Acoustic emission and stress homologous coupling acquisition terminal

Abstract

The invention discloses an acoustic emission and stress homologous coupling acquisition terminal, belongs to the technical field of mining monitoring, and aims to realize the functions of stress and acoustic emission signal synchronous acquisition, waveform recording, data storage, field analysis, data remote transmission and the like. The system comprises an acquisition probe, a processing terminal and an upper computer, wherein the acquisition probe comprises a data acquisition module, an analog-to-digital conversion module and a controller module; the processing terminal is composed of a data processing unit, a man-machine interaction module, a data storage module, a communication module, a power supply module and an RTC module. And the upper computer is data processing software and is responsible for data processing, display, storage and control of a processing terminal. By monitoring impact dynamic disasters of a mine, accidents are effectively avoided. The impact dynamic disaster condition in mining engineering can be effectively predicted and analyzed under the condition that the mining engineering does not make contact with a mine, the stability of surrounding rock of an underground roadway can be comprehensively evaluated at any time, and technical support is provided for safe and efficient production of the mine.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of mine exploitation monitoring, and particularly relates to a sound emission and stress homologous coupling acquisition terminal system. BACKGROUND

[0002] In the process of deep mine exploitation, a series of dynamic disaster problems such as rock burst and rock burst come along, which threaten the personal safety of mining personnel and cause property losses. Therefore, in the deep mine exploitation activities, effectively monitoring and early warning the occurrence of rock burst and rock burst and other impact dynamic disasters is a powerful measure to ensure mine safety.

[0003] At present, the monitoring of mine impact dynamic disasters has stress, strain, displacement, microseismic, acoustic emission and other main technical means. Among them, the stress, strain, displacement and other monitoring and early warning methods have poor dynamic characteristics, monitoring lag, and cannot effectively and dynamically monitor the evolution process from the occurrence to the development of impact dynamic disasters. Microseismic monitoring technology has shown significant advantages in overall monitoring of the mine area due to its wide monitoring range and low frequency characteristics, but in the fine monitoring of local areas such as working faces, it is difficult to capture high-frequency, subtle effective signals due to its lower monitoring frequency domain, and there are certain defects in local effective early warning.

[0004] The acoustic emission monitoring method has a higher monitoring frequency domain and makes up for the shortcomings of microseismic monitoring in local monitoring of working faces. The acoustic emission monitoring method can accurately capture and analyze the small fracture signals from the rock mass without directly contacting the mine structure, providing strong technical support for the impact dynamic disaster early warning of key areas such as working faces.

[0005] Therefore, in recent years, there have been more and more studies on the use of acoustic emission method for monitoring and early warning of mine impact dynamic disasters. For example, the existing patent document with document number CN107764899B proposes a rock mass acoustic emission monitoring and positioning method, device and system, which includes: monitoring the acoustic emission events of the rock mass online; analyzing the data of each acoustic emission sensor channel in real time; determining whether there is an acoustic emission event, if yes, positioning the acoustic emission source; if not, continue to monitor the acoustic emission events of the rock mass online. A rock mass acoustic emission monitoring and positioning device includes: an online monitoring module; a real-time analysis module; a judgment module. A rock mass acoustic emission monitoring and positioning system includes: an acoustic emission sensor, a monitoring host and a data processing computer. The existing technology realizes real-time analysis of waveform data of all channels at all times.

[0006] The prior art with the document number CN120260246A discloses an open-pit mine landslide disaster early warning method and system, and relates to the technical field of open-pit mine safety monitoring. The method comprises the following specific steps: multi-source information collection and preprocessing: arranging microseismic and acoustic emission monitoring equipment in the potential landslide area to collect signals, collecting geological, topographic and meteorological data, and preprocessing the collected microseismic signals, acoustic emission signals and multi-source data; the invention can monitor the micro-fracture condition inside the landslide body in real time and in depth by combining microseismic and acoustic emission monitoring technology, and capture internal damage information that is difficult to obtain by traditional surface monitoring methods. At the same time, the invention can accurately extract early features of landslide disasters by analyzing and processing monitoring data using pattern recognition and machine learning methods, and realize timely early warning of landslide disasters.

[0007] However, the existing technology still does not realize the functions of synchronous acquisition of stress and acoustic emission signals, waveform recording, data storage, on-site analysis and data remote transmission, and cannot comprehensively record the complete process of rock mass under the action of ground stress, from elastic deformation to crack generation, to crack development, to rock burst, and cannot analyze the recorded acoustic emission waveform data in real time. SUMMARY

[0008] The technical problem to be solved by the present application is:

[0009] In view of the above problems, the present application provides an acoustic emission and stress homologous coupling acquisition terminal. The device is a mine impact dynamic disaster monitoring equipment that integrates the functions of synchronous acquisition of stress and acoustic emission signals, waveform recording, data storage, on-site analysis and data remote transmission. It can record the acoustic emission signal waveform released by the rock mass during the complete process of rock mass under the action of ground stress, from elastic deformation to crack generation, to crack development, to rock burst. It can analyze the recorded acoustic emission waveform data in real time, extract and summarize the characteristic parameters of the waveform data, analyze the dynamic disaster evolution process of the rock mass, and provide real-time early warning and guidance for the mine to make treatment plans.

[0010] The technical solution adopted by the present application to solve the above technical problems is: an acoustic emission and stress homologous coupling acquisition terminal, characterized in that the acquisition terminal is composed of three parts: an acquisition probe, a processing terminal and an upper computer.

[0011] The acquisition probe comprises a data acquisition module, an analog-digital conversion module and a controller module; the data acquisition module comprises an acoustic emission matrix, an inclination sensor and a stress sensor, the acoustic emission matrix comprises three groups of pickup sensors, and each two groups of pickup sensors are perpendicular to each other and arranged in a coordinate system, the acoustic emission measurement coordinate system formed by the pickup sensors coincides with the measurement coordinate system of the inclination sensor, and the stress measurement surface of the stress sensor is parallel to the upper bottom surface of the circular truncated cone shell; the analog-digital conversion module comprises an AD acquisition chip, an amplification circuit and a filter circuit to form a preliminary acquisition circuit for acquiring acoustic emission signals and stress data; the controller module comprises a controller unit MCU, a power module, a communication module and an RTC module;

[0012] The processing terminal comprises a data processing unit, a man-machine interaction module, a data storage module, a communication module, a power module and an RTC module; the man-machine interaction module comprises a TFT-LCD liquid crystal display, a state indicator light matrix and a function key matrix; the data storage module has two storage modes of a U disk and an SD card to realize real-time storage of original waveform data and analysis data; the communication module comprises a 485 bus, TCP communication and / or optical fiber communication to realize uplink and downlink communication with the acquisition probe and the host computer.

[0013] The host computer is used for receiving and processing data sent by the processing terminal, and is responsible for acoustic emission signal waveform spectrum feature extraction, result display, data storage and control of the processing terminal; the control of the processing terminal comprises data transmission format, acquisition cycle, alarm mode and threshold setting.

[0014] The beneficial technical effects of the present application are:

[0015] The present application aims to construct a comprehensive mine impact dynamic disaster monitoring system, and realizes core functions of synchronous acquisition of stress and acoustic emission signals, waveform recording, data storage, on-site analysis and data remote transmission through multi-sensor data acquisition technology, records acoustic emission signal waveforms released by rock mass in a series of complete processes from elastic deformation, to crack generation, to crack development and to rock burst under the action of ground stress, and analyzes and extracts feature parameters of the recorded acoustic emission waveform data in real time.

[0016] The application comprises three core components of a collection probe, a processing terminal and an upper computer, constructs a disaster detection data collection module taking an acoustic emission matrix, an inclination sensor and a stress sensor as cores, carries out data cleaning by using a wavelet threshold denoising method, effectively processes time series data containing noise, and ensures the accuracy and effectiveness of disaster data analysis. In the field of time domain analysis of acoustic emission signals, the system introduces a signal waveform simplification waveform feature extraction technology, accurately depicts the occurrence and development process of disasters by extracting and summarizing multiple simplified waveform feature parameters, and combines a frequency spectrum feature extraction algorithm combining a fast Fourier transform method and a Wigner-Ville distribution (WVD) as a supplementary method of the simplified waveform feature parameter analysis method, which significantly improves the recognition accuracy and efficiency of acoustic sources.

[0017] The technical scheme provided by the application realizes a mine impact dynamic disaster monitoring device integrating functions of stress and acoustic emission signal synchronous collection, waveform recording, data storage, on-site analysis and data remote transmission, which can record acoustic emission signal waveforms released by a rock mass in a complete process from elastic deformation, to crack generation, to crack development, to rock burst under the action of ground stress; real-time analysis of recorded acoustic emission waveform data, extraction and summarization of feature parameters of the waveform data, analysis of rock mass dynamic disaster evolution process, real-time early warning and guidance for the mine to make a processing plan. The application effectively avoids accidents by monitoring the impact dynamic disaster of the mine, and provides technical support for eliminating coal mine safety hazards. The device can effectively predict and analyze the impact dynamic disaster in mining engineering without contacting the mine, can comprehensively evaluate the stability of underground roadway surrounding rock at any time, and provides technical support for safe and efficient production of the mine. Through the acoustic emission technology, the stability of a slope, the safety of a mountain rock stratum and the overall safety level of a mining area can be evaluated in real time and dynamically, and a solid defense line is built for coal mine safety production.

[0018] The applicant has been committed to the research and application of a mine impact dynamic disaster monitoring and early warning device (system) for many years, and has cooperated with others to develop a mine single-channel acoustic emission collection extension, a mine multi-channel acoustic emission collection extension and a mine acoustic emission and stress same-source coupling collection and analysis terminal on acoustic emission products. In order to optimize the performance of the products, the applicant now further upgrades the mine acoustic emission and stress same-source coupling collection and analysis terminal. BRIEF DESCRIPTION OF DRAWINGS

[0019] Fig. 1 is a whole structure schematic view of the acoustic emission and stress same-source coupling collection terminal of the application;

[0020] Fig. 2 is a collection probe structure schematic view of the acoustic emission and stress same-source coupling collection terminal of the application;

[0021] Fig. 3 This is a structural block diagram of the acoustic emission and stress co-source coupling acquisition terminal of the present invention;

[0022] Fig. 4 This is a schematic diagram illustrating the definition of simplified waveform parameters for the acoustic emission signal of this invention;

[0023] Fig. 5 This is a block diagram of the display interface of the acoustic emission and stress co-source coupling acquisition terminal of the present invention. Detailed Implementation

[0024] Combined with appendix Figs. 1 to 5 The acoustic emission and stress co-source coupling acquisition terminal described in this invention, referred to as the acquisition terminal, includes three parts: an acquisition probe, a processing terminal, and a host computer.

[0025] The acquisition probe includes a data acquisition module, an analog-to-digital conversion module, and a controller module (its housing can be a combination of a flat-circular cone and a cylinder, with longitudinal patterns on the cylinder for easy installation). The data acquisition module consists of an acoustic emission matrix, a tilt sensor, and a stress sensor. The acoustic emission matrix contains three sets of pickup sensors, with each pair of pickup sensors perpendicular to each other and arranged in a coordinate system. The resulting acoustic emission measurement coordinate system coincides with the measurement coordinate system of the tilt sensor. The stress sensor's force-bearing surface is parallel to the top surface of the conical frustum of the housing. The analog-to-digital conversion module, through an AD acquisition chip, an amplifier circuit, and a filter circuit, forms a four-channel acquisition circuit, three channels for acquiring acoustic emission signals and one channel for acquiring stress data. The controller module consists of a controller unit (MCU), a power supply module, a communication module, and an RTC module.

[0026] The processing terminal includes a data processing unit, a human-machine interaction module, a data storage module, a communication module, a power supply module, and an RTC module. The human-machine interaction module consists of a TFT-LCD liquid crystal display screen, a status indicator matrix, and a function button matrix. The data storage module supports both USB flash drive and SD card storage, enabling real-time storage of raw waveform data and analyzed data. The communication module includes three communication methods: 485 bus, TCP communication, and fiber optic communication, enabling uplink and downlink communication with the acquisition probe and the host computer.

[0027] The host computer is data processing software responsible for data processing, display, storage, and control of the processing terminal (including data transmission format, acquisition cycle, alarm mode, and threshold settings).

[0028] Furthermore, the acoustic emission matrix in the data acquisition module of the acquisition probe contains three sets of sound pickup sensors, which constitute a set of acoustic emission signal acquisition coordinate system. That is, three sets of sound pickup sensors are arranged inside the acquisition probe, and each pair of sound pickup sensors is perpendicular to each other.

[0029] The pickup sensor selects a piezoelectric ceramic type sensor with a center frequency of 20KHz and a sensitivity of not less than 80dB (0dB=1V / m / s) as a pickup unit, and collects acoustic emission signals with center frequencies of 1.5KHz and 20KHz. The processing terminal extracts features from each frame of the acoustic emission raw waveform data, and obtains the required acoustic emission feature parameters. The calculation and acquisition of the acoustic emission feature parameters are completed by the data processing unit 21, and then the local storage is done according to the acoustic emission event trigger time, and the parameters to be uploaded are selected according to certain configuration and sent to the upper computer device.

[0030] The acoustic emission feature parameters and their definitions are as follows:

[0031] (1) Amplitude, the maximum amplitude of the signal waveform in a complete acoustic emission event, usually expressed in dB, the amplitude is 0dB when the sensor output is 1uV, and the conversion formula is:

[0032]

[0033] Wherein, U represents the maximum voltage value of the signal, dB represents the amplitude of the signal, and lg represents the logarithmic function with base 10. The amplitude parameter is independent of the threshold value, and is often used as a criterion for the intensity of the acoustic emission source.

[0034] (2) Energy, the energy of a single complete acoustic emission event is quantified by calculating the area under the envelope curve of the signal waveform; it is not very sensitive to the threshold value and the propagation characteristics of the acoustic emission wave, and can be used to identify the type of wave source and evaluate the activity of the acoustic emission source.

[0035] (3) Ring count, when the acoustic emission signal voltage value exceeds the threshold value, a pulse waveform is generated, and each pulse wave exceeding the threshold value is counted as one ring count; the ring count value is related to the threshold value, and is easily affected by the sensitivity of the acoustic emission system. Ring count is suitable for continuous and burst acoustic emission signal analysis, and represents the frequency and intensity of acoustic emission signal occurrence, and is widely used to evaluate the frequency of acoustic emission activity.

[0036] (4) Event count, a group of pulse decay waves exceeding the threshold value is called an acoustic emission event, and the event count can be represented by the counting rate or the total count, reflecting the frequency and intensity of acoustic emission events, and used for positioning and activity evaluation of the acoustic emission source. Similarly, the event count is affected by the setting of the threshold value.

[0037] (5) Rise time, defined as the time from the first time the signal waveform exceeds the threshold value to the maximum amplitude, usually in us, and the rise time of the acoustic emission signal is usually 10 -8 ~10-4 s. Rise time is often used in the identification and filtering of electromechanical noise in acoustic emission.

[0038] (6) Duration, defined as the time experienced from the first time the signal waveform crosses the threshold value to the final time it falls below the threshold value, which is affected by the setting of the threshold value.

[0039] (7) Effective voltage, which is the root mean square value of the signal level in the sampling time, that is:

[0040]

[0041] Where N is the number of sampling points, the effective voltage is related to the size of the acoustic emission signal voltage value, and is not affected by the threshold voltage value, and is often used for the activity evaluation of continuous acoustic emission signals.

[0042] (8) Average frequency, which is defined as the ratio of the ringing frequency to the acoustic emission duration, that is, average frequency = ringing frequency / acoustic emission duration; the greater this ratio, the more frequent the ringing event occurs during the acoustic emission event duration, which may mean that the change of microstructure is more intense.

[0043] (9) Peak frequency, in the spectral feature analysis of acoustic emission signals, the peak frequency refers to the frequency point at which the signal energy is most concentrated, that is, the frequency component with the largest amplitude in the fast Fourier transform (FFT) operation result. This frequency component often represents the main feature of the acoustic emission signal in the frequency domain.

[0044] The feature extraction of the acoustic emission signal, the commonly used methods mainly have the simplified waveform feature extraction and the signal waveform spectrum feature extraction, the present application combines the two methods to achieve the purpose of better identifying the sound source. The acoustic emission signal simplified waveform feature extraction mainly analyzes the acoustic emission signal from the time domain, and represents the acoustic emission signal by a plurality of simplified waveform feature parameters, such as amplitude, energy, ringing count, event count, rise time, duration, effective voltage, and then analyzes and statistically processes these feature parameters. This method is affected by the parameter setting of the signal acquisition device, and the feature description of the acoustic emission signal is not very perfect, and the waveform spectrum analysis method analyzes the essential characteristics of the sound source from the frequency domain, such as average frequency and peak frequency, as a complementary method to the simplified waveform feature parameter analysis method, which provides more information about the acoustic emission source in the frequency domain.

[0045] Further, the inclination sensor in the data acquisition module of the acquisition probe is perpendicular to the stress sensor force surface in the acquisition probe and parallel or perpendicular to the acoustic emission sensor group, that is, the xyz axes of the measurement coordinate system coincide with the measurement coordinate system of the acoustic emission matrix, and the inclination measurement range is 0-360°. The role of the inclination sensor is to facilitate the on-site installer to know the orientation of the sensor probe installation hole during on-site installation, and to facilitate the use of the acoustic emission sensor to determine the specific position of the acoustic emission source.

[0046] The installation of the acquisition probe first needs to connect the acquisition probe with the processing terminal, adjust to the installation mode, and according to the inclination and stress data fed back by the processing terminal, match the installation tool to ensure that the probe can be stably placed at the bottom of the hole and seamlessly fit. Then modulate the cement mortar into the hole, so that the acquisition probe is completely covered, and after a period of time, when the cement mortar gradually solidifies, the installation tool is taken out, and the hole is filled with cement to ensure the stability of the probe. When selecting the installation position of the acquisition probe, the geological conditions of the mine should be fully considered. In addition, the vertical depth of each drill hole is at least 1 m to ensure more effective pressure relief effect and effectively collect high-quality acoustic emission data.

[0047] Further, the stress data acquisition should be synchronized with the acoustic emission data, that is, the stress data should be collected at the same time when the acoustic emission event occurs, and the stress measurement range is 0-8 MPa; in addition, the stress data also needs to be collected at regular intervals when there is no acoustic emission event, and the stress data is stored and uploaded.

[0048] The acquisition of the acoustic emission signal should set a reasonable acquisition trigger threshold (trigger threshold, i.e. threshold value) to shield background noise, on-site noise and other interference, and only record valid acoustic emission events. For valid acoustic emission events, record the complete waveform data and trigger time of acoustic emission and stress wave, and also need to filter the recorded waveform to further reduce the influence of environmental clutter.

[0049] Further, the processing terminal communicates with the acquisition probe and the upper computer through the communication module, and the data transmission between the acquisition probe and the processing terminal is completed through the 485 bus, while the data transmission between the processing terminal and the upper computer integrates three communication modes, which are: 485 bus, TCP communication and optical fiber communication.

[0050] The three communication modes between the processing terminal and the upper computer are as follows:

[0051] (1) 485 bus communication mode, in this communication mode, the upper computer is the host computer, and the processing terminal is the slave computer, the baud rate can be configured as 1200-115200 (the specific baud rate can be configured in the processing terminal setting menu), and the data format is N,8,1;

[0052] (2) TCP communication mode, in which the network interface can automatically identify and adapt to the rate of the remote network device, and automatically work at the rate of 10Mbps, 100Mbps or 1000Mbps, and the mode can be compatible with network devices of different rates;

[0053] (3) Fiber communication mode, which follows the Ethernet protocol, is not affected by electromagnetic interference and radio frequency interference, and supports high-speed data transmission over long distances.

[0054] The processing terminal and the acquisition probe adopt a high-speed 485 communication mode, because the data amount transmitted in the acquisition probe is relatively large, the high-speed 485 communication mode is selected, and the communication cable between the two adopts a network cable with a shielding layer and an explosion-proof certificate, and the processing terminal also supplies power to the acquisition probe through the network cable.

[0055] The data transmission is controlled by a data processing unit, which needs to process the data packets transmitted by the acquisition probe, including extracting acoustic emission data, stress data and inclination data, and storing and backing up the original waveform data of acoustic emission, the characteristic parameter data after analysis, the stress data and the inclination data, and also needs to analyze the command data of the upper computer and upload the relevant data to the upper computer device according to the analysis result.

[0056] The data storage work is completed by a data storage module, which has two storage modes of U disk and SD card, and the U disk supports up to 128G (about 30G can store data continuously for one year), which is used for recording the original waveform of the overline trigger; the SD card is used for recording the characteristic parameters and characteristic waveforms. When the storage space is insufficient, the device will prompt and indicate the user through the screen liquid crystal and the indicator light; when the data storage exceeds the memory limit, the device will continue the storage work by cyclically covering the historical data; at the same time, the setting parameters of the processing terminal itself are stored and backed up to prevent loss after power failure or restart.

[0057] Further, the human-computer interaction function is realized by a human-computer interaction module, including liquid crystal display, indicator light indication, key operation, threshold alarm and other functions. The human-computer interaction function mainly includes two aspects of interaction, the first is screen data interaction, and the second is key operation interaction. The present application selects a 4.3-inch TFT-LCD liquid crystal display screen as the display screen, and the display main interface includes a standby interface and a parameter setting interface. The standby interface is used to display the processing terminal number, version number, acoustic emission waveform, inclination value, stress value, alarm value, alarm state and time information; the parameter setting interface needs to be entered through the operation of the keys, and includes a debugging function menu, an installation mode menu and a parameter setting menu.

[0058] The debugging function menu requires a password to access and includes sub-menus for acoustic emission debugging mode, stress debugging mode, and tilt angle debugging mode. After selecting the appropriate debugging mode through human-computer interaction, the acquisition terminal performs sensor calibration. After completing the data calibration, the acquired data is displayed cyclically and stored, while the raw data information is simultaneously displayed on the standby screen.

[0059] The installation mode menu can be accessed by pressing the corresponding function key on the standby screen without entering a password. Under this menu, the tilt angle data and stress data of the acquisition probe are displayed in real time, which allows on-site installers to know the orientation and position of the acquisition probe in the mounting hole based on the feedback data, thereby ensuring the correct installation of the acquisition probe.

[0060] The parameter settings menu can be accessed by pressing the corresponding function key from the standby screen, but a password is required before entry. The settings menu items include the following:

[0061] (1) System settings menu. Under this menu, you can perform operations such as setting the processing terminal number and setting the time.

[0062] (2) Communication Settings Menu: Under this menu, you need to set the uplink and downlink 485 speeds according to the transmission speeds of the acquisition probe and the host computer.

[0063] The baud rate required for communication, the communication method with the host computer, and the data format to be uploaded;

[0064] (3) Trigger Settings Menu: Under this menu, users can configure the trigger settings of the acquisition probe according to the monitoring site conditions.

[0065] A threshold is set to filter out background noise, ambient noise, and other interference, recording only effective acoustic emission events.

[0066] (4) Over-limit setting menu: Under this menu, the device's alarm function can be enabled. If the collected data exceeds the warning value, then...

[0067] The alarm is triggered using a dual alarm notification method, employing both LCD and indicator light prompts.

[0068] (5) Storage settings menu: This menu allows you to view historical data and also includes a storage reset function to clear the data.

[0069] Process the data stored on the terminal.

[0070] Further, the acquisition probe and the hardware system of the processing terminal are both equipped with RTC modules. The modules provide a unified time reference for the acquisition probe and the processing terminal, ensuring that the two have consistent time labels when collecting data. Even in the case of main power off or system accidental restart, the accuracy of the system time can be maintained, thereby ensuring that the data obtained by each acquisition point has a correct time stamp, facilitating subsequent data analysis and processing work, and providing accurate historical data query function.

[0071] Further, the filtering processing is a wavelet threshold denoising method. Using the wavelet threshold denoising method to optimize the monitoring data can effectively remove noise and extract useful information in the signal. The steps of the wavelet threshold denoising method are as follows:

[0072] (1) Wavelet analysis. Receive the original monitoring data, select a suitable wavelet basis function and decomposition layer number to perform wavelet decomposition on the actual vibration signal, and obtain wavelet coefficients of different scales;

[0073] (2) Threshold processing. After decomposition, select a suitable threshold and use a threshold function to quantize the wavelet coefficients of each layer. If the value is greater than the threshold, it is retained; if the value is less than the threshold, it is discarded, so as to suppress the interference noise;

[0074] (3) Signal reconstruction. The processed wavelet coefficients are reconstructed by inverse wavelet transform to obtain the useful signal, i.e. the denoised signal;

[0075] The selection of a suitable wavelet basis function generally considers the vanishing moment, symmetry, regularity and similarity. Expanding the signal with different wavelet basis functions will obtain different time-frequency characteristics. For the instantaneity and burstiness of acoustic emission signals, the Symlets wavelet basis function with vanishing moment characteristics is selected, and its formula is as follows:

[0076]

[0077] Wherein, u(t) is the unit function.

[0078] The selection of a suitable layer number can realize the separation of noise and useful signal. The greater the decomposition layer number, the more obvious the characteristics of the signal and the useful signal, which is more conducive to the separation of the two. The greater the decomposition layer number, the more obvious the difference between signal and noise, which is more conducive to signal and noise decomposition. However, the reconstructed signal will also be more distorted, affecting the denoising effect. The frequency range of wavelet decomposition is related to the sampling frequency. When the decomposition layer number is N, the frequency range is:

[0079] F S / 2 N+1 =f max / 2 N (4)

[0080] Wherein, F S is the sampling frequency, f max is the maximum signal frequency.

[0081] The wavelet decomposition is the process of wavelet change, that is, the original signal is decomposed into signals of different frequency bands, and the analysis of signals of different frequencies at different times can be realized.

[0082]

[0083] Wherein, is a wavelet function, and the wavelet function in the application is a Symlets wavelet; x (τ) is an original signal.

[0084] The corresponding wavelet inverse transform is:

[0085]

[0086] Wherein, C ψ is a constant of the wavelet function.

[0087] The threshold processing, after wavelet transform, the wavelet coefficient of the useful signal in the wavelet domain is greater than the wavelet coefficient of the noise signal, and the noise can be set to 0 by selecting the threshold value, but if the threshold value is too high, part of the useful signal will be deleted, and if the threshold value is too low, the noise processing is not comprehensive. Therefore, a threshold value processing with self-adaptive adjustment performance should be selected for the noise leakage acoustic emission signal. The threshold value selected in the application is:

[0088]

[0089] Wherein, σ is the standard deviation of the noise, J is the number of decomposition layers, N is the signal length, and ln represents the logarithmic function with e as the base.

[0090] The commonly used threshold function is divided into soft threshold and hard threshold, the soft threshold denoising is to reduce the wavelet coefficient whose absolute value is greater than the "threshold" threshold by the threshold value, and the wavelet coefficient whose absolute value is less than the "threshold" threshold is set to 0, as shown in formula (8); the hard threshold denoising is to retain the wavelet coefficient whose absolute value is greater than the "threshold" threshold, and the wavelet coefficient whose absolute value is less than the "threshold" threshold is set to 0, as shown in formula (9).

[0091]

[0092] Wherein, is the wavelet coefficient after threshold function processing, ω j,k is the original wavelet coefficient, and λ is the selected threshold value.

[0093] Further, the Fourier transform method and Wigner-Ville distribution (WVD) are combined to extract the waveform spectrum features of the acoustic emission signals. The Fourier transform method is used to analyze the spectrum of the acoustic emission signals, and the smooth pseudo-Wigner-Ville distribution method is used to observe the frequency component changes of the acoustic emission signals with time.

[0094] The Fourier transform is an effective mathematical tool for analyzing the spectrum features of signals. In the frequency domain analysis of signals, the frequency is taken as the independent variable, the signal x(t) is taken as the function of frequency f, x(f), and the graph plotted with the frequency f as the horizontal coordinate and the amplitude as the vertical coordinate is called the amplitude spectrum of the signal. The graph plotted with the frequency f as the horizontal coordinate and the phase as the vertical coordinate is called the phase spectrum of the signal. The acoustic emission signals have the characteristics of transience and randomness, and cannot be expressed as the sum of several cosine functions according to the Fourier series of the periodic signals. The Fourier transform is needed to realize the transformation from the time domain to the frequency domain. The expression of the Fourier transform is:

[0095]

[0096] or

[0097]

[0098] The inverse Fourier transform thereof is:

[0099]

[0100] As can be seen from equation (12), X(ω)e jωt dω is an infinitesimal quantity, which represents that the amplitude of the harmonic component of the signal x(t) at the angular frequency ω approaches zero. Only in a part of the frequency range (frequency band) the harmonic component has a non-zero value, i.e. the amplitude value is used after the integration of ω in the frequency range, which is called the spectral density of the signal x(t) and is also commonly called the spectrum. The spectrum analysis of x(t) is the process of obtaining the spectrum function and spectrum graph of the signal x(t).

[0101] The Wigner-Ville distribution W x (t,f) is a quadratic time-frequency representation with energy, which satisfies the time-frequency edge property and time and shift invariance. However, the Wigner-Ville distribution may have negative values. In order to associate the W x (t,f) with the actual physical performance, the W x (t,f) can be considered as the spectrum window flowing through the time interval A measure of the energy of a signal. This method can effectively detect the start and end time and frequency information of the mutation and non-stationary disturbance signal.

[0102] The Wigner-Ville distribution of the actual signal s(t) is defined as:

[0103]

[0104] Or

[0105]

[0106] Wherein x(t) is the analytic signal of the actual signal s(t). The analytic signal x(t) of the actual signal s(t) is obtained by taking the actual signal s(t) as the real part and taking the sequence obtained by Hilbert transforming the signal s(t) as the imaginary part.

[0107] W x The total integral of (t, f) with respect to time t and frequency f on the t-f plane is equal to the energy E of the signal, that is

[0108]

[0109] From equation (11), it can be seen that the WVD is the energy distribution of the signal on the t-f plane.

[0110] From equations (9) and (10), it can be seen that the WVD is bilinear, and therefore, it has cross terms. Assuming that x(t) is the sum of n x k (t) components:

[0111]

[0112] According to the definition, it can be seen that:

[0113]

[0114] The cross term is a false spectral distribution, which affects the physical interpretation of the WVD.

[0115] The smoothed pseudo Wigner-Ville distribution simultaneously windows the time domain variable and the frequency domain variable to reduce the cross term. Its definition is:

[0116]

[0117] Here h and g are two real symmetric windows, and h(0)=g(0)=1, the scales of the time domain smoothing and the frequency domain smoothing are easily controlled, and the length of the window function h(t) and g(t) can be independently selected. According to the W xThe law of (t, f) changing with the change of time domain variable and frequency domain variable, time-frequency analysis of acoustic emission signals is carried out.

[0118] Embodiment

[0119] The acoustic emission and stress homologous coupling acquisition terminal comprises an acquisition probe 1, a processing terminal 2 and an upper computer.

[0120] The acquisition probe 1 comprises a data acquisition module 11, an analog-digital conversion module 12 and a controller module 13, and the shell thereof is in the form of a combination of a truncated right circular cone and a cylinder, and the column body has longitudinal patterns, facilitating installation. The data acquisition module 11 is composed of an acoustic emission matrix 111, an inclination sensor 112 and a stress sensor 113, and the acoustic emission matrix 111 comprises three groups of pickup sensors. The analog-digital conversion module 12 comprises four acquisition circuits formed by an AD acquisition chip, an amplification circuit and a filter circuit, wherein three circuits are used for acquiring acoustic emission signals, and one circuit is used for acquiring stress data. The controller module 13 is composed of a controller unit (MCU) 131, a power module 132, a communication module 133 and an RTC module 134.

[0121] The three groups of pickup sensors in the acoustic emission matrix 111 form an acoustic emission signal acquisition coordinate system, that is, the three groups of pickup sensors are arranged in the acquisition probe 1, and each two groups of pickup sensors are perpendicular to each other, the formed acoustic emission measurement coordinate system coincides with the measurement coordinate system of the inclination sensor 112, and the measurement force surface of the stress sensor 113 is parallel to the upper bottom surface of the circular truncated cone shell. The pickup sensors select piezoelectric ceramic type sensors with a center frequency of 20 KHz and a sensitivity of not less than 80 dB (0 dB = 1 V / m / s) as pickup units, and acoustic emission signals are collected at center frequencies of 1.5 KHz and 20 KHz.

[0122] The man-machine interaction module 22 is composed of a TFT-LCD liquid crystal display 221, a state indicating lamp matrix 222 and a function button matrix 223, and comprises liquid crystal display, indicating lamp indication, button operation and threshold alarm functions. The man-machine interaction function mainly comprises two aspects of interaction, that is, screen data interaction and button operation interaction. The display main interface comprises a standby interface and a parameter setting interface, the standby interface is used for displaying the processing terminal number, version number, acoustic emission waveform, inclination value, stress value, alarm value, alarm state and time information, and the parameter setting interface needs to be entered by operating the buttons and comprises a debugging function menu, an installation mode menu and a parameter setting menu.

[0123] The data storage work is completed by a data storage module 23, which has two storage modes of U disk and SD card, the U disk supports 128G (about 30G can store data continuously for one year) at most, and is used for recording the original waveform of the trigger overline; the SD card is used for recording the characteristic parameters and characteristic waveform. When the storage space is insufficient, the device will prompt and indicate the user through the screen liquid crystal and the indicator light.

[0124] The processing terminal 2 communicates with the acquisition probe 1 and the upper computer through the communication module 23, the data transmission between the acquisition probe 1 and the processing terminal 2 is completed through the 485 bus, and the data transmission between the processing terminal 2 and the upper computer is integrated with three communication modes, which are 485 bus, TCP communication and optical fiber communication.

[0125] The power supply system of the application is composed of the power module 132 of the acquisition probe and the power module of the processing terminal, wherein the acquisition probe is powered by the processing terminal. In view of the particularity and complexity of the mine environment, the application strictly follows the standard requirements of the intrinsically safe circuit at the beginning of the design, and ensures that all electrical parameters can meet the safe use under the harsh conditions of the mine. The same as the specific embodiment.

[0126] The above is only the preferred specific embodiment of the application, but the protection scope of the application is not limited to this, any person skilled in the art can easily think of the changes or replacements within the technical range disclosed by the application, which should be covered in the protection scope of the application.

Claims

1. A terminal for acquiring acoustic emission and stress co-source coupling, characterized in that, The acquisition terminal consists of three parts: an acquisition probe (1), a processing terminal (2), and a host computer. The acquisition probe (1) includes a data acquisition module (11), an analog-to-digital conversion module (12), and a controller module (13). The data acquisition module (11) includes an acoustic emission matrix (111), an tilt sensor (112), and a stress sensor (113). The acoustic emission matrix (111) contains three sets of pickup sensors, and each pair of pickup sensors is perpendicular to each other and arranged in a coordinate system. The acoustic emission measurement coordinate system formed by these two sets coincides with the measurement coordinate system of the tilt sensor (112). The force measurement surface of the stress sensor (113) is parallel to the bottom surface of the circular platform of the shell. The analog-to-digital conversion module (12) is a preliminary acquisition circuit consisting of an AD acquisition chip, an amplification circuit, and a filtering circuit, used to acquire acoustic emission signals and stress data. The controller module (13) includes a controller unit MCU (131), a power module (132), a communication module (133), and an RTC module (134). The processing terminal (2) includes a data processing unit, a human-machine interaction module (22), a data storage module, a communication module (24), a power supply module, and an RTC module; wherein, the human-machine interaction module (22) includes a TFT-LCD liquid crystal display screen (221), a status indicator matrix (222), and a function key matrix (223); the data storage module has two storage methods, USB flash drive and SD card, to realize real-time storage of raw waveform data and parsed data; the communication module (24) includes 485 bus, TCP communication and / or fiber optic communication, used to realize uplink and downlink communication with the acquisition probe (1) and the host computer; The host computer is used to receive and process the data sent by the processing terminal (2). The host computer is responsible for extracting the waveform spectrum features of the acoustic emission signal, displaying the results, storing the data, and controlling the processing terminal. The control of the processing terminal includes data transmission format, acquisition cycle, alarm mode, and threshold setting.

2. The acoustic emission and stress co-source coupling acquisition terminal according to claim 1, characterized in that, The three sets of pickup sensors in the acoustic emission matrix (111) constitute a set of acoustic emission signal acquisition coordinate system, that is, three sets of pickup sensors are arranged in the acquisition probe (1), and each pair of pickup sensors is perpendicular to each other. The pickup sensor is a piezoelectric ceramic sensor with a center frequency of 20KHz and a sensitivity of not less than 80dB (0dB=1V / m / s) as the pickup unit, and the acoustic emission signal is collected with a center frequency of 1.5KHz and 20KHz; the processing terminal extracts features from each frame of the monitoring signal of the original acoustic emission waveform data to obtain the required acoustic emission feature parameters. The calculation and acquisition of acoustic emission characteristic parameters are completed by the data processing unit (21), and then the data is stored locally according to the acoustic emission event trigger time. The parameters to be uploaded are selected according to a certain configuration and sent to the host computer device. The acoustic emission characteristic parameters and their definitions are as follows: (1) Amplitude: The maximum amplitude of the signal waveform in a complete acoustic emission event, expressed in dB. The amplitude is 0 dB when the sensor outputs 1 μV. The conversion calculation formula is as follows: Where U represents the maximum voltage value of the signal, dB represents the amplitude of the signal, and lg represents the logarithmic function to base 10. The amplitude parameter is independent of the threshold value and is used to determine the intensity of the acoustic emission source. (2) Energy: The energy of a single complete acoustic emission event is quantified by calculating the area under the envelope of the signal waveform. It is not very sensitive to the threshold value and the propagation characteristics of the acoustic emission wave. It is used to identify the type of wave source and evaluate the activity of the acoustic emission source. (3) Ringing count: When the voltage value of the acoustic emission signal exceeds the threshold value, a pulse waveform is generated. Each pulse waveform that exceeds the threshold value is counted as a ringing count. The value of the ringing count is related to the threshold value and is easily affected by the sensitivity of the acoustic emission system. Ringing count is suitable for the analysis of continuous and burst acoustic emission signals and represents the frequency and intensity of the acoustic emission signal. (4) Event counting: A set of pulse decay waves exceeding the threshold value is called an acoustic emission event. Event counting can be expressed by count rate or total count, reflecting the frequency and intensity of acoustic emission events, and is used for the location and activity evaluation of acoustic emission sources. Similarly, event counting is affected by the threshold value setting. (5) Rise time, defined as the time it takes for the signal waveform to first cross the threshold value and reach its maximum amplitude, is usually measured in microseconds (µs). Typically, the rise time of an acoustic emission signal is around 10µs. -8 ~10 -4 us, rise time is used for electromechanical noise identification and filtering in acoustic emission; (6) Duration is defined as the time it takes for the signal waveform to first cross the threshold value and eventually fall below the threshold value. Its magnitude is affected by the threshold value setting. (7) RMS voltage, which is the root mean square value of the signal level during the sampling time, i.e.: Where N is the number of sampling points, the effective value voltage is related to the magnitude of the acoustic emission signal voltage value and is not affected by the threshold voltage value, and is used for the activity evaluation of continuous acoustic emission signals; (8) Average frequency, which is defined as the ratio of the number of rings to the duration of acoustic emission, i.e., average frequency = number of rings / duration of acoustic emission; the larger this ratio is, the more frequently the ringing events occur during the duration of the acoustic emission event, which means that the changes in the microstructure are more drastic. (9) Peak frequency: In the spectral characteristic analysis of acoustic emission signals, the peak frequency refers to the frequency point where the signal energy is most concentrated, which is the frequency component with the largest amplitude in the result of the Fast Fourier Transform (FFT) operation. The feature extraction of acoustic emission signals combines simplified waveform feature extraction with signal waveform spectrum feature extraction.

3. The acoustic emission and stress co-source coupling acquisition terminal according to claim 2, characterized in that, The processing terminal (2) communicates with the acquisition probe (1) and the host computer through the communication module (23). Data transmission between the terminal (2) and the acquisition probe (1) is completed through the 485 bus. Data transmission between the terminal (2) and the host computer integrates the communication methods of 485 bus, TCP communication and fiber optic communication. The three communication methods between the processing terminal and the host computer are as follows: (1) 485 bus communication mode. In this communication mode, the host computer is the master and the processing terminal is the slave. The baud rate can be configured to 1200-115200 and the data format is N,8,1. (2) TCP communication mode. In this communication mode, the network interface can automatically identify and adapt to the speed of the remote network device and automatically work at a speed of 10Mbps, 100Mbps or 1000Mbps. This mode is compatible with network devices of different speeds. (3) Fiber optic communication method: Fiber optic communication follows the Ethernet protocol, is not affected by electromagnetic interference and radio frequency interference, and supports high-speed data transmission over long distances. The processing terminal and the acquisition probe use high-speed 485 communication. Because the amount of data transmitted in the acquisition probe is relatively large, high-speed 485 communication is selected. The communication cable between the two uses a shielded network cable with an explosion-proof certificate, and the processing terminal also needs to supply power to the acquisition probe through this network cable. The data transmission is controlled by the data processing unit (21). The data processing unit needs to process the data packets transmitted from the acquisition probe, including the extraction of acoustic emission data, stress data, and tilt angle data, as well as the storage and backup of the original waveform data of acoustic emission, the analyzed characteristic parameter data, stress data, and tilt angle data. It also needs to parse the command data of the host computer and package and upload the relevant data to the host computer device according to the parsing results. The data storage is performed by a data storage module (23), which has two storage modes: USB flash drive and SD card. The USB flash drive supports up to 128G and is used to record the original waveform that triggers the line crossing. The SD card is used to record characteristic parameters and characteristic waveforms. When the storage space is insufficient, the device will alert the user through the LCD screen and indicator lights. When the data storage exceeds the memory limit, the storage work will continue by overwriting the historical data in a loop. At the same time, the settings parameters of the processing terminal itself are stored and backed up to prevent loss after power failure or restart.

4. The acoustic emission and stress co-source coupling acquisition terminal according to claim 3, characterized in that, The human-computer interaction function is realized through the human-computer interaction module (22), which includes LCD display, indicator light indication, button operation, and threshold alarm function. The human-computer interaction function mainly includes two aspects of interaction: the first is screen data interaction, and the second is button operation interaction. A 4.3-inch TFT-LCD liquid crystal display screen is selected as the display screen. The main display interface includes a standby interface and a parameter setting interface. The standby interface is used to display the processing terminal number, version number, acoustic emission waveform, tilt angle value, stress value, alarm value, alarm status, and time information. The parameter setting interface needs to be entered through the operation button and includes the debugging function menu, the installation mode menu, and the parameter setting menu. The debugging function menu requires a password to enter. Its sub-menus include: acoustic emission debugging mode, stress debugging mode, and tilt angle debugging mode. After selecting the corresponding debugging mode through human-computer interaction, the acquisition terminal performs sensor calibration. After completing the data calibration, the acquired data is displayed in a loop and stored. At the same time, the acquired raw data information is displayed synchronously on the standby screen. The installation mode menu can be accessed by pressing the corresponding function key on the standby screen without entering a password. Under this menu, the tilt angle data and stress data of the acquisition probe are displayed in real time, which makes it convenient for on-site installers to know the orientation and position of the acquisition probe in the mounting hole based on the feedback data, thereby ensuring the correct installation of the acquisition probe. The parameter settings menu can be accessed by pressing the corresponding function key from the standby screen, but a password is required before entry. The settings menu items include the following: (1) System settings menu. Under this menu, you can perform operations such as setting the processing terminal number and setting the time. (2) Communication settings menu. Under this menu, you need to set the baud rate value required for uplink and downlink 485 communication according to the transmission speed of the acquisition probe and the host computer, as well as select the communication method with the host computer and the data format to be uploaded. (3) Trigger setting menu. Under this menu, users can set the trigger threshold of the acquisition probe according to the monitoring site conditions to shield background noise, site noise and other interference, and only record effective acoustic emission events. (4) Over-limit setting menu: Under this menu, the alarm function of the device can be enabled. If the collected data exceeds the warning value, an alarm will be triggered. The alarm is triggered by a dual alarm method using LCD display and indicator light. (5) Storage settings menu. Under this menu, you can view historical data and also have a storage reset function to clear the data stored on the processing terminal.

5. The acoustic emission and stress co-source coupling acquisition terminal according to claim 4, characterized in that, The tilt sensor (112) is perpendicular to the stress sensor's force-bearing surface inside the acquisition probe, and parallel or perpendicular to a set of acoustic emission sensors. That is, the xyz axes of the measurement coordinate system coincide with the measurement coordinate system of the acoustic emission matrix. The tilt measurement range is 0 to 360°. The tilt sensor serves two purposes: first, it helps on-site installers to know the orientation of the sensor probe's mounting hole during installation; second, it facilitates use with the acoustic emission sensor to determine the specific location of the acoustic emission source. The installation of the acquisition probe first requires connecting the acquisition probe to the processing terminal and adjusting it to the installation mode. Based on the tilt angle and stress data fed back by the processing terminal, and with the help of installation tools, ensure that the probe can be stably placed at the bottom of the hole and that it fits seamlessly. Then, prepare cement mortar and pour it into the hole to completely cover the acquisition probe. After letting it dry for a period of time, wait for the cement mortar to gradually solidify, remove the installation tools, and then fill the hole with cement to ensure the stability of the acquisition probe. When selecting the installation location of the acquisition probe, the geological conditions of the mine should be fully considered. Each borehole has a vertical depth of at least 1m to ensure more effective pressure relief and efficient collection of high-quality acoustic emission data.

6. The acoustic emission and stress co-source coupling acquisition terminal according to claim 5, characterized in that, Stress data acquisition should be synchronized with acoustic emission data acquisition. That is, stress data should be acquired simultaneously when an acoustic emission event occurs, with the stress measurement range being 0–8 MPa. When no acoustic emission event occurs, stress data should also be acquired periodically, and the stress data should be stored and uploaded. The acquisition of acoustic emission signals should be based on a reasonable acquisition trigger threshold (trigger threshold, i.e., threshold value) to shield background noise and on-site noise interference, and only record valid acoustic emission events. For valid acoustic emission events, the complete waveform data of acoustic emission and stress wave and the trigger time should be recorded. The recorded waveform also needs to be filtered by the filtering circuit to further reduce the influence of environmental noise.

7. The acoustic emission and stress co-source coupling acquisition terminal according to claim 6, characterized in that, The hardware systems of the acquisition probe (1) and the processing terminal (2) are both equipped with an RTC module. This module provides a unified time reference for the acquisition probe and the processing terminal, ensuring that both use consistent time stamps when acquiring data. Even if the main power is turned off or the system is restarted unexpectedly, the system time can be kept accurate, thereby ensuring that the data acquired by each acquisition point has the correct timestamp, which facilitates subsequent data analysis and processing, as well as providing accurate historical data query functions.

8. The acoustic emission and stress co-source coupling acquisition terminal according to claim 7, characterized in that, The filtering process is wavelet threshold denoising, which optimizes the monitoring data to effectively remove noise and extract useful information from the signal. The steps of wavelet threshold denoising are as follows: (1) Wavelet analysis: Receive the original monitoring data, select appropriate wavelet basis functions and decomposition levels to perform wavelet decomposition on the actual vibration signal, and obtain wavelet coefficients at different scales; (2) Threshold processing: After decomposition, an appropriate threshold is selected, and the wavelet coefficients of each layer are quantized by a threshold function. If the coefficients are greater than the threshold, they are retained, and if they are less than the threshold, they are discarded, so as to suppress interference noise. (3) Signal reconstruction: The processed wavelet coefficients are reconstructed by inverse wavelet transform to obtain the useful signal, i.e. the denoised signal; Choosing a suitable wavelet basis function generally requires comprehensive consideration of vanishing moment, symmetry, regularity, and similarity. Expanding the signal using different wavelet basis functions will yield different time-frequency characteristics. Considering the instantaneous and sudden nature of acoustic emission signals, the Symlets wavelet basis function, which has a vanishing moment characteristic, is selected. Its formula is as follows: Where u(t) is a unit function; Choosing an appropriate number of decomposition layers is crucial for separating noise from the useful signal. A higher decomposition layer number leads to more distinct characteristics of the signal and the useful signal, thus facilitating separation. A larger decomposition layer number also increases the difference between signal and noise, improving signal-to-noise decomposition. However, it also increases signal distortion during reconstruction, affecting denoising performance. The frequency range of wavelet decomposition is related to the sampling frequency. When the decomposition layer number is N, the frequency range is: F S / 2 N+1 =f max / 2 N (4) Among them, F S f is the sampling frequency. max Maximum signal frequency; The wavelet decomposition is a process of wavelet transformation, which decomposes the original signal into signals of different frequency bands. This enables the analysis of signals of different frequencies at different times. The wavelet function formula is as follows: frequency is located by parameter a, and time is located by parameter b: in, It is a wavelet function, specifically a Symlets wavelet; x(τ) is the original signal; Its corresponding inverse wavelet transform is: in, C ψ It is a constant of the wavelet function; The threshold processing, after wavelet transform, results in the wavelet coefficients of the useful signal being greater than those of the noise signal in the wavelet domain. Selecting a threshold can set the noise to zero; however, if the threshold is too high, it will delete some useful signals, and if the threshold is too low, the noise processing will be incomplete. Therefore, a threshold with adaptive adjustment capabilities should be selected to process the noisy leakage sound emission signal. The selected threshold is: Where σ is the noise standard deviation; J is the number of decomposition layers; N is the signal length; and ln represents the logarithmic function with base e. Commonly used threshold functions are divided into soft thresholding and hard thresholding. Soft thresholding denoising subtracts the threshold from the wavelet coefficients whose absolute value is greater than the threshold and sets the wavelet coefficients whose absolute value is less than the threshold to 0, as shown in formula (8). Hard thresholding denoising retains the wavelet coefficients whose absolute value is greater than the threshold and sets the wavelet coefficients whose absolute value is less than the threshold to 0, as shown in formula (9). in, The wavelet coefficients are after thresholding, ω j,k λ represents the original wavelet coefficients, and λ is the selected threshold.

9. The acoustic emission and stress co-location acquisition terminal according to claim 8, characterized in that, The spectral feature extraction of the acoustic emission signal waveform is accomplished by combining the Fast Fourier Transform (FFT) method with the Wigner-Ville distribution (WVD), specifically as follows: First, the fast Fourier transform method is used to perform spectral analysis of the acoustic emission signal to analyze the spectral components of the acoustic emission signal as a whole. Then, the smooth pseudo-Wigner-Ville distribution method is used to observe the changes in the frequency components of the acoustic emission signal over time. In frequency domain analysis of signals, frequency is taken as the independent variable, and the signal x(t) is considered as a function x(f) of frequency f. The graph plotted with frequency f as the abscissa and amplitude as the ordinate is called the amplitude spectrum of the signal, and the graph plotted with frequency f as the abscissa and phase as the ordinate is called the phase spectrum of the signal. Acoustic emission signals have transient and random characteristics, and cannot be represented by a sum of several sine (cosine) functions using Fourier series as a periodic signal. Therefore, Fourier transform is needed to transform the signal from the time domain to the frequency domain. The expression for the Fourier transform is: or Its inverse Fourier transform is: In equation (12), X(ω)e jωt dω indicates that the amplitude of the harmonic component of signal x(t) at angular frequency ω approaches zero. The harmonic component has a non-zero value only in a certain frequency range (band). That is, the amplitude value is taken after integrating ω in these frequency ranges. X(ω) is called the spectral density of signal x(t), and is also often called the spectrum. Spectral analysis of x(t) is the process of finding its spectral function and spectrum graph from signal x(t). The Wigner-Ville distribution W x (t,f) is a quadratic time-frequency representation with energy, satisfying time-frequency boundary conditions and time and shift invariance. It exhibits negative values ​​based on the Wigner-Ville distribution, which is used to represent W... x (t,f) is related to actual physical properties, and W is considered to be x (t,f) is the time interval Intra-flow through spectral window A measure of energy to detect the start and end times and frequency information of abrupt, non-stationary disturbance signals; The Wigner-Ville distribution of a real signal s(t) is defined as follows: or Where: x(t) is the analytic signal of the actual signal s(t); the analytic signal x(t) is obtained by taking the actual signal s(t) as the real part and the sequence obtained by performing the Hilbert transform on the signal s(t) as the imaginary part. W x The total integral of (t,f) with respect to time t and frequency f in the tf plane equals the energy E of the signal, i.e. From equation (11), we can see that WVD is the energy distribution of the signal on the tf plane; From equations (9) and (10), it can be seen that WVD is bilinear and has cross terms. Assume that x(t) is n x k (t) Sum of components: According to the definition: The interaction term is used to reduce the spurious spectral distribution that affects the physical analysis of WVD. A smoothed pseudo-Wigner-Ville distribution is used to window both time-domain and frequency-domain variables simultaneously to reduce the interaction term, which is defined as follows: In the formula, h and g are two real symmetric windows, and h(0) = g(0) = 1. The lengths of the window functions h(t) and g(t) are chosen independently to control the scale of time-domain smoothing and frequency-domain smoothing. According to W x The time-frequency analysis of acoustic emission signals is performed by analyzing the changes in (t,f) with respect to time-domain and frequency-domain variables.

Images