristics. The smaller noise part is removed by the cutoff value, and the meaningful signal part is retained or reduced. Less than the preset threshold limit , the processing of wavelet coefficients can be further refined to retain the main characteristics of the signal while effectively removing noise; through thresholding and truncation of wavelet coefficients, the time-frequency diagram of the signal can be clearer, and the signal and noise can be effectively separated.

[0042] In the embodiment of the present invention, after thresholding the frequency domain wavelet transform coefficients, the thresholded value is obtained. , and the frequency bandwidth of the band-limited eigenmode components is combined for classification denoising. The frequency domain wavelet transform coefficients , and perform corresponding attenuation; for It is determined to be noise and is directly reset to zero. (Right now ), then smoothing is performed using a nonlinear function.

[0043] In the SS-CWT time-frequency diagram, different frequency bandwidths usually contain different signal characteristics. The frequency bandwidth of the band-limited eigenmode component determines the frequency components containing the main signal components and noise components. Among them, the narrowband frequency bandwidth corresponds to the local frequency components of the signal and may contain useful signals; the broadband frequency bandwidth may contain more noise components. Different threshold processing methods can be selected based on the frequency bandwidth characteristics of the band-limited eigenmode component. For narrowband frequencies, the difference between the signal and the noise is not significant, and a soft threshold (preset threshold limit) is used. ) method is used to smooth the wavelet coefficients, retain the main information of the wavelet coefficients, and gradually reduce the impact of noise on the signal to retain more signal features; for broadband frequencies, the difference between signal and noise is significant, and a hard threshold (truncation value) is used ) method, will be less than the cutoff value The signal part corresponding to the wavelet coefficient is directly set to zero to remove the noise.

[0044] Step S204: According to the instantaneous frequency distribution and local energy distribution of the SS-CWT time-frequency diagram of the band-limited intrinsic mode component, based on the adaptive time-frequency segmentation algorithm, feature extraction is performed on different time-frequency regions of the SS-CWT time-frequency diagram after the classification and denoising of the band-limited intrinsic mode component, so as to perform time-frequency region segmentation on the SS-CWT time-frequency diagram after the classification and denoising.

[0045] When rock burst occurs, the time span of the mining earthquake event is short and the energy is relatively low (compared with local or regional earthquakes). Therefore, there are other high-power features (noise or long-period signals) in the SS-CWT time-frequency diagram, and the high-power features will affect the performance of time-frequency denoising and reduce the denoising efficiency. Therefore, in the embodiment of the present invention, different denoising strategies are used to distinguish and eliminate the influence of high-power features in microseismic signals to improve sparsity and denoising efficiency.

[0046] In the continuous wavelet representation of passive microseismic records, high-power features are distributed throughout the entire data length in some limited frequency bands, among which some high-power features are related to the main noise in a relatively narrow frequency band within the frequency range of the microseismic event, while other features appear in relatively low frequency bands outside the main frequency range of the microseismic event. In addition, the high-power features and time-localized features corresponding to the seismic event do not appear in the entire duration of the time series, but appear from time to time in the form of short pulses. Based on this, in the embodiments of the present invention, Figure 4 As shown in the figure, by analyzing the time-frequency characteristics of the SS-CWT time-frequency of the band-limited intrinsic mode component, the instantaneous frequency distribution and local energy distribution of the SS-CWT time-frequency diagram of the band-limited intrinsic mode component are obtained, and then the high-power low-frequency characteristics are separated from the frequency range of the microseismic signal of the top plate hydraulic fracturing.

[0047] Specifically, according to the instantaneous frequency distribution and local energy distribution of the SS-CWT time-frequency diagram of the band-limited eigenmode component, according to the formula: The superposition amplitude of the frequency domain wavelet transform coefficients corresponding to the band-limited eigenmode components is calculated to represent the distribution of the frequency domain wavelet transform coefficients along the scale axis. is the frequency domain wavelet transform coefficient obtained by transforming the SS-CWT time-frequency diagram of the band-limited eigenmode component; is the total number of scaling scales when performing continuous synchronous compression wavelet transform on the band-limited eigenmode components; In accordance with Scale Superposition amplitudes of successive synchronously compressed wavelet transforms of band-limited eigenmode components.

[0048] Here, the characteristics of microseismic signals in different frequency ranges are represented by the scale axis (the scaling scale of microseismic signals during wavelet transform). The distribution of the magnitude of the frequency domain wavelet transform coefficients along the scale axis helps to understand the vibration characteristics of different frequencies generated during hydraulic fracturing. The scaling scale is inversely proportional to the frequency. A larger scaling scale corresponds to a lower frequency, while a smaller scaling scale corresponds to a higher frequency. By analyzing different scaling scales, the characteristics of microseismic signals in various frequency ranges can be captured. By calculating the superposition amplitude of the frequency domain wavelet transform coefficients corresponding to the band-limited eigenmode components , the energy distribution at different extension scales can be observed to facilitate the identification of the dominant characteristics and potential anomalies of microseismic signals.

[0049] The existence of low-frequency characteristics in the microseismic signal of roof hydraulic fracturing leads to the superposition amplitude of the frequency domain wavelet transform coefficients. There are multiple different peaks on the scale axis, that is, there are overlaps of different frequency components in the microseismic signal, resulting in signal aliasing, making it difficult to distinguish specific vibration features. To this end, in the embodiment of the present invention, the superposition amplitude is constructed The probability histogram is obtained, and the noise is segmented based on the adaptive time-frequency segmentation algorithm to improve the accuracy of microseismic signal analysis. Specifically, according to the formula: Determine the minimum weighted within-class variance , in order to perform time-frequency region segmentation on the SS-CWT time-frequency diagram after classification and denoising, and obtain the shock wave characteristic area and noise area in the SS-CWT time-frequency diagram with limited eigenmode components.

[0050] In the formula, For telescopic scale The weighted intra-class variance under ; The total number of expansion and contraction scales when performing continuous synchronous compression wavelet transform on microseismic signals; The telescoping scale corresponding to minimizing the weighted intra-class variance; , Is a positive integer.

[0051] In the embodiment of the present invention, by minimizing the weighted intra-class variance Determine the best segmentation points for different categories (seismic wave features and noise) to perform time-frequency region segmentation on the SS-CWT time-frequency graph after classification and denoising. Specifically, by minimizing the weighted intra-class variance , the different areas on the SS-CWT time-frequency diagram after classification and denoising are divided into seismic wave characteristic areas and noise areas. Among them, the seismic wave characteristic area has stronger signal energy, which is manifested as a higher amplitude accumulation value ( ), and shows consistency in intra-class variance; while the noise area has weaker signal energy, and the energy distribution is scattered and unconcentrated. Then, according to the energy density of different areas, the segmentation threshold is dynamically adjusted to achieve accurate segmentation of high-energy signal areas and low-energy noise areas in the time-frequency diagram.

[0052] Step S205: According to the time-frequency region segmentation result of the SS-CWT time-frequency diagram after the classification and denoising processing, the SS-CWT time-frequency diagram after the classification and denoising processing of the band-limited intrinsic mode component is analyzed by the AIC arrival time picking method to pick up the first arrival time of the microseismic waveform of the roof hydraulic fracturing.

[0053] Here, the first arrival time refers to the arrival time, the initial arrival time, that is, the position where the microseismic waveform first starts (the starting time of the waveform when the seismic wave first arrives at the sensor). Specifically, through the AIC arrival time picking method, according to the SS-CWT time-frequency diagram of the band-limited intrinsic mode component after removing the noise area, the mutation point of the microseismic signal of the hydraulic fracturing of the coal seam roof is identified, and the first arrival time of the microseismic waveform of the hydraulic fracturing of the roof is picked up. In other words, based on the adaptive time-frequency segmentation, the time-frequency area segmentation of the seismic wave characteristic area and the noise area in the SS-CWT time-frequency diagram of the band-limited intrinsic mode component is realized; then, the seismic wave characteristic area is restored to a "time-amplitude" time series, and the AIC arrival time picking method is used to determine the peak value of the "time-amplitude" time series. The time point corresponding to the peak value is the starting time of the microseismic waveform, such as Figure 5 shown.

[0054] At the same time, after the time-frequency region segmentation of the seismic wave characteristic region and the noise region in the SS-CWT time-frequency diagram of the band-limited intrinsic mode component is realized, the reconstructed microseismic signal of the hydraulic fracturing of the coal seam roof is obtained by adding the seismic wave characteristic region in the SS-CWT time-frequency diagram of the band-limited intrinsic mode component, such as Figure 6 Then, the microseismic signal of hydraulic fracturing in the coal seam roof is evaluated by using the pre-determined denoising effect evaluation index. The denoising effect is evaluated.

[0055] Specifically, according to the formula: Microseismic signals of hydraulic fracturing in coal seam roof The denoising effect is evaluated. Microseismic signal of hydraulic fracturing in coal seam roof Reconstructing microseismic signals signal-to-noise ratio; Microseismic signal of hydraulic fracturing in coal seam roof The energy percentage of the reconstructed microseismic signal; Microseismic signal of hydraulic fracturing in coal seam roof The root mean square error with the reconstructed microseismic signal; Microseismic signal of hydraulic fracturing in coal seam roof and the signal smoothness of the reconstructed microseismic signal; for Microseismic signals of hydraulic fracturing in the coal seam roof collected at all times; for Microseismic signals of hydraulic fracturing in coal seam roof collected at all times Reconstructed microseismic signals; for Microseismic signals of hydraulic fracturing in the coal seam roof collected at all times; for Reconstructed microseismic signals of hydraulic fracturing microseismic signals of coal seam roof collected at all times; , is the number of collected microseismic signals of hydraulic fracturing in the coal seam roof.

[0056] The method for reducing noise of microseismic signals of hydraulic fracturing in the coal seam roof provided by the embodiment of the present invention decomposes the microseismic signals of hydraulic fracturing through VMD, and proposes a criterion for judging the band-limited intrinsic mode components, reveals the multi-scale structure of the microseismic signals, and provides a more stable and controllable basis for signal denoising processing; uses continuous synchronous compression wavelet transform for joint noise reduction processing, and performs continuous synchronous compression wavelet transform and single-channel customized threshold adaptive filtering on each band-limited intrinsic mode component; and uses an adaptive time-frequency segmentation algorithm to process high frequency and low frequency, and realizes the high-focus and high-resolution characteristics of the microseismic signals of hydraulic fracturing in the coal seam roof through the "high-frequency signal, low-frequency noise" analysis mode of the time-frequency characteristics of the microseismic signals of hydraulic fracturing. In this way, the potential threat of sudden occurrence of fault-type mine earthquakes in key layers and inducing rock bursts during mining is avoided, which is an effective measure for the precise prevention and control of rock bursts, provides a theoretical basis for the accurate prevention and control of mines with frequent mine earthquakes, realizes the source control of rock burst disasters, reduces the cost of rock burst prevention and control, and improves the safe and efficient mining of coal mines.

[0057] The following is a description of the coal seam roof hydraulic fracturing microseismic signal denoising system provided by the present invention. The coal seam roof hydraulic fracturing microseismic signal denoising system described below and the coal seam roof hydraulic fracturing microseismic signal denoising method described above can be referenced to each other.

[0058] Figure 7 A schematic diagram of the structure of a coal seam roof hydraulic fracturing microseismic signal noise reduction system provided by an embodiment of the present invention is shown in FIG. Figure 7 As shown, the coal seam roof hydraulic fracturing microseismic signal noise reduction system provided by the embodiment of the present invention includes: The decomposition module 701 is used to decompose the coal seam roof hydraulic fracturing microseismic signal into multiple band-limited intrinsic mode components through variational mode decomposition, and perform continuous synchronous compression wavelet transform analysis on each band-limited intrinsic mode component that passes the validity judgment to obtain a synchronous compression wavelet transform time-frequency diagram of the corresponding band-limited intrinsic mode component; Denoising module 702, used for performing classification denoising on the synchronous compression wavelet transform time-frequency diagram based on wavelet threshold denoising according to different denoising strategies to obtain denoised modal components; The segmentation module 703 is used to segment different time-frequency regions of the synchronous compression wavelet transform time-frequency graph according to the instantaneous frequency distribution and local energy distribution of the synchronous compression wavelet transform time-frequency graph, and pick up the first arrival time of the microseismic waveform of the roof hydraulic fracturing according to the time-frequency region segmentation result; The reconstruction module 704 is used to screen the denoised modal components using the first arrival time of the microseismic waveform of the roof hydraulic fracturing to obtain effective modal components, reconstruct the effective modal components, and obtain the reconstructed microseismic signal of the coal seam roof hydraulic fracturing.

[0059] The coal seam roof hydraulic fracturing microseismic signal denoising system provided by the present invention decomposes the coal seam roof hydraulic fracturing microseismic signal into multiple band-limited intrinsic mode components through variational mode decomposition, performs continuous synchronous compression wavelet transform analysis on each band-limited intrinsic mode component that passes the effectiveness judgment, and obtains a synchronous compression wavelet transform time-frequency diagram of the corresponding band-limited intrinsic mode component; based on wavelet threshold denoising, the synchronous compression wavelet transform time-frequency diagram is classified and denoised according to different denoising strategies to obtain denoised mode components; according to the instantaneous frequency distribution and local energy distribution of the synchronous compression wavelet transform time-frequency diagram, different time-frequency regions of the synchronous compression wavelet transform time-frequency diagram are segmented, and according to the time-frequency regions, the synchronous compression wavelet transform time-frequency diagram is segmented and the local energy distribution is divided into different time-frequency regions. The domain segmentation result picks up the first arrival time of the microseismic waveform of the roof hydraulic fracturing; the denoised modal component is screened by using the first arrival time of the microseismic waveform of the roof hydraulic fracturing to obtain the effective modal component, and the effective modal component is reconstructed to obtain the reconstructed microseismic signal of the coal seam roof hydraulic fracturing, so as to realize signal-to-noise separation and high-performance noise reduction of the hydraulic fracturing microseismic signal with low signal-to-noise ratio and non-steady-state characteristics, provide a basis for the precise prevention and control of mines with frequent mine earthquakes, effectively avoid the potential threat of sudden mine earthquakes caused by fault-type mine earthquakes in key layers during mining and then inducing rock bursts, realize the precise prevention and control of rock bursts and support the source control of disasters, reduce the cost of rock burst prevention and control, and improve the safe and efficient mining of coal mines.

[0060] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the relevant technology can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiment.

[0061] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for reducing noise of microseismic signals of hydraulic fracturing of coal seam roof, characterized in that: include: The microseismic signal of hydraulic fracturing in the coal seam roof is decomposed into multiple band-limited intrinsic mode components by variational mode decomposition. A continuous synchronous compression wavelet transform analysis is performed on each band-limited intrinsic mode component that passes the validity judgment, and the synchronous compression wavelet transform time-frequency diagram of the corresponding band-limited intrinsic mode component is obtained. Based on wavelet threshold denoising, the synchronous compression wavelet transform time-frequency diagram is subjected to classification denoising according to different denoising strategies to obtain denoised modal components; According to the instantaneous frequency distribution and local energy distribution of the synchronous compression wavelet transform time-frequency diagram, different time-frequency regions of the synchronous compression wavelet transform time-frequency diagram are segmented, and the first arrival time of the microseismic waveform of the roof hydraulic fracturing is picked up according to the time-frequency region segmentation result; The denoised modal components are screened using the first arrival time of the microseismic waveform of the roof hydraulic fracturing to obtain effective modal components, and the effective modal components are reconstructed to obtain a reconstructed microseismic signal of the coal seam roof hydraulic fracturing.

2. The method for reducing noise of microseismic signals of hydraulic fracturing of coal seam roof according to claim 1, characterized in that: The method of decomposing the coal seam roof hydraulic fracturing microseismic signal into multiple band-limited eigenmode components by variational mode decomposition includes: When performing variational mode decomposition, the coal seam roof hydraulic fracturing microseismic signal is decomposed into multiple band-limited eigenmode components according to the following formula; In the formula, for The time corresponding to band-limited eigenmode components, for The instantaneous phase corresponding to the moment, for The instantaneous amplitude corresponding to the moment; ; is a positive integer; Sure Moment Band-limited eigenmode components The instantaneous frequency : According to the instantaneous frequency A plurality of band-limited eigenmode components are obtained, and each band-limited eigenmode component contains different proportions of effective components and noise components.

3. The method for reducing noise of microseismic signals of hydraulic fracturing of coal seam roof according to claim 1, characterized in that: The effectiveness judgment includes: The microseismic signal of hydraulic fracturing of the coal seam roof is obtained according to the following formula Microseismic signals of hydraulic fracturing in coal seam roof No. Band-limited eigenmode components The waveform correlation coefficient between and variance contribution : In the formula, is the preset similarity threshold, for Moment The microseismic signals of hydraulic fracturing in the coal seam roof are collected by the signal collection channels; is the number of collected microseismic signals of hydraulic fracturing in the coal seam roof; ; is a positive integer; Microseismic signal of hydraulic fracturing in coal seam roof Microseismic signals of hydraulic fracturing in coal seam roof No. Band-limited eigenmode components The waveform correlation coefficient between them; Microseismic signal of hydraulic fracturing in coal seam roof No. Band-limited eigenmode components Microseismic signals of hydraulic fracturing in coal seam roof The variance contribution rate of Microseismic signal of hydraulic fracturing in coal seam roof The total number of band-limited eigenmode components obtained after decomposition; The number of effective band-limited eigenmode components to be determined for adaptive judgment; By waveform correlation coefficient , variance contribution rate and preset similarity threshold , for the microseismic signal of hydraulic fracturing in coal seam roof No. Band-limited eigenmode components The validity of the band-limited eigenmode component is judged. When the band-limited eigenmode component meets the judgment condition, the corresponding band-limited eigenmode component is judged to be a valid component; otherwise, the corresponding band-limited eigenmode component is judged to be an invalid component.

4. The method for reducing noise of microseismic signals of hydraulic fracturing of coal seam roof according to claim 1 or 2, characterized in that: The method of performing continuous synchronous compression wavelet transform analysis on each band-limited eigenmode component that passes the validity judgment to obtain a synchronous compression wavelet transform time-frequency diagram of the corresponding band-limited eigenmode component includes: Compression frequency by sync Concentrate the signal energy to the corresponding frequency, and through the band-limited characteristics, synchronize the compression frequency Select the corresponding frequency interval and extract the frequency component of the response; The corresponding frequency components in the frequency range are reversed through the inverse process of wavelet transform to obtain the corresponding band-limited eigenmode components. Using the band-limited characteristics, the modal signals in each frequency range are screened twice, and the wavelet transform coefficients in the time-frequency domain are inversely transformed back to the time domain. The reconstructed modes are displayed in the time-frequency domain, and the synchronous compressed wavelet transform time-frequency diagram of the band-limited eigenmode components is generated.

5. The method for reducing noise of microseismic signals of hydraulic fracturing of coal seam roof according to claim 1, characterized in that: The wavelet threshold based denoising is to classify and denoise the synchronous compression wavelet transform time-frequency graph according to different denoising strategies, including: After thresholding the frequency domain wavelet transform coefficients, the thresholded value is obtained. , and classify and denoise the frequency bandwidth of the band-limited eigenmode components: for The frequency domain wavelet transform coefficients , and perform corresponding attenuation; for It is determined to be noise and directly reset to zero; for , then smoothing is performed through a nonlinear function.

6. The method for reducing noise of microseismic signals of hydraulic fracturing of coal seam roof according to claim 5, characterized in that: The wavelet threshold denoising is based on which the synchronous compression wavelet transform time-frequency graph is subjected to classification denoising according to different denoising strategies, and further includes: Different threshold processing methods are selected through the frequency bandwidth characteristics of the band-limited eigenmode components. For narrowband frequencies, the soft threshold method is used to smooth the wavelet coefficients, retain the main information of the wavelet coefficients, and gradually reduce the impact of noise on the signal to retain more signal features. For broadband frequencies, the difference between signal and noise is significant, and the hard threshold method is used to reduce the signal below the cutoff value. The signal part corresponding to the wavelet coefficient is directly set to zero to remove the noise.

7. The method for reducing noise of microseismic signals of hydraulic fracturing of coal seam roof according to claim 1, characterized in that: The segmenting of different time-frequency regions of the synchronous compression wavelet transform time-frequency graph according to the instantaneous frequency distribution and the local energy distribution of the synchronous compression wavelet transform time-frequency graph comprises: Determine the minimum weighted within-class variance , to perform time-frequency region segmentation on the synchronous compression wavelet transform time-frequency image after classification denoising, and obtain the shock wave characteristic area and noise area in the SS-CWT time-frequency image of the band-limited eigenmode component; By minimizing the weighted intra-class variance Determine the best segmentation point of seismic wave characteristics and noise to perform time-frequency region segmentation on the synchronous compression wavelet transform time-frequency graph after classification denoising, where the seismic wave characteristic region is the amplitude accumulation value ( ) is greater than the preset segmentation threshold and shows consistency in the intra-class variance; while the energy distribution in the noise area is scattered and unconcentrated; the segmentation threshold is dynamically adjusted according to the energy density of different areas.

8. The method for reducing noise of microseismic signals of hydraulic fracturing of coal seam roof according to claim 7, characterized in that: The method of picking up the first arrival time of the microseismic waveform of the roof hydraulic fracturing according to the time-frequency region segmentation result includes: Restoring the seismic wave characteristic region into a time-amplitude time series; The AIC arrival time picking method is used to determine the peak value of the time-amplitude time series. The time point corresponding to the peak value is the first arrival time of the microseismic waveform.

9. The method for reducing noise of microseismic signals of hydraulic fracturing of coal seam roof according to claim 1, characterized in that: The reconstructing the effective modal components to obtain the reconstructed microseismic signal of the coal seam roof hydraulic fracturing includes: Through the first arrival time of the microseismic waveform, the effective modal component is screened as the characteristic area of ​​the seismic wave containing strong signal energy; The seismic wave characteristic regions containing relatively strong signal energy are added together to obtain a reconstructed microseismic signal of hydraulic fracturing of the coal seam roof.

10. A coal seam roof hydraulic fracturing microseismic signal noise reduction system, characterized in that: include: A decomposition module is used to decompose the coal seam roof hydraulic fracturing microseismic signal into multiple band-limited intrinsic mode components through variational mode decomposition, and perform continuous synchronous compression wavelet transform analysis on each band-limited intrinsic mode component that passes the validity judgment to obtain the synchronous compression wavelet transform time-frequency diagram of the corresponding band-limited intrinsic mode component; A denoising module, which is used for performing classification denoising on the synchronous compression wavelet transform time-frequency diagram based on wavelet threshold denoising according to different denoising strategies to obtain denoised modal components; A segmentation module is used to segment different time-frequency regions of the synchronous compression wavelet transform time-frequency graph according to the instantaneous frequency distribution and local energy distribution of the synchronous compression wavelet transform time-frequency graph, and pick up the first arrival time of the microseismic waveform of the roof hydraulic fracturing according to the time-frequency region segmentation result; The reconstruction module is used to screen the denoised modal components using the first arrival time of the microseismic waveform of the roof hydraulic fracturing to obtain effective modal components, reconstruct the effective modal components, and obtain the reconstructed microseismic signal of the coal seam roof hydraulic fracturing.

Images