Heading machine working abnormal sound detection method and system based on sound wave signal analysis

Abstract

The invention provides a heading machine working abnormal sound detection method and system based on sound wave signal analysis, and relates to the technical field of heading machine detection.The method comprises the steps that dynamic continuous sound monitoring is conducted on a heading machine through a distributed piezoelectric ceramic sound wave probe, and original array sound wave data is preprocessed and analyzed according to a sound wave preprocessing mechanism; and calling an energy focusing estimation strategy to carry out estimation analysis on the reflection matrix, forming a working sound visual graph of the heading machine according to the wave velocity distribution information, obtaining an abnormal sound source part, predicting to obtain an abnormal type of the abnormal sound source part, and carrying out early warning processing on the abnormal type of the abnormal sound source part. The technical problems that in the prior art, heading machine abnormal sound detection depends on manual and simple sensor monitoring, abnormal sound positioning and recognition accuracy is low, and real-time performance is poor are solved. The technical effects of improving the accuracy and the real-time performance of abnormal sound positioning and identification of the heading machine and guaranteeing the running safety and the fault early warning capability of the heading machine are achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of tunneling machine detection, in particular to a tunneling machine working abnormal sound detection method and system based on sound wave signal analysis. BACKGROUND

[0002] In the field of modern coal mining and underground engineering construction, the tunneling machine as a key construction equipment, its running state is directly related to the progress and safety of the project. The current tunneling machine working abnormal sound detection mainly relies on manual inspection and traditional simple monitoring means. Manual inspection is easily affected by subjective factors, and different personnel have different abilities to perceive and judge abnormal sounds. In addition, in complex and harsh construction environment, manual inspection is difficult to accurately capture abnormal sound signals. The traditional simple monitoring means generally have problems such as limited sensor layout, insufficient spatial coverage, large noise environment interference, and insufficient signal processing capability, which leads to the inability to accurately identify and locate abnormal sound in time, and is difficult to guarantee the construction safety and fault warning capability.

[0003] The existing technology of tunneling machine abnormal sound detection relies on manual and sensor simple monitoring, which has the technical problems of low accuracy of abnormal sound positioning and identification and poor real-time performance. SUMMARY

[0004] The present application provides a tunneling machine working abnormal sound detection method and system based on sound wave signal analysis, which is used to solve the technical problems of low accuracy of abnormal sound positioning and identification and poor real-time performance in the existing technology of tunneling machine abnormal sound detection relying on manual and sensor simple monitoring.

[0005] In view of the above problems, the present application provides a tunneling machine working abnormal sound detection method and system based on sound wave signal analysis.

[0006] In a first aspect of the present application, a tunneling machine working abnormal sound detection method based on sound wave signal analysis is provided, which comprises: continuously monitoring the sound of the tunneling machine by a distributed piezoelectric ceramic sound wave probe to obtain original array sound wave data; pre-processing and analyzing the original array sound wave data according to a sound wave preprocessing mechanism to obtain a reflection matrix; estimating and analyzing the reflection matrix by calling an energy focusing estimation strategy to obtain wave velocity distribution information; forming a working sound visual graph of the tunneling machine according to the wave velocity distribution information, and obtaining an abnormal sound source part in the working sound visual graph; predicting the abnormal type of the abnormal sound source part, and performing early warning processing of the abnormal type on the abnormal sound source part.

[0007] Preferably, the method further comprises: the distributed piezoelectric ceramic acoustic wave probe is arranged on the surface of the cutter head of the tunneling machine based on a radial line layout plan, and is completed by a rotation completion plan; wherein the radial line layout plan refers to arranging 6 radial lines with an included angle of 60° on the surface of the cutter head, and arranging M probes on each line, M is an integer greater than or equal to 3 and less than or equal to 5; wherein the rotation completion plan refers to automatically relocking the surface of the cutter head after rotating 60°, 120° or 180° clockwise, wherein the coupling state of the distributed piezoelectric ceramic acoustic wave probe and the surface of the cutter head remains unchanged.

[0008] Preferably, the original array acoustic wave data is preprocessed and analyzed according to the acoustic wave preprocessing mechanism to obtain a reflection matrix, including: preprocessing the original array acoustic wave data according to a primary preprocessing scheme in the acoustic wave preprocessing mechanism to obtain first-level acoustic wave data; preprocessing the first-level acoustic wave data according to a secondary preprocessing scheme in the acoustic wave preprocessing mechanism to obtain second-level acoustic wave data; and establishing the reflection matrix based on the second-level acoustic wave data; wherein the primary preprocessing scheme includes preprocessing steps of direct wave cutting, band-pass filtering, FK filtering, energy equalization and gain compensation.

[0009] Preferably, the first-level acoustic wave data is preprocessed according to the secondary preprocessing scheme in the acoustic wave preprocessing mechanism to obtain second-level acoustic wave data, including: according to the secondary preprocessing scheme, activating a prediction filter to preprocess the first-level acoustic wave data to obtain the second-level acoustic wave data; wherein before activating the prediction filter, it includes: calculating a first local inclination of a first acoustic wave signal received by a first acoustic wave probe according to a plane wave deconstruction strategy, wherein the first acoustic wave probe refers to any one of the distributed piezoelectric ceramic acoustic wave probes; calculating a first inclination variance based on the first local inclination, and taking the first inclination variance as a first structural complexity index of a first part, wherein the first part refers to a position on the surface of the cutter head monitored by the first acoustic wave probe; matching a first filter factor length corresponding to the first structural complexity index, and constructing the prediction filter based on the first filter factor length.

[0010] Preferably, calculating the first local tilt angle of the first acoustic signal received by the first acoustic probe according to the plane wave deconstruction strategy includes: performing a Fourier transform on the first acoustic signal to obtain a first frequency domain signal; calculating the first local tilt angle based on the first frequency domain signal, wherein calculating the first local tilt angle includes: obtaining the first adjacent probe of the first acoustic probe; comparing the first acoustic probe with the first adjacent probe to obtain a first comparison parameter, wherein the first comparison parameter includes a first phase delay and a first probe spacing; obtaining the first signal frequency of the first frequency domain signal, and calculating the first local tilt angle according to the plane wave deconstruction strategy, combined with the first phase delay and the first probe spacing.

[0011] Preferably, the method further includes: the first structural complexity index is proportional to the length of the first filter factor.

[0012] Preferably, the energy focusing estimation strategy is invoked to estimate and analyze the reflection matrix to obtain wave velocity distribution information, including: performing Radon transform on the reflection matrix and extracting the energy focusing width at the slope k=-1 of the main diagonal; and analyzing the energy focusing width according to the energy focusing estimation strategy to obtain the wave velocity distribution information.

[0013] Preferably, forming the working sound visualization of the tunneling machine based on the wave velocity distribution information includes: calculating the first pseudospectral energy intensity of the first acoustic signal and forming a pseudospectral energy function; using the azimuth angle corresponding to the maximum value of the pseudospectral energy function as a weighting coefficient to perform weighted correction on the Green function to obtain a target Green function; processing the target Green function using a time reversal operator to form the working sound visualization, wherein forming the working sound visualization includes: reversing the target Green function on the time axis and convolving the reversed target Green function with the first acoustic signal to obtain a first convolution result; integrating the first convolution result to obtain the working sound visualization.

[0014] Preferably, predicting the anomaly type of the abnormal noise source location includes: collecting real-time operating status information of the abnormal noise source location; traversing the vectorized real-time operating status information in the historical operation database to obtain the most similar historical operating status information; extracting the historical anomaly type from the most similar historical record corresponding to the most similar historical operating status information, and using it as the anomaly type.

[0015] In a second aspect of the present application, a tunneling machine abnormal sound detection system based on acoustic signal analysis is provided, and the system comprises: an acoustic monitoring module configured to perform dynamic and continuous sound monitoring on the tunneling machine by using a distributed piezoelectric ceramic acoustic probe to obtain original array acoustic data; a data preprocessing and analysis module configured to perform preprocessing and analysis on the original array acoustic data according to an acoustic preprocessing mechanism to obtain a reflection matrix; an energy focusing estimation module configured to call an energy focusing estimation strategy to perform estimation and analysis on the reflection matrix to obtain wave velocity distribution information; an abnormal sound source acquisition module configured to form a working sound visual map of the tunneling machine according to the wave velocity distribution information, and acquire an abnormal sound source part in the working sound visual map; and an abnormality prediction and early warning module configured to predict an abnormal type of the abnormal sound source part, and perform early warning processing of the abnormal type on the abnormal sound source part.

[0016] The one or more technical solutions provided in the present application have at least the following technical effects or advantages: The method provided by the embodiments of the present application performs dynamic and continuous sound monitoring on the tunneling machine by using a distributed piezoelectric ceramic acoustic probe to obtain original array acoustic data; performs preprocessing and analysis on the original array acoustic data according to an acoustic preprocessing mechanism to obtain a reflection matrix; calls an energy focusing estimation strategy to perform estimation and analysis on the reflection matrix to obtain wave velocity distribution information; forms a working sound visual map of the tunneling machine according to the wave velocity distribution information, and acquires an abnormal sound source part in the working sound visual map; predicts an abnormal type of the abnormal sound source part, and performs early warning processing of the abnormal type on the abnormal sound source part. The technical effect of improving the accuracy and real-time performance of abnormal sound positioning and identification of the tunneling machine, and guaranteeing the operation safety and fault early warning capability of the tunneling machine is achieved. BRIEF DESCRIPTION OF DRAWINGS

[0017] Figure 1 A flowchart of a tunneling machine abnormal sound detection method based on acoustic signal analysis provided by the present application is shown.

[0018] Figure 2 A structural diagram of a tunneling machine abnormal sound detection system based on acoustic signal analysis provided by the present application is shown.

[0019] Marked with reference signs: acoustic monitoring module 10, data preprocessing and analysis module 20, energy focusing estimation module 30, abnormal sound source acquisition module 40, and abnormality prediction and early warning module 50. DETAILED DESCRIPTION

[0020] This application provides a method and system for detecting abnormal noises in tunnel boring machines (TBMs) based on acoustic signal analysis. This addresses the technical problems of existing TBM noise detection methods, which rely on manual labor and simple sensor monitoring, resulting in low accuracy and poor real-time performance in noise location and identification. The method improves the accuracy and real-time performance of abnormal noise location and identification in TBMs, thereby enhancing construction safety and fault early warning capabilities.

[0021] The technical solutions of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. It should be understood that the present invention is not limited to the exemplary embodiments described herein. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention. It should also be noted that, for ease of description, only the parts related to the present invention are shown in the accompanying drawings, not all of them.

[0022] Example 1, as Figure 1 As shown, this application provides a method for detecting abnormal noises during tunneling machine operation based on acoustic signal analysis. The method includes: Dynamic and continuous sound monitoring of the tunneling machine is performed using distributed piezoelectric ceramic acoustic probes to obtain raw array acoustic data.

[0023] Specifically, piezoelectric ceramic acoustic wave probes are sensitive elements that utilize the piezoelectric effect to convert the mechanical vibration signals of sound waves into measurable electrical signals. They feature high sensitivity, fast response speed, and adaptability to harsh environments, enabling them to capture subtle structural sound changes in high-noise and high-vibration environments. By employing a distributed layout, multiple piezoelectric ceramic acoustic wave probes are installed on the cutterhead surface of a tunneling machine. This distributed layout allows for the capture of sound signals generated during the cutterhead's operation from different positions and angles, avoiding monitoring blind spots and ensuring the completeness and representativeness of the acquired sound information.

[0024] During the operation of the tunneling machine, the distributed piezoelectric ceramic acoustic probes are activated to track the changes in the working status of the tunneling machine in real time and continuously monitor the sound, acquiring multiple real-time acoustic data. The acoustic data from different probes are arranged and combined according to the spatial distribution order and time order of the probe array to form the original array acoustic data, which can completely reflect the sound field changes at different positions of the cutterhead throughout the construction process, improving the comprehensiveness and reliability of the data.

[0025] Further, the method further comprises: the distributed piezoelectric ceramic acoustic wave probe is arranged on the surface of the cutter head of the tunneling machine based on a radial survey line plan, and is completed by a rotation completion plan; wherein the radial survey line plan refers to that six radial survey lines with an included angle of 60° are arranged on the surface of the cutter head, and M probes are arranged on each survey line, M is an integer greater than or equal to 3 and less than or equal to 5; wherein the rotation completion plan refers to that the surface of the cutter head is automatically relocked after rotating 60°, 120° or 180° clockwise, wherein the coupling state of the distributed piezoelectric ceramic acoustic wave probe and the surface of the cutter head is unchanged.

[0026] Specifically, the distributed piezoelectric ceramic acoustic wave probe is arranged on the surface of the tunneling machine in a radial survey line plan, and is completed in combination with a rotation completion plan. The radial survey line plan is the basic framework of the entire probe arrangement. On the surface of the cutter head of the tunneling machine, the cutter head surface takes the cutter head center as the reference, six survey lines are arranged in a radial distribution, and the included angle between adjacent two survey lines is accurately set to 60°. M piezoelectric ceramic acoustic wave probes are uniformly arranged on each radial survey line, and the value range of M is an integer greater than or equal to 3 and less than or equal to 5. This uniform radial layout can form a uniform spatial sampling grid without increasing the number of probes, fully cover the surface of the cutter head, and ensure that the sound signals during the operation of the cutter head are captured from different angles. In order to further improve the comprehensiveness and accuracy of acoustic monitoring, a rotation completion plan is introduced. The rotation completion plan refers to that after the initial radial survey line plan is arranged, the surface of the cutter head is rotated in a clockwise direction, and the rotation angle is 60°, 120° or 180°. After each rotation, the cutter head is automatically relocked, so that the probe array repeatedly covers the monitoring area in multiple different spatial orientations while maintaining the coupling state of the probe and the surface of the cutter head unchanged. The coupling state refers to the close contact state between the probe and the measured surface in terms of acoustics and mechanics, so as to ensure the stability of signal acquisition and truly reflect the structural acoustic characteristics. In this way, the area originally not covered by the initial radial survey line can be monitored by the probe at the new position, thereby realizing the full coverage of the sound monitoring of the surface of the cutter head. For example, after the cutter head rotates 60° clockwise, the area originally between two survey lines is now covered by a new survey line, and the probe can collect the sound signals of the area, further enriching the monitoring data. Through the combination of the radial survey line plan and the rotation completion plan, the sound signals during the operation of the cutter head of the tunneling machine can be comprehensively and meticulously collected, the comprehensive and accurate original array acoustic wave data can be obtained, and the accurate positioning and type identification of the abnormal sound source can be effectively supported.

[0027] According to the acoustic wave preprocessing mechanism, the original array acoustic wave data is preprocessed and analyzed to obtain a reflection matrix.

[0028] Specifically, the original array acoustic wave data collected by monitoring is subjected to noise suppression, interference removal and feature enhancement by using a pre-set acoustic wave preprocessing mechanism. The acoustic wave preprocessing mechanism includes a series of signal processing techniques, such as direct wave cutting, filtering, energy equalization, gain compensation and the like, and in combination with targeted algorithms such as predictive filtering, a plurality of characteristic data capable of reflecting the propagation and reflection characteristics of the acoustic wave in the heading machine structure are extracted, and the plurality of characteristic data are integrated according to the spatial and temporal correspondence of the probe array to form a reflection matrix, thereby providing effective data support for abnormal sound source positioning.

[0029] Further, the original array acoustic wave data is preprocessed and analyzed according to the acoustic wave preprocessing mechanism to obtain a reflection matrix, including: preprocessing the original array acoustic wave data according to a primary preprocessing scheme in the acoustic wave preprocessing mechanism to obtain first-level acoustic wave data; preprocessing the first-level acoustic wave data according to a secondary preprocessing scheme in the acoustic wave preprocessing mechanism to obtain second-level acoustic wave data; and establishing the reflection matrix based on the second-level acoustic wave data; wherein the primary preprocessing scheme includes preprocessing steps of direct wave cutting, band-pass filtering, FK filtering, energy equalization and gain compensation.

[0030] Specifically, the original array acoustic wave data is preprocessed and analyzed according to the acoustic wave preprocessing mechanism. First, a primary preprocessing scheme is executed, that is, for the original array acoustic wave data collected by the distributed piezoelectric ceramic acoustic wave probe, direct wave cutting, band-pass filtering, FK filtering, energy equalization and gain compensation are sequentially performed. The direct wave cutting removes the direct propagation signal between the probe and the sound source by time window interception to reduce its interference with the reflection characteristics. The band-pass filtering suppresses low-frequency and high-frequency noise by limiting the frequency range, and only retains the effective frequency band related to the structural characteristics. The FK filtering refers to frequency-wavenumber filtering, which selectively retains the wave field components of a specific propagation direction by using the wavenumber domain characteristics, thereby suppressing irrelevant wave energy. The energy equalization adjusts the signal energy of different probes and different time slices to make the overall data amplitude distribution more consistent, avoiding that some strong energy channels mask weak reflection signals. The gain compensation corrects the amplitude of signals with long propagation distance or multiple reflections that cause energy attenuation. After the above processing, first-level acoustic wave data that suppresses interference and retains main structural reflection characteristics is obtained.

[0031] Then a secondary preprocessing scheme is entered, in which a prediction filter is used to further eliminate residual noise and non-stationary interference, parameters of the prediction filter being dynamically determined according to a plane wave decomposition strategy, a filter factor length being matched by calculating a local dip angle and a variance thereof, so that the filter is adaptively adjusted in monitoring areas with different structural complexities. After the secondary processing, secondary acoustic wave data with higher signal-to-noise ratio and spatial resolution are obtained, and a reflection matrix is established based on the obtained secondary acoustic wave data. The reflection matrix is a two-dimensional data array in which inter-channel cross-correlation or reflection coefficient information is structured and stored according to a spatial layout and a time sequence of a probe array, and can comprehensively describe propagation and reflection characteristics of acoustic waves in a structure. The original array acoustic wave data are preprocessed by the primary preprocessing scheme, and the data are further optimized by the secondary preprocessing scheme, so that the data quality is improved, and the accuracy and reliability of abnormal sound source detection are improved.

[0032] Further, the primary acoustic wave data are preprocessed according to the secondary preprocessing scheme in the acoustic wave preprocessing mechanism to obtain secondary acoustic wave data, including: activating a prediction filter to preprocess the primary acoustic wave data according to the secondary preprocessing scheme to obtain the secondary acoustic wave data; wherein, before the prediction filter is activated, including: calculating a first local dip angle of a first acoustic wave signal received by a first acoustic wave probe according to a plane wave decomposition strategy, wherein the first acoustic wave probe refers to any one of the distributed piezoelectric ceramic acoustic wave probes; calculating a first dip angle variance based on the first local dip angle, and taking the first dip angle variance as a first structural complexity index of a first part, wherein the first part refers to a position on a cutter disc surface monitored by the first acoustic wave probe; matching a first filter factor length corresponding to the first structural complexity index, and constructing the prediction filter based on the first filter factor length.

[0033] The first structural complexity index is directly proportional to the first filter factor length.

[0034] Specifically, after the original array acoustic wave data is preprocessed according to the primary preprocessing scheme in the acoustic wave preprocessing mechanism to obtain first-level acoustic wave data, in order to further suppress residual acoustic wave signal noise and enhance the resolution capability of the structure reflection signal, according to the secondary preprocessing scheme in the acoustic wave preprocessing mechanism, a prediction filter is activated to preprocess the first-level acoustic wave data. The prediction filter is a digital filter based on local signal characteristics for adaptive modeling and interference suppression, and its filtering performance depends on the set filter factor length. Before activating the prediction filter, the prediction filter needs to be constructed, and the specific construction process is as follows: select any one probe from the distributed piezoelectric ceramic acoustic wave probe as the first acoustic wave probe, and the signal received by the first acoustic wave probe is the first acoustic wave signal. Through the plane wave deconstruction strategy, the first local inclination is obtained by analyzing and calculating the first acoustic wave signal. The first local inclination reflects the propagation direction of the acoustic wave at a specific position and time. After obtaining the first local inclination, the first inclination variance is further calculated based on the first local inclination by statistical analysis of the first local inclination within a certain spatio-temporal sampling range. Taking the monitoring position of the first acoustic wave probe as the center, a continuous sampling point in the time sequence is selected, and the inclination values of the adjacent measuring points in space are combined to calculate the average value of multiple inclination samples, and then the deviation square of each sample relative to the average value is calculated and averaged to obtain the first local inclination variance. The first local inclination variance is a quantitative description of the fluctuation degree of the local inclination, which can reflect the fluctuation degree of the acoustic wave propagation direction with time and space. The first inclination variance corresponds to the position of the cutter head surface, i.e. the first part monitored by the first acoustic wave probe. The first structure complexity index is the first inclination variance, and the larger the first structure complexity index, the more intense the acoustic field change at the part. Based on experimental data and expert experience, a proportionality coefficient or a mapping function is preset to map the structure complexity index value obtained from the first inclination variance to the corresponding filter factor length range. The first structure complexity index is directly proportional to the first filter factor length, for example, the minimum and maximum filter factor length value interval is specified in the filter design, and the lowest complexity corresponds to the minimum length and the highest complexity corresponds to the maximum length, and the intermediate values are linearly interpolated or calculated by a segmented function to obtain the specific length parameter. After matching the first filter factor length, the first filter factor length is input as the order parameter of the prediction filter into the filter design process, the type of the prediction filter is selected, such as the linear prediction filter based on the autoregressive model, and then the first-level acoustic wave data is used as the training input to iteratively solve the filter coefficients by the least mean square error or least square adaptive algorithm, so that the prediction filter can predict the current sampling point using the historical sampling points, and effectively separate and suppress the components in the prediction residual that are irrelevant to the structure reflection characteristics, thereby forming an adaptive optimized prediction filter.The first-order sound wave data is filtered by a prediction filter, so that second-order sound wave data with higher signal-to-noise ratio and clearer waveform characteristics is obtained, and the detection accuracy and reliability of the tunneling machine working abnormal sound are improved.

[0035] Further, the first local inclination of the first sound wave signal received by the first sound wave probe is calculated according to the plane wave deconstruction strategy, including: performing Fourier transform on the first sound wave signal to obtain a first frequency domain signal; and calculating the first local inclination according to the first frequency domain signal, wherein the calculation of the first local inclination includes: obtaining a first adjacent probe of the first sound wave probe; comparing the first sound wave probe with the first adjacent probe to obtain a first comparison parameter, wherein the first comparison parameter includes a first phase delay and a first probe spacing; obtaining a first signal frequency of the first frequency domain signal, and calculating the first local inclination according to the plane wave deconstruction strategy, in combination with the first phase delay and the first probe spacing.

[0036] Specifically, in the process of calculating the first local inclination of the first sound wave signal received by the first sound wave probe according to the plane wave deconstruction strategy, the first sound wave signal is first subjected to Fourier transform to obtain a first frequency domain signal. Fourier transform is a mathematical tool that can convert a time domain signal into a frequency domain signal. Through this conversion, the first frequency domain signal is obtained. Then, a first adjacent probe of the first sound wave probe is obtained, which refers to a piezoelectric ceramic sound wave probe adjacent to the first sound wave probe in spatial position. The sound wave signal data of the signals received by the first sound wave probe and the first adjacent probe in the same time window are extracted and compared, and a first comparison parameter is obtained, wherein the first comparison parameter includes a first phase delay and a first probe spacing. Specifically, the sound wave data signals of the first sound wave probe and the first adjacent probe are analyzed using a cross-correlation function. The position offset corresponding to the function peak value can be used to calculate the first phase delay, i.e., the first phase delay refers to the difference in phase of the same signal received by two probes, which reflects the time sequence of the arrival of sound waves at different probes. At the same time, the physical distance between the first sound wave probe and the first adjacent probe is directly measured or obtained through preset probe installation parameters with the aid of coordinate information of the probe layout, and the first probe spacing is obtained.

[0037] The first frequency domain signal is subjected to frequency spectrum analysis, the amplitude corresponding to each frequency point is traversed, the frequency corresponding to the peak value of the amplitude in the first frequency domain signal is extracted as the first signal frequency, and is substituted into the calculation formula of the plane wave deconstruction strategy, in combination with the first phase delay and the first probe spacing, to calculate the first local inclination, wherein the calculation formula is: wherein σ is the inclination, For phase delay, f is signal frequency, and Dx is probe spacing. Namely, the first phase delay, the first signal frequency and the first probe spacing obtained are substituted into the formula to calculate the first local dip angle. Through Fourier transform, the signal is converted to the frequency domain, the frequency and phase information of the signal can be more accurately extracted, the adjacent probes are acquired and the contrast parameters are calculated, the difference characteristics of the sound wave propagation at different positions are utilized, and finally the local dip angle is calculated, thereby further improving the detection accuracy and reliability of the tunneling machine working abnormal sound.

[0038] The energy focus estimation strategy is called to estimate and analyze the reflection matrix to obtain wave velocity distribution information.

[0039] Further, the energy focus estimation strategy is called to estimate and analyze the reflection matrix to obtain wave velocity distribution information, including: performing Radon transformation on the reflection matrix, and extracting the energy focus width at k=-1 of the main diagonal line slope; and analyzing the energy focus width according to the energy focus estimation strategy to obtain the wave velocity distribution information.

[0040] Specifically, first, a Radon transform is performed on the reflection matrix, which is an integral transform method of projecting a two-dimensional signal to a parameterized straight line space, and can map the energy distribution of the sound wave at different inclination angles to the slope domain, for identifying the sound wave propagation direction and velocity. After completing the Radon transform, by locating the energy distribution curve in the direction of the main diagonal line slope k = -1 in the slope domain of the transform result, the maximum energy point in this curve is found as the peak position, and with the peak as the center, the position points where the energy decays to a certain threshold, for example, 50% of the peak energy, on both sides of the curve are sought, and the interval length between the two threshold points is calculated, which is the energy focusing width at the main diagonal line slope k = -1. The energy focusing width reflects the degree of concentration of energy in a certain direction, and the narrower the width, the more concentrated the energy. Subsequently, the extracted energy focusing width is analyzed according to the energy focusing estimation strategy, which is an analysis rule and method based on the principle of sound wave propagation and the law of energy distribution. For example, due to the difference in wave velocity, the propagation and reflection of the sound wave in the medium are affected, resulting in differences in energy distribution in the reflection matrix. When the medium wave velocity is high, the energy at a certain slope will present a narrow and concentrated peak in the Radon domain, while when the wave velocity is low, the energy peak will be wider, and the peak position may be shifted. Therefore, the energy focusing estimation strategy of establishing the width-wave velocity mapping relationship can be derived from the sound wave propagation formula or calibrated through experiments by corresponding different wave velocity ranges through peak width and peak position characteristics. Then the energy focusing width is input into the energy focusing estimation strategy for analysis to obtain the corresponding wave velocity distribution information. The wave velocity distribution information refers to the distribution of the propagation velocity of the sound wave at different positions, so that the subsequent sound source visualization imaging and accurate positioning of abnormal parts can be based on the real wave velocity model, thereby improving the imaging accuracy and positioning reliability and improving the abnormal detection accuracy.

[0041] A working sound visual map of the heading machine is formed according to the wave velocity distribution information, and an abnormal sound source position in the working sound visual map is obtained.

[0042] Specifically, the obtained wave velocity distribution information is analyzed, and the wave velocity data corresponding to different positions are converted into visual elements, and then a working sound visual map of the heading machine is constructed, which directly presents the propagation characteristics of sound in each region. The working sound visual map directly and accurately identifies and locates the energy concentration abnormal region, i.e. the abnormal sound source position, by directly displaying the acoustic energy intensity distribution at different positions through different colors or brightness. By converting the wave velocity distribution into a visual map and locating the abnormal sound source, potential faults or abnormal working conditions of the heading machine equipment can be found in time, providing accurate basis for equipment maintenance and fault diagnosis, and ensuring safe and efficient operation of the heading machine.

[0043] Further, the working sound visual map of the tunneling machine is formed according to the wave velocity distribution information, including: calculating the first pseudo-spectrum energy intensity of the first sound wave signal according to the wave velocity distribution information, and forming a pseudo-spectrum energy function; taking the azimuth corresponding to the maximum value of the pseudo-spectrum energy function as a weight coefficient, the Green function is weighted and corrected to obtain a target Green function; the target Green function is processed by using a time reversal operator to form the working sound visual map, wherein forming the working sound visual map includes: inverting the target Green function on the time axis, and convolving the inverted target Green function with the first sound wave signal to obtain a first convolution result; integrating the first convolution result to obtain the working sound visual map.

[0044] Specifically, the wave velocity distribution information reflects the propagation characteristics of sound waves in different positions and directions. According to the wave velocity distribution information, the first sound wave signal received by the first sound wave probe is subjected to pseudo-spectrum analysis, and the spatial array signal processing method such as the MUSIC or Capon algorithm is used to analyze and process the first sound wave signal, extract its frequency and azimuth information, and calculate the first pseudo-spectrum energy intensity. The pseudo-spectrum energy intensity can obtain high-resolution wave source azimuth characteristics under low signal-to-noise ratio conditions. The energy intensity corresponding to each azimuth is combined to form a pseudo-spectrum energy function, which describes the pseudo-spectrum energy distribution of signals at different azimuths.

[0045] The azimuth corresponding to the maximum value of the pseudo-spectrum energy function is used as a weight coefficient to correct the Green function. The Green function is a mathematical function that describes the response of any point in the field under point source excitation. In the problem of sound wave propagation, the Green function reflects the sound field distribution generated by sound waves in space. Due to the influence of various factors in the actual environment, the propagation characteristics of different azimuths are different, so it is necessary to correct them. The target Green function obtained by weighting the Green function with the azimuth corresponding to the maximum value of the pseudo-spectrum energy function as a weight coefficient can more accurately reflect the actual sound wave propagation.

[0046] The time reversal operator is called to process the target Green's function. Time reversal is a signal processing method that reverses the propagation process of a wave in time. Its principle is to play the received signal in reverse. The wave will refocus at the original sound source position during reverse propagation. Specifically, the target Green's function is first reversed in the time axis to obtain the time reversal Green's function, and then convolved with the first sound wave signal. The convolution result reflects the refocusing of the sound wave in time, which is recorded as the first convolution result. Convolution is a mathematical operation that slides the time reversal Green's function and the first sound wave signal in time and multiplies and integrates them, effectively fusing the two. Through this fusion, the key information related to the sound source position and characteristics in the first sound wave signal is extracted. Because the time reversal Green's function carries prior knowledge of the sound propagation path and sound source characteristics, when combined with the first sound wave signal, it can highlight the components in the signal that are closely related to the position and characteristics of the abnormal sound source. Finally, the first convolution result is arranged according to the spatial position and time sequence to form a space-time distribution matrix. Then, the convolution value corresponding to each spatial position in the matrix is numerically integrated in the time dimension. The integration can use numerical methods such as cumulative summation or trapezoidal method to converge the energy information of the position in the entire time period. The integral results of all spatial positions are mapped to the corresponding spatial coordinate system, and visualized through color gradient or brightness change, thus generating a complete two-dimensional or three-dimensional working sound visualization. For example, there are 30 distributed piezoelectric ceramic probes on the surface of the roadheader cutter, and after the previous steps, the first convolution result is obtained. The first convolution result is a two-dimensional matrix with probe positions as rows and time sampling points as columns, such as a 30x1000 matrix, 30 positions, and 1000 convolution values for each position. The 1000 convolution values for each row, i.e., each probe position, are integrated using cumulative summation, resulting in the total energy value of the position in the entire observation period. For example, the first probe integration result is 85, the second probe is 210, and the third probe is 45. These values represent the degree of acoustic energy concentration at each position. According to the actual installation coordinates of each probe on the cutter, the energy values obtained by integration are mapped to the spatial positions of the cutter, such as using two-dimensional coordinates (x, y) to mark the probe positions. Visual rendering is performed, colors are selected, such as blue representing low energy and red representing high energy, and the integral energy values of each probe are converted into color filling. If a more continuous image is needed, spatial interpolation can be performed on the energy values to generate a smooth cutter sound distribution map. The rendered image becomes a working sound visualization, where the area with the most concentrated color corresponds to the position with the strongest energy focus, which is likely to be an abnormal sound source.

[0047] The work sound visual diagram intuitively displays the acoustic energy distribution of each spatial position through the change of color or brightness, so that the area where the energy is abnormally concentrated, i.e., the abnormal sound source position, can be intuitively identified. By calculating the pseudo-spectrum energy intensity and constructing the pseudo-spectrum energy function, the energy characteristics of the first sound wave signal can be accurately grasped, the Green function is weighted and corrected, the accuracy of the sound wave propagation model is improved, the time reversal operator is used for processing and forming the work sound visual diagram, the abstract sound wave signal is converted into an intuitive image, the abnormal sound in the working sound of the roadheader can be accurately and directly analyzed, the potential fault of the roadheader can be found in time, and strong support is provided for the safe operation and maintenance of the roadheader.

[0048] The abnormal type of the abnormal sound source position is predicted, and the abnormal sound source position is subjected to early warning processing of the abnormal type.

[0049] Specifically, after the work sound visual diagram is constructed and the abnormal sound source position is located, real-time running state information of the abnormal sound source position is collected, the abnormal type of the abnormal sound source position is judged in combination with historical data, for example, the abnormal type of the current abnormal sound source position is predicted and identified by comparing and analyzing the currently collected real-time running state information with the historical data. According to the severity of the abnormal type, different levels of early warning signals are set, for example, a first-level early warning represents a slight abnormality, a second-level early warning represents a moderate abnormality, and a third-level early warning represents a serious abnormality. After the predicted type is obtained, the early warning processing mechanism is triggered, early warning information is automatically generated according to the danger level of the abnormal type, and can be displayed on the operation terminal or sound and light alarm is issued, and automatic shutdown or maintenance prompt is triggered if necessary. The sound source positioning result is combined with historical experience data to realize the processing from discovering abnormal sound to judging abnormal type, and then to active early warning, which improves the accuracy and response speed of abnormal diagnosis in the running process of the roadheader, thereby reducing the fault risk of the roadheader and ensuring the safety of equipment operation.

[0050] Further, the abnormal type of the abnormal sound source position is predicted, including: collecting real-time running state information of the abnormal sound source position; traversing the vectorized real-time running state information in a historical operation database to obtain the most similar historical running state information; extracting the historical abnormal type in the most similar historical record corresponding to the most similar historical running state information as the abnormal type.

[0051] Specifically, real-time running state information is collected for the located abnormal sound source part, which contains not only acoustic characteristics but also working condition parameters of the tunneling machine collected by various high-precision sensors installed on the tunneling machine, such as rotating speed, torque, thrust, vibration acceleration, etc., to ensure that the data comprehensively reflects the running state of the abnormal sound source part. Then, the collected data is vectorized, i.e., different types and dimensions of features are converted into a unified feature vector through normalization or standardization methods, and the vectorized real-time running state information is iterated in a pre-constructed historical operation database which stores a large amount of running state information and corresponding abnormal records in the past operation process of the tunneling machine. The similarity between the vectorized real-time running state information and each record in the historical operation database is calculated through a specific similarity algorithm, such as the cosine similarity algorithm, to obtain the most similar historical running state information to the current feature vector. The historical abnormal type corresponding to the most similar historical running state information is extracted from the most similar historical record and directly used as the predicted type of the current abnormal sound source part. The combination of historical experience and real-time monitoring enables the determination of abnormal types to be based on a large number of existing cases and feature matching, which not only quickly and accurately predicts the abnormal type of the abnormal sound source part, but also provides a direct decision basis for subsequent early warning and maintenance, thereby achieving rapid response and efficient disposal, ensuring the safe and stable operation of the tunneling machine, and improving production efficiency and equipment service life.

[0052] In the second embodiment, based on the same inventive concept as the tunneling machine working abnormal sound detection method in the foregoing embodiments, as shown in FIG. 2, the present application provides a tunneling machine working abnormal sound detection system based on sound wave signal analysis, wherein the system comprises: Figure 2 A sound wave monitoring module 10 is configured to perform dynamic and continuous sound monitoring on the tunneling machine through a distributed piezoelectric ceramic sound wave probe to obtain original array sound wave data. A data preprocessing and analysis module 20 is configured to perform preprocessing and analysis on the original array sound wave data according to a sound wave preprocessing mechanism to obtain a reflection matrix. An energy focusing estimation module 30 is configured to call an energy focusing estimation strategy to perform estimation and analysis on the reflection matrix to obtain wave velocity distribution information. An abnormal sound source acquisition module 40 is configured to form a working sound visual map of the tunneling machine according to the wave velocity distribution information and acquire an abnormal sound source part in the working sound visual map. An abnormality prediction and early warning module 50 is configured to predict an abnormal type of the abnormal sound source part and perform early warning processing of the abnormal type on the abnormal sound source part.

[0053] ​Further, the sound wave monitoring module 10 of the tunneling machine working abnormal sound detection system based on sound wave signal analysis is further used for: the distributed piezoelectric ceramic sound wave probe is arranged on the surface of the cutter head of the tunneling machine based on a radial survey line plan, and is completed by a rotation completion plan; wherein the radial survey line plan refers to that six radial survey lines with an included angle of 60° are arranged on the surface of the cutter head, and M probes are arranged on each survey line, and M is an integer greater than or equal to 3 and less than or equal to 5; wherein the rotation completion plan refers to that the surface of the cutter head is automatically relocked after rotating 60°, 120° or 180° clockwise, and the coupling state of the distributed piezoelectric ceramic sound wave probe and the surface of the cutter head is unchanged.

[0054] Further, the data preprocessing analysis module 20 of the tunneling machine working abnormal sound detection system based on sound wave signal analysis is further used for: preprocessing the original array sound wave data according to a primary preprocessing scheme in the sound wave preprocessing mechanism to obtain first-level sound wave data; preprocessing the first-level sound wave data according to a secondary preprocessing scheme in the sound wave preprocessing mechanism to obtain second-level sound wave data; and establishing the reflection matrix based on the second-level sound wave data; wherein the primary preprocessing scheme includes preprocessing steps of direct wave cutting, band-pass filtering, FK filtering, energy equalization and gain compensation.

[0055] Further, the data preprocessing analysis module 20 of the tunneling machine working abnormal sound detection system based on sound wave signal analysis is further used for: according to the secondary preprocessing scheme, activating a prediction filter to preprocess the first-level sound wave data to obtain the second-level sound wave data; wherein before activating the prediction filter, the following steps are included: calculating a first local inclination angle of a first sound wave signal received by a first sound wave probe according to a plane wave deconstruction strategy, wherein the first sound wave probe refers to any one of the distributed piezoelectric ceramic sound wave probes; calculating a first inclination angle variance based on the first local inclination angle, and taking the first inclination angle variance as a first structure complexity index of a first part, wherein the first part refers to a surface position of the cutter head monitored by the first sound wave probe; matching a first filter factor length corresponding to the first structure complexity index, and constructing the prediction filter based on the first filter factor length.

[0056] Further, the data preprocessing and analysis module 20 of the sound wave signal analysis-based tunneling machine abnormal noise detection system is further configured to: perform Fourier transform on the first sound wave signal to obtain a first frequency domain signal; and calculate the first local inclination based on the first frequency domain signal, wherein the calculation of the first local inclination comprises: obtaining a first adjacent probe of the first sound wave probe; comparing the first sound wave probe with the first adjacent probe to obtain a first comparison parameter, wherein the first comparison parameter comprises a first phase delay and a first probe spacing; obtaining a first signal frequency of the first frequency domain signal, and calculating the first local inclination based on the plane wave deconstruction strategy and in combination with the first phase delay and the first probe spacing.

[0057] Further, the data preprocessing and analysis module 20 of the sound wave signal analysis-based tunneling machine abnormal noise detection system is further configured to: the first structural complexity index is directly proportional to the first filter factor length.

[0058] Further, the energy focusing estimation module 30 of the sound wave signal analysis-based tunneling machine abnormal noise detection system is further configured to: perform Radon transform on the reflection matrix, and extract an energy focusing width at a main diagonal line slope k=-1; and analyze the energy focusing width based on the energy focusing estimation strategy to obtain the wave velocity distribution information.

[0059] Further, the abnormal noise source acquisition module 40 of the sound wave signal analysis-based tunneling machine abnormal noise detection system is further configured to: calculate a first pseudo-spectrum energy intensity of the first sound wave signal based on the wave velocity distribution information, and form a pseudo-spectrum energy function; take an azimuth angle corresponding to a maximum value of the pseudo-spectrum energy function as a weight coefficient to perform weighted correction on a Green function to obtain a target Green function; and process the target Green function by using a time reversal operator to form the working sound visual map, wherein the formation of the working sound visual map comprises: inverting the target Green function on a time axis, and performing convolution processing on the inverted target Green function and the first sound wave signal to obtain a first convolution result; and performing integration on the first convolution result to obtain the working sound visual map.

[0060] Further, the abnormal prediction and early warning module 50 of the sound wave signal analysis-based tunneling machine abnormal noise detection system is further configured to: collect real-time running state information of the abnormal noise source part; traverse the vectorized real-time running state information in a historical operation database to obtain most similar historical running state information; extract a historical abnormal type in a most similar historical record corresponding to the most similar historical running state information, and take the historical abnormal type as the abnormal type.

[0061] The foregoing description of the disclosed embodiments enables a person skilled in the art to make or use the application. Numerous modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein can be applied to other embodiments without the use of the inventive faculty. Therefore, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0062] It will be readily apparent to one skilled in the art that varying substitutions and modifications can be made to the application disclosed herein without departing from the scope and spirit of the application. Moreover, it is the intent that all such variations and modifications be considered as falling within the scope of the application, and that the application be limited only by the definitions contained in the appended claims.

Claims

1. A method for detecting abnormal sound of a boring machine in operation by analyzing a sound wave signal, characterized by, The method comprises the steps of: obtaining original array acoustic wave data by performing dynamic and continuous sound monitoring on the roadheader through a distributed piezoelectric ceramic acoustic wave probe; preprocessing and analyzing the original array acoustic wave data according to an acoustic wave preprocessing mechanism to obtain a reflection matrix; estimating and analyzing the reflection matrix by calling an energy focusing estimation strategy to obtain wave velocity distribution information; forming a working sound visual graph of the roadheader according to the wave velocity distribution information, and obtaining an abnormal sound source position in the working sound visual graph; predicting the abnormal type of the abnormal sound source position, and performing early warning processing on the abnormal sound source position according to the abnormal type.

2. The method according to claim 1, wherein the method is characterized by: The distributed piezoelectric ceramic acoustic wave probe is arranged on the surface of the cutter head of the roadheader based on a radial survey line plan, and is completed through a rotation completion plan. The radial survey line plan refers to arranging six radial survey lines with an included angle of 60° on the surface of the cutter head, and arranging M probes on each survey line, where M is an integer greater than or equal to 3 and less than or equal to 5. The rotation completion plan refers to automatically relocking the surface of the cutter head after rotating 60°, 120° or 180° clockwise, and the coupling state of the distributed piezoelectric ceramic acoustic wave probe and the surface of the cutter head remains unchanged.

3. The method according to claim 1, wherein the method is characterized by: The method comprises the steps of: preprocessing the original array acoustic wave data according to a primary preprocessing scheme in the acoustic wave preprocessing mechanism to obtain first-level acoustic wave data; preprocessing the first-level acoustic wave data according to a secondary preprocessing scheme in the acoustic wave preprocessing mechanism to obtain second-level acoustic wave data; establishing the reflection matrix based on the second-level acoustic wave data; The primary preprocessing scheme comprises preprocessing steps of direct wave cutting, band-pass filtering, FK filtering, energy equalization and gain compensation.

4. The method according to claim 3, wherein the method is characterized by: The method comprises the steps of: According to the secondary preprocessing scheme, activating a prediction filter to preprocess the first-level acoustic wave data to obtain the second-level acoustic wave data; Before activating the prediction filter, the method comprises the steps of: calculating a first local inclination angle of a first acoustic wave signal received by a first acoustic wave probe according to a plane wave deconstruction strategy, wherein the first acoustic wave probe refers to any one of the distributed piezoelectric ceramic acoustic wave probes; calculating a first inclination angle variance based on the first local inclination angle, and taking the first inclination angle variance as a first structural complexity index of a first position, wherein the first position refers to a surface position of the cutter head monitored by the first acoustic wave probe; matching a first filter factor length corresponding to the first structural complexity index, and constructing the prediction filter based on the first filter factor length.

5. The method according to claim 4, wherein the method is characterized by: The method comprises the steps of: performing Fourier transform on the first acoustic wave signal to obtain a first frequency domain signal; calculating the first local inclination angle based on the first frequency domain signal, wherein the calculation of the first local inclination angle comprises: acquiring a first adjacent probe of the first acoustic wave probe; obtaining a first contrast parameter by comparing the first acoustic wave probe and the first adjacent probe, wherein the first contrast parameter comprises a first phase delay and a first probe spacing; acquiring a first signal frequency of the first frequency domain signal, and calculating the first local dip angle according to the plane wave deconstruction strategy, in combination with the first phase delay and the first probe spacing.

6. The method according to claim 4, wherein the method is characterized by: The first structural complexity index is directly proportional to the first filter factor length.

7. The method according to claim 1, wherein the method is characterized by: The energy focusing estimation strategy is called to estimate and analyze the reflection matrix to obtain wave velocity distribution information, including: performing Radon transformation on the reflection matrix, and extracting the energy focusing width at k=-1 of the main diagonal line slope; analyzing the energy focusing width according to the energy focusing estimation strategy to obtain the wave velocity distribution information.

8. The method according to claim 4, wherein the method is characterized by: forming a working sound visual graph of the tunneling machine according to the wave velocity distribution information, including: calculating the first pseudo-spectrum energy intensity of the first acoustic wave signal according to the wave velocity distribution information, and forming a pseudo-spectrum energy function; taking the azimuth angle corresponding to the maximum value of the pseudo-spectrum energy function as a weight coefficient to perform weighted correction on the Green function to obtain a target Green function; processing the target Green function by using a time reversal operator to form the working sound visual graph, wherein forming the working sound visual graph includes: inverting the target Green function on the time axis, and performing convolution processing on the inverted target Green function and the first acoustic wave signal to obtain a first convolution result; integrating the first convolution result to obtain the working sound visual graph.

9. The method according to claim 8, wherein the method is a method for detecting abnormal sound of a boring machine, and the sound signal is a sound signal of the boring machine. predicting an abnormal type of the abnormal sound source part, including: collecting real-time running state information of the abnormal sound source part; traversing the vectorized real-time running state information in a historical operation database to obtain most similar historical running state information; extracting a historical abnormal type in a most similar historical record corresponding to the most similar historical running state information as the abnormal type.

10. A boring machine operation abnormality detection system for acoustic signal analysis, characterized by, Steps for implementing the method of any one of claims 1 to 9, including: an acoustic wave monitoring module (10) for dynamically and continuously monitoring a tunneling machine by a distributed piezoelectric ceramic acoustic wave probe to obtain original array acoustic wave data; a data preprocessing analysis module (20) for preprocessing and analyzing the original array acoustic wave data according to an acoustic wave preprocessing mechanism to obtain a reflection matrix; an energy focusing estimation module (30) for calling an energy focusing estimation strategy to estimate and analyze the reflection matrix to obtain wave velocity distribution information; an abnormal sound source acquisition module (40) for forming a working sound visual graph of the tunneling machine according to the wave velocity distribution information, and acquiring an abnormal sound source part in the working sound visual graph; an abnormality prediction and early warning module (50) for predicting an abnormal type of the abnormal sound source part, and performing early warning processing of the abnormal type on the abnormal sound source part.

Images