R-wave pickup method for actually measuring blasting vibration signals based on engineering scale

Abstract

The invention discloses an R wave pickup method based on an engineering scale actual measurement blasting vibration signal, and relates to the field of rock blasting negative effect safety evaluation, and the method comprises the steps: arranging a blasting vibration meter, and monitoring a blasting induced vibration signal; analyzing blasting actually-measured vibration data, and obtaining a blasting signal time-frequency structure; carrying out filtering processing on an R-wave region in the time-frequency structure based on wavelet transform to obtain R-wave signal time-frequency data; and analyzing and processing the R-wave time-frequency data based on S inverse transformation to obtain an R-wave signal. According to the R-wave pickup method for actually measuring the blasting vibration signal based on the engineering scale, high-precision extraction of the R-wave signal is realized by improving the S transformation and wavelet filtering technology, and meanwhile, damage to the body wave signal is avoided. The method has the beneficial effects of high-precision extraction, engineering applicability, simplicity and convenience in operation, improvement of analysis accuracy, capability of providing support for engineering safety evaluation and the like, and can provide reliable technical support for blasting vibration signal analysis and engineering safety evaluation.

Description

Technical Field

[0001] The present invention relates to the field of safety evaluation of negative effects of rock blasting, and particularly to an R-wave picking method based on measured blasting vibration signals at the engineering scale. Background Art

[0002] During the process of rock excavation, drilling blasting is widely used due to its advantages such as high construction efficiency, convenience, and low cost. However, the blasting seismic waves generated by the conversion of the energy released during the explosion of explosives and propagating outward will have a certain impact on surrounding buildings, facilities, etc. Existing research shows that blasting seismic waves can be divided into body waves and surface waves. Among them, the surface wave (R-wave) carries most of the vibration energy, with a proportion of up to 67%. At the same time, due to the characteristics of low frequency and long duration of the surface wave, the degree of impact and damage on surrounding buildings is higher than that of body waves. Therefore, in the field of safety evaluation of negative effects of blasting, R-wave picking has always been one of the hot issues studied by researchers.

[0003] Most of the existing achievements in R-wave picking directly come from seismological research. Compared with natural earthquakes, blasting seismic waves have disadvantages such as low signal-to-noise ratio and short propagation distance, resulting in unclear separation. The existing achievements have large errors when directly applied to R-wave picking of vibration signals at the engineering scale. Therefore, based on this current situation, a research is carried out to propose an R-wave picking method based on measured blasting vibration signals at the engineering scale. Summary of the Invention

[0004] The purpose of the present invention is to provide an R-wave picking method based on measured blasting vibration signals at the engineering scale, aiming at the problems of large errors and poor effects in R-wave picking of measured vibration signals of blasting at the engineering scale of meters to kilometers, by comprehensively utilizing the advantages of improved S-transform to obtain the time-frequency space differences between body waves and surface waves and wavelet transform R-wave filtering, so as to overcome the defects of the existing technology.

[0005] The technical solution adopted by the present invention is as follows:

[0006] An R-wave picking method based on measured blasting vibration signals at the engineering scale, comprising the following steps:

[0007] Step 1, arranging blasting vibration monitors to monitor the vibration signals induced by blasting;

[0008] Step 2, based on the improved S-transform, analyzing the measured blasting vibration data to obtain the time-frequency structure of the blasting signal;

[0009] Step 3, based on wavelet transform, filtering the R-wave region in the time-frequency structure to obtain the time-frequency data of the R-wave signal;

[0010] Step 4, based on the inverse S-transform, analyzing and processing the time-frequency data of the R-wave to obtain the R-wave signal.

[0011] Further, in step 1, a blasting vibration monitor is used to monitor the vibration signals induced by blasting; an RTK surveying instrument is used to measure the center of the blasting source and the vibration monitoring points to obtain coordinate information; and the distances between each measuring point and the center of the blasting source are calculated based on the coordinate information.

[0012] Further, in step 2, the improved S transform includes the following steps:

[0013] Step 2.1 EMD processing: The measured signal s(i) is subjected to empirical mode decomposition (EMD) to remove the IMF components of high-frequency noise; the remaining components are reconstructed to obtain the reconstructed signal s′(i);

[0014] Step 2.2 S transform: The reconstructed signal s′(i) is subjected to the S transform to obtain its time-frequency structure;

[0015] The S transform formula is:

[0016]

[0017] In the formula, ST(τ, f) represents the S transform result of the signal x(t), which is a time-frequency domain representation; represents the input time-domain signal; τ is the time translation parameter, representing the center position of the window function; f is the frequency variable, which is used to determine the frequency points of interest in time-frequency analysis and is also used to adjust the width of the window function. The higher the frequency, the narrower the window; x(t) represents the input time-domain signal; t is the time variable; |f| is the absolute value of the frequency, which is used to adjust the width of the window function. The higher the frequency; is the normalization coefficient to ensure that the integral area of the window function is 1; is the Gaussian function, which is used to localize the signal on the time axis; e -i2πft is the complex exponential function, which is used to transform the signal from the time domain to the frequency domain; i is the imaginary unit, satisfying i 2 = -1; dt represents the integral with respect to the time variable t;

[0018] Step 2.3 Determine the time-frequency space of the R wave: After determining the arrival times of the P wave and the S wave, the time-frequency space of the R wave is preliminarily determined.

[0019] Further, in step 3, the R wave time-frequency domain filtering based on wavelet transform includes the following steps:

[0020] Step 3.1, R wave filter design:

[0021]

[0022] In the formula, Fil represents the output value of the filter; t is the time variable; f is the frequency variable; Z R is the time-frequency space of the R wave;

[0023] Step 3.2, filtering process:

[0024] Filter the signal in the time-frequency space of the R wave using a filter to obtain the time-frequency data of the R wave signal:

[0025] ST R = ST(s′(t))·Fil(t,f)

[0026] In the formula, ST R represents the time-frequency data of the filtered R wave signal; ST(s′(t)) represents the S-transform result of the reconstructed signal s′(t); ST is the operator of the S-transform; s′(t) is the signal after preprocessing, that is, EMD denoising and reconstruction.

[0027] Furthermore, in step 4, the inverse S-transform processing includes the following steps:

[0028] Step 4.1, construct an inverse S-transform model:

[0029] Construct an inverse S-transform model, and the expression is as follows:

[0030]

[0031] Among them, the window function g(τ - t, f) satisfies:

[0032]

[0033] In the formula, x(t) represents the reconstructed time-domain signal; ST(τ, f) represents the S-transform result of the signal; τ is the time variable; f is the frequency variable; dτ represents the integral with respect to the time variable τ; df represents the integral with respect to the frequency variable f; e i2πft is the complex exponential function; g(τ - t, f) is the window function in the S-transform;

[0034] Step 4.2, inverse S-transform processing:

[0035] Perform inverse S-transform processing on the time-frequency data of the R wave signal to obtain the R wave signal:

[0036] R = ST -1 (ST R ) = ST -1 [ST(s′(t))·Fil(t,f)]

[0037] In the formula, R represents the reconstructed R wave time-domain signal; ST -1 represents the inverse S-transform operator; ST R represents the time-frequency data of the filtered R wave signal; ST(s′(t)) represents the S-transform result of the reconstructed signal s′(t); s′(t) represents the reconstructed signal after preprocessing; Fil(t,f) represents the time-frequency domain filter for extracting the R wave signal.

[0038] The beneficial effects of the present invention are as follows:

[0039] 1. High-precision R-wave extraction

[0040] Improved S-transform: Through the improved S-transform method, the time-frequency structure of blasting vibration signals can be obtained more accurately, and the time-frequency characteristics of body waves and R-waves can be effectively distinguished. Wavelet filtering technology: Combining wavelet filtering means, filtering is performed on the time-frequency region of the R-wave to avoid the interference of body wave signals and achieve high-precision extraction of the R-wave.

[0041] 2. No damage to body wave information

[0042] Selective filtering: Selective filtering is performed on the R-wave region in the time-frequency domain to ensure the integrity of body wave signals and avoid damage to body wave information. Independent processing: There are obvious differences between R-waves and body waves in the time-frequency space. Through time-frequency analysis, the R-wave signal can be processed independently without affecting the quality of body wave signals.

[0043] 3. Applicable to engineering scales

[0044] Engineering applicability: This method is applicable to the processing of blasting vibration signals at engineering scales (meters to kilometers), solving the problems of large R-wave picking errors and poor effects of existing technologies at engineering scales. Support from measured data: Based on measured blasting vibration signals, the method has high engineering practicability and reliability.

[0045] 4. Simple operation and easy to implement

[0046] Clear process: The method steps are clear, including signal acquisition, time-frequency analysis, filtering processing, and signal reconstruction. The operation is simple and easy to promote and apply in actual projects. Automation potential: Each step can be automatically processed through algorithms, reducing human intervention and improving processing efficiency.

[0047] 5. Improve the accuracy of blasting vibration signal analysis

[0048] Advantages of time-frequency analysis: Through the improved S-transform and wavelet filtering technology, the time-frequency characteristics of blasting vibration signals can be analyzed more accurately, providing reliable data support for the study of blasting vibration effects. Precise extraction of R-waves: R-waves are important components in blasting vibration signals, and their precise extraction helps to more comprehensively evaluate the impact of blasting vibration.

[0049] 6. Provide support for engineering safety assessment

[0050] Blasting vibration monitoring: Accurately extracting the R-wave signal helps to more accurately evaluate the impact of blasting vibration on the surrounding environment and structures, providing a scientific basis for engineering safety. Vibration control optimization: Based on the analysis results of the R-wave signal, the design of blasting parameters can be optimized to reduce the negative impact of vibration on the surrounding environment.

[0051] 7. Technical innovation

[0052] Improved S-transform: By improving the S-transform method, the accuracy and resolution of time-frequency analysis are enhanced, providing a new technical means for blasting vibration signal processing. Wavelet filtering combination: Combining wavelet filtering technology with time-frequency analysis realizes the accurate extraction of the R-wave signal, with high technical innovation.

[0053] In summary, the method for picking up the R-wave based on the measured blasting vibration signal at the engineering scale realizes the high-precision extraction of the R-wave signal by improving the S-transform and wavelet filtering technologies, while avoiding damage to the body wave signal. Its beneficial effects include high-precision extraction, engineering applicability, simple operation, improved analysis accuracy, and support for engineering safety assessment. This method has significant technical advantages and application value, and can provide reliable technical support for blasting vibration signal analysis and engineering safety assessment. Description of the drawings

[0054] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, for those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0055] Figure 1 It is the flowchart of the method for picking up the R-wave based on the measured blasting vibration signal at the engineering scale of the present invention;

[0056] Figure 2 It is the schematic diagram of the measuring point layout for the single-hole blasting test in step 1 of the present invention;

[0057] Figure 3 It is the time-frequency structure diagram of the measured blasting vibration signal in step 2 of the present invention;

[0058] Figure 4 It is the R-wave filtering region diagram of the measured blasting vibration signal in step 3 of the invention;

[0059] Figure 5 It is the R-wave picking result diagram of the measured blasting vibration signal in step 4 of the present invention. Detailed implementation manners

[0060] Next, in combination with the accompanying drawings of the embodiments of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0061] Aiming at the problem of large R-wave picking error in vibration signals at the engineering scale, this embodiment provides an R-wave picking method based on measured blasting vibration signals at the engineering scale. As Figure 1 shown, the R-wave picking method based on measured blasting vibration signals at the engineering scale includes the following steps:

[0062] Step 1: Arrange blasting vibration measuring instruments to monitor the vibration signals induced by blasting.

[0063] Equipment arrangement: Use blasting vibration measuring instruments to monitor the vibration signals induced by blasting.

[0064] Coordinate measurement: Use RTK surveying instruments to measure the center of the blast source and the vibration monitoring points to obtain coordinate information.

[0065] Calculation of the distance from the blast center: Calculate the distance between each measuring point and the center of the blast source according to the coordinate information, that is, the distance from the blast center.

[0066] Step 2: Based on the improved S-transform, analyze the measured blasting vibration data to obtain the time-frequency structure of the blasting signal.

[0067] Specifically, the improved S-transform includes the following steps:

[0068] Step 2.1 EMD processing: Perform empirical mode decomposition (EMD) on the measured signal s(i) to remove the IMF components of high-frequency noise; reconstruct the remaining components to obtain the reconstructed signal s′(i);

[0069] Step 2.2 S-transform: Perform the S-transform on the reconstructed signal s′(i) to obtain its time-frequency structure;

[0070] The principle of the S-wave transform is as follows:

[0071] First, for the time-frequency analysis calculation of any signal, the calculation formula is as follows:

[0072] F(f) = ∫x(t)g(τ - t, f)e -i2πft dt

[0073] Wherein, F(f) represents the time-frequency analysis result of the signal x(t), which is a function of the frequency f and describes the time-frequency characteristics of the signal at the frequency f; x(t) represents the input time-domain signal; t is the time variable; g(τ - t, f) is the window function in time-frequency analysis, which is used to localize the signal in the time-frequency domain; is the time translation parameter, which is used to determine the center position of the window function g(τ - t, f) and determines the time point of interest in time-frequency analysis; f is the frequency variable, which is used to determine the frequency point of interest in time-frequency analysis and is also used to adjust the width of the window function. The higher the frequency, the narrower the window; e -i2πft is the complex exponential function, which is used to transform the signal from the time domain to the frequency domain; i is the imaginary unit, satisfying i 2 = -1; the role of the complex exponential function is to perform Fourier transform on the signal and extract frequency information; dt represents the integration with respect to the time variable t.

[0074] The specific form of the window function is:

[0075]

[0076] Wherein, τ is the time translation parameter, representing the center position of the window function; f is the frequency variable, which is used to adjust the width of the window function. The higher the frequency, the narrower the window; is the normalization coefficient; is the Gaussian function, which is used to localize the signal on the time axis.

[0077] Among them, the window function g(t) is a Gaussian function, which is used to localize the signal in time-frequency analysis, and its form is:

[0078]

[0079] Wherein, |f| is the absolute value of the frequency, which is used to adjust the width of the window function. The higher the frequency, the narrower the window; is the normalization coefficient, ensuring that the integral area of the window function is 1; is the core part of the Gaussian function, which is used to localize the signal on the time axis.

[0080] Through the above calculation, by performing time-frequency analysis on the signal x(t), the time-frequency domain representation F(f) is obtained; the window function g(τ - t, f) is used to localize the signal in the time-frequency domain, and the complex exponential function e -i2πft is used to achieve frequency domain transformation; the final result F(f) is a function of the frequency f and can reflect the time-frequency characteristics of the signal at the frequency f.

[0081] The S transform can be expressed as follows:

[0082]

[0083] Wherein, ST(τ,f) represents the S-transform result of the signal x(t), which is a time-frequency domain representation.

[0084] Step 2.3 Determine the R-wave time-frequency space: After determining the arrival times of the P-wave and S-wave, preliminarily determine the time-frequency space of the R-wave.

[0085] Step 3: Based on wavelet transform, filter the R-wave region in the time-frequency structure to obtain the R-wave signal time-frequency data.

[0086] Specifically, the R-wave time-frequency domain filtering based on wavelet transform includes the following steps:

[0087] Step 3.1: Based on the determined R-wave time-frequency space, carry out the R-wave filter design;

[0088]

[0089] Wherein, Fil represents the output value of the filter, which is a binary function with values of 1 or 0, used to indicate whether to retain the component of the signal at the time-frequency point (t,f); t is the time variable, which is the time position of the signal in the time-frequency domain; f is the frequency variable, which is the frequency position of the signal in the time-frequency domain; Z R is the R-wave time-frequency space, which is a region in the time-frequency domain. In this region, the main component of the signal is the R-wave. The specific range of Z R needs to be determined through time-frequency analysis, that is, the improved S-transform; (t,f)∈Z R indicates that the time-frequency point (t,f) is located in the R-wave time-frequency space Z R If this condition is met, the filter output Fil = 1, and the signal component at this time-frequency point is retained; indicates that the time-frequency point (t,f) is not in the R-wave time-frequency space. If this condition is met, the filter output Fil = 0, and the signal component at this time-frequency point is filtered out.

[0090] Z R is the R-wave time-frequency space, which needs to be determined through time-frequency analysis, that is, the improved S-transform; Fil is a binary filter used to extract the R-wave signal in the time-frequency domain; the design of this filter is simple and intuitive, which can effectively separate the R-wave signal and avoid affecting the body wave signal at the same time.

[0091] Step 3.2: Use the filter Fil to filter the signal in the R-wave time-frequency space to obtain the corresponding R-wave signal time-frequency data:

[0092] ST R = ST(s′(t))·Fil(t,f)

[0093] Wherein, ST RRepresents the time-frequency data of the filtered R-wave signal, which is obtained by multiplying the time-frequency representation ST(s′(t)) of the original signal by the filter Fil(t,f). ST(s′(t)) represents the S-transform result of the reconstructed signal s′(t), which is a time-frequency domain representation; ST is the operator of the S-transform, which converts the time-domain signal s′(t) into a time-frequency domain representation; s′(t) is the signal after preprocessing, i.e., EMD denoising and reconstruction, and it is the original signal s(t).

[0094] This formula extracts the time-frequency data ST of the R-wave signal by applying the filter Fil(t,f) to the time-frequency representation ST(s′(t)) of the signal R ; the function of the filter is to retain the signal components in the R-wave time-frequency space Z R while filtering out other components, such as: body wave signals; the final result ST R is the time-frequency representation of the R-wave signal, which can be used for further analysis or inverse transformation to obtain the time-domain R-wave signal; thus achieving the goal of accurately extracting the R-wave signal in the time-frequency domain.

[0095] Step 4: Based on the inverse S-transform, analyze and process the R-wave time-frequency data to obtain the R-wave signal.

[0096] Specifically, the inverse S-transform processing includes the following steps:

[0097] Step 4.1: Construct an inverse S-transform model:

[0098] Construct an inverse S-transform model, and the expression is as follows:

[0099]

[0100] Among them, the window function g(τ - t, f) satisfies:

[0101]

[0102] In the formula, x(t) represents the reconstructed time-domain signal, which is the target output of the inverse S-transform. It is obtained by performing an inverse transform on the time-frequency domain data ST(τ, f); ST(τ, f) represents the S-transform result of the signal, which is a time-frequency domain representation; τ is the time variable, representing the time position in the time-frequency domain; f is the frequency variable, representing the frequency position in the time-frequency domain; dτ represents the integral with respect to the time variable τ; df represents the integral with respect to the frequency variable f; e i2πft is the complex exponential function, which is used to convert the frequency-domain signal back to the time domain; i is the imaginary unit, satisfying i 2 = -1; the role of the complex exponential function is to perform an inverse Fourier transform on the signal to restore the time-domain information; g(τ - t, f) is the window function in the S-transform, which is used to localize the signal in the time-frequency domain.

[0103] The inverse S transform recovers the original time-domain signal x(t) by integrating and performing an inverse Fourier transform on the time-frequency domain data ST(τ, f); the normalization condition of the window function g(τ - t, f) ensures the correctness of the inverse transform; this formula realizes the signal reconstruction from the time-frequency domain to the time domain.

[0104] Step 4.2, inverse S transform processing:

[0105] Perform inverse S transform processing on the time-frequency data of the R-wave signal to obtain the R-wave signal:

[0106] R = ST -1 (ST R ) = ST -1 [ST(s′(t))·Fil(t, f)]

[0107] In the formula, R represents the reconstructed R-wave time-domain signal, which is obtained by performing an inverse S transform on the R-wave time-frequency data ST R ; ST -1 represents the inverse S transform operator, which is used to convert the time-frequency domain data back to the time-domain signal; ST R represents the time-frequency data of the filtered R-wave signal; ST(s′(t)) represents the S transform result of the reconstructed signal s′(t); s′(t) represents the reconstructed signal after preprocessing; Fil(t, f) represents the time-frequency domain filter, which is used to extract the R-wave signal.

[0108] In summary, the method for picking up the R wave based on the measured blasting vibration signal at the engineering scale improves the S transform method to obtain the time-frequency structure of the body wave and the R-wave region of the vibration signal, combines wavelet filtering means, processes the R-wave region, filters the R-wave region without affecting the body wave signal, and realizes the picking up of the R wave of the measured blasting vibration signal.

[0109] Furthermore, to verify the actual effect of the method for picking up the R wave based on the measured blasting vibration signal at the engineering scale, taking the measured vibration waveform of a certain engineering rock mass blasting excavation production test as an example, the picking up of the R wave of the blasting vibration signal at the engineering scale is carried out.

[0110] First, arrange blasting vibration measuring instruments, as Figure 2 shown, use RTK surveying instruments to measure the center of the blasting source and the vibration monitoring points, and calculate the blast center distance of the corresponding measuring points as Figure 2 shown; then, based on the improved S transform, analyze the measured blasting vibration data of the 2# vibration measuring point to obtain the time-frequency structure of the blasting signal as Figure 3 shown; as Figure 4 shown, based on wavelet transform, filter the R-wave region in the time-frequency structure to obtain the time-frequency data of the R-wave signal; finally, based on the inverse S transform of the S wave, analyze and process the R-wave time-frequency data to obtain the R-wave signal, asFigure 5 As shown, in the figure, a is the measured blasting vibration signal at the 6# vibration measurement point, and b in the figure is the result of R-wave picking.

[0111] It can be seen that this R-wave picking method based on the measured blasting vibration signal at the engineering scale obtains the time-frequency structure of the body wave and the R-wave region of the vibration signal by improving the S-transform method, and combines wavelet filtering means to process the R-wave region. On the premise of not affecting the body wave signal, filtering processing is carried out on the R-wave region to achieve accurate picking of the R-wave in the measured blasting vibration signal.

[0112] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments. The above embodiments and the descriptions in the specification only illustrate the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention will have various changes and improvements, and these changes and improvements all fall within the scope of the present invention claimed. The scope of the present invention claimed is defined by the appended claims and their equivalents.

Claims

1. A R-wave picking method based on engineering-scale measured blasting vibration signals, characterized in that: The R-wave picking method based on the engineering-scale measured blasting vibration signal comprises the following steps: Step 1: Arrange a blasting vibration meter to monitor blasting-induced vibration signals; Step 2: Based on the improved S transform, the measured vibration data of the blasting is analyzed to obtain the time-frequency structure of the blasting signal; Step 3, based on wavelet transform, filtering is performed on the R wave region in the time-frequency structure to obtain the time-frequency data of the R wave signal; Step 4: Based on the inverse S transform, analyze and process the R wave time-frequency data to obtain the R wave signal.

2. The R-wave picking method based on engineering-scale measured blasting vibration signals according to claim 1 is characterized in that: In step 1, a blasting vibrometer is used to monitor the vibration signal induced by the blasting; an RTK measuring instrument is used to measure the blasting source center and the vibration monitoring point to obtain coordinate information; and the distance between each measuring point and the blasting source center is calculated based on the coordinate information.

3. The R-wave picking method based on engineering-scale measured blasting vibration signals according to claim 1 is characterized in that: In step 2, the improved S transform includes the following steps: Step 2.1 EMD processing: Perform EMD on the measured signal s(i) to remove the IMF component of the high-frequency noise; reconstruct the remaining components to obtain the reconstructed signal s′(i); Step 2.2 S transform: Perform S transform on the reconstructed signal s′(i) to obtain its time-frequency structure; The S-transform formula is: In the formula, ST(τ,f) represents the S-transform result of the signal x(t), which is a time-frequency domain representation; represents the input time-domain signal; τ is the time shift parameter, which represents the center position of the window function; f is the frequency variable, which is used to determine the frequency point of interest in time-frequency analysis. It is also used to adjust the width of the window function. The higher the frequency, the narrower the window; x(t) represents the input time-domain signal; t is the time variable; |f| is the absolute value of the frequency, which is used to adjust the width of the window function. The higher the frequency; is the normalization coefficient, ensuring that the integral area of ​​the window function is 1; is a Gaussian function used to localize the signal on the time axis; e -i2πft is a complex exponential function used to convert signals from the time domain to the frequency domain; i is an imaginary unit, satisfying i 2 =-1; dt represents the integral of the time variable t; Step 2.3 Determine the time-frequency space of the R wave: After determining the first arrival times of the P wave and the S wave, preliminarily determine the time-frequency space of the R wave.

4. The R-wave picking method based on engineering-scale measured blasting vibration signals according to claim 1 is characterized in that: In step 3, the R-wave time-frequency domain filtering based on wavelet transform includes the following steps: Step 3.1, R wave filter design: Where Fil represents the output value of the filter; t is the time variable; f is the frequency variable; Z R is the R wave time-frequency space; Step 3.2, filtering: Use the filter to filter the signal in the R wave time-frequency space to obtain the time-frequency data of the R wave signal: ST R =ST(s′(t))·Fil(t,f) In the formula, ST R represents the time-frequency data of the filtered R wave signal; ST(s′(t)) represents the S-transform result of the reconstructed signal s′(t); ST is the operator of S-transform; s′(t) is the signal after preprocessing, i.e., EMD denoising and reconstruction.

5. The R-wave picking method based on engineering-scale measured blasting vibration signals according to claim 1 is characterized in that: In step 4, the S-based inverse transform process includes the following steps: Step 4.1, construct the S inverse transformation model: Construct the S inverse transformation model, the expression is as follows: Among them, the window function g(τ-t,f) satisfies: Where x(t) represents the reconstructed time domain signal; ST(τ,f) represents the S-transform result of the signal; τ is the time variable; f is the frequency variable; dτ represents the integral of the time variable τ; df represents the integral of the frequency variable f; e i2πft is a complex exponential function; g(τ-t,f) is the window function in the S transform; Step 4.2, S inverse transform processing: Perform S inverse transform on the time-frequency data of the R wave signal to obtain the R wave signal: R=ST -1 (ST R )=ST -1 [ST(s′(t))·Fil(t,f)] Where R represents the reconstructed R wave time domain signal; ST -1 Represents the S inverse transformation operator; ST R represents the time-frequency data of the R wave signal after filtering; ST(s′(t)) represents the S-transform result of the reconstructed signal s′(t); s′(t) represents the reconstructed signal after preprocessing; Fil(t,f) represents the time-frequency domain filter used to extract the R wave signal.

Images