Accurate positioning method for sound emission source with complex structure in high-noise environment

Abstract

The invention belongs to the technical field of aviation strength tests, and particularly relates to a high-noise environment complex structure acoustic emission source accurate positioning method. The method comprises the following steps: step 1, constructing a time difference mapping library for positioning; 2, collecting an acoustic emission signal, and carrying out composite noise reduction on the acoustic emission signal; 3, processing the acoustic emission signals through a time window energy ratio method and an AIC method to obtain the arrival time of each acoustic emission signal; step 4, performing signal matching according to the similarity of the arrival time of each acoustic emission signal to obtain an event group for positioning; and 5, carrying out positioning calculation according to the time difference mapping library and the event group to obtain a damage signal source positioning coordinate. According to the invention, the interference of noise is effectively reduced, the detection precision of the sound wave arrival time is greatly improved, and the structure damage source can be positioned in a high-noise complex structure fatigue test.

Description

Technical Field

[0001] The present application belongs to the technical field of aviation strength testing, and in particular relates to a method for accurately locating an acoustic emission source of a complex structure in a high-noise environment. Background Art

[0002] Acoustic emission damage monitoring technology is widely used in damage monitoring / detection of structures or materials due to its advantages of good real-time performance, high sensitivity, and insensitivity to the shape of the object being detected. Acoustic emission damage monitoring technology can not only identify the acoustic emission signals released by the structure during the initiation and expansion of damage, but also locate the sound source through the sensor array. The damage localization function of acoustic emission technology can not only provide the location information of the damage, but also effectively filter out the noise interference outside the monitoring area through spatial filtering, effectively improving the reliability of damage monitoring.

[0003] At present, the most widely used and mature positioning technology in engineering is the time of arrival (TOA)-based positioning technology. This technology realizes the positioning of the acoustic emission source through the time difference and wave velocity parameters of the acoustic emission wave reaching different sensors in a positioning array. Its positioning accuracy depends on the accuracy and precision of the arrival time and wave velocity measurement. In actual detection, it is usually necessary to set the detection threshold voltage to filter out the noise with small amplitude and reduce the data volume of the monitoring signal. The time when the damage signal "crosses" the threshold for the first time is taken as the arrival time of the signal. This arrival time detection method can accurately locate the damage position in a detection environment with less noise. However, from its detection principle, it can be found that the time of "crossing" the threshold for the first time is later than the actual arrival time of the signal. Therefore, this method will inevitably introduce errors when calculating the positioning through the time difference. More seriously, when monitoring damage in an environment with very harsh noise, the noise signal presents a "continuous" signal form, and when the signal amplitude is close to the damage signal, the noise signal will frequently cross the threshold, resulting in a very large error in the measurement of the arrival time by "crossing the threshold". In addition, in the location of damage to complex structures or anisotropic materials, the assumptions of the wave propagation in a straight line and the constant wave speed in TOA positioning technology will not hold. For example, when there is a large opening in the wave propagation path, the wave will propagate along a curve to bypass the opening and eventually be detected by the sensor. Composite materials are anisotropic materials, and their wave speeds vary greatly in different directions. Therefore, using a fixed wave speed in positioning will also result in a large positioning error. In summary, the two parameters that traditional TOA positioning technology relies on, arrival time and wave speed, both have the problem of inaccuracy when facing complex structures in high-noise environments. In actual engineering applications, low positioning rates or large positioning errors usually occur. This seriously weakens the effectiveness of acoustic emission technology in positioning.

[0004] Therefore, it is desired to have a technical solution to overcome or at least alleviate at least one of the above-mentioned defects of the prior art. Summary of the invention

[0005] The purpose of this application is to provide a method for accurately locating acoustic emission sources of complex structures in high-noise environments, so as to solve the current problems of low localization rate and large localization error of damage sources of complex structures in high-noise environments.

[0006] The technical solution of this application is:

[0007] A method for accurately locating acoustic emission sources of complex structures in a high noise environment, comprising:

[0008] Step 1: Build a time difference mapping library for positioning;

[0009] Step 2: collecting acoustic emission signals and performing composite noise reduction on the acoustic emission signals;

[0010] Step 3: Processing the acoustic emission signal by using the time window energy ratio method and the AIC method to obtain the arrival time of each acoustic emission signal;

[0011] Step 4: perform signal matching based on the similarity of the arrival time of each acoustic emission signal to obtain an event group for positioning;

[0012] Step 5: Perform positioning calculation based on the time difference mapping library and the event group to obtain the positioning coordinates of the damage signal source.

[0013] In at least one embodiment of the present application, in step 1, constructing a time difference mapping library for positioning includes:

[0014] S101, determining a detection area and a sensor layout plan, and dividing the detection area into grids;

[0015] S102, using artificial lead breaking to simulate the acoustic emission source to excite each grid node, and collecting lead breaking waveform data at the same time;

[0016] S103, processing all lead-break waveform data to obtain the arrival time of each acoustic emission signal;

[0017] S104, calculating the arrival time difference of all sensor pairs on each grid node according to the arrival time, and constructing an initial time difference mapping library of each sensor pair on all grid nodes;

[0018] S105, eliminating abnormal data in the initial time difference mapping library based on the quartile method, and using a bilinear interpolation fitting algorithm to refine the data in the initial time difference mapping library and fill in missing data, so as to obtain a time difference mapping library for positioning.

[0019] In at least one embodiment of the present application, in step 2, performing composite noise reduction on the acoustic emission signal includes:

[0020] S201, constructing a deep residual neural network model, inputting the material damage acoustic emission signal and the environmental noise signal obtained through the experiment into the deep residual neural network model for training, and performing time domain waveform feature recognition and noise reduction on the acoustic emission signal through the trained deep residual neural network model;

[0021] S202: Perform frequency domain filtering and noise reduction on the acoustic emission signal using a Butterworth digital bandpass filter.

[0022] In at least one embodiment of the present application, in step three, the acoustic emission signal is processed by a time window energy ratio method and an AIC method to obtain the arrival time of each acoustic emission signal, including:

[0023] S301, processing the acoustic emission signal by a time window energy ratio method to obtain a damage characteristic time window position;

[0024] S302: Process the damage characteristic time window position by AIC method to obtain the arrival time of each acoustic emission signal.

[0025] In at least one embodiment of the present application, in S301, the acoustic emission signal is processed by a time window energy ratio method to obtain a damage feature time window position, including:

[0026] Assuming that L1 and L2 are two adjacent time windows, the time window energy ratio of the acoustic emission signal is calculated;

[0027]

[0028] Among them, x(t) is the time series of the acoustic emission signal, t is the sample point, T 1 is the starting point of the first time window, T 0 is the end point of the first time window, which is also the starting point of the second time window, T2 is the end point of the second time window, ω is the relative energy of the acoustic emission signal waveform, and α is the stability coefficient;

[0029] The damage feature time window position is obtained according to the maximum value of the time window energy ratio.

[0030] In at least one embodiment of the present application, in S302, the damage feature time window position is processed by the AIC method to obtain the arrival time of each acoustic emission signal, including:

[0031] Calculate the AIC value of the damage feature time window position:

[0032] AIC(t)=tlog 10 (var({x(1:t)})+(Tt-1)log 10 (var({x(t:T}))

[0033] Among them, var({x(1:t}) represents the variance of x, and T is the last sample point;

[0034] The arrival time of each acoustic emission signal is obtained according to the minimum value of the AIC value.

[0035] In at least one embodiment of the present application, step 4, performing signal matching according to the similarity of the arrival time of each acoustic emission signal to obtain an event group for positioning, includes:

[0036] S401, converting the order of magnitude of the arrival time of each acoustic emission signal to the 0.1 ms level;

[0037] S402, extracting the integer part of the arrival time of each acoustic emission signal as the matching time;

[0038] S403: Match the acoustic emission signals with the same matching time one by one, and output an event group for positioning.

[0039] In at least one embodiment of the present application, in step 5, performing positioning calculation according to the time difference mapping library and the event group to obtain the positioning coordinates of the damage signal source includes:

[0040] S501, calculating the time difference of the acoustic emission signal from the test point in the event group reaching each sensor pair

[0041] S502, calculate time difference The time difference of each grid node in the time difference mapping library with the same sensor pair The difference

[0042] S503, calculating the time difference weight f of the test point in each sensor pair K ;

[0043] S504: Fusion of the time difference of each sensor pair at the test point According to the time difference And the time difference weight f K The weighted time difference corresponding to each grid node is calculated, and the position coordinates of the grid node where the minimum weighted time difference is located are used as the damaged signal source positioning coordinates of the test point.

[0044] In at least one embodiment of the present application, the method further includes step six, graphically visualizing the structural damage location according to the damage signal source positioning coordinates, evaluating the structural damage state, and outputting the damage positioning result.

[0045] The invention has at least the following beneficial technical effects:

[0046] The method for accurately locating acoustic emission sources of complex structures in high-noise environments of the present application effectively reduces noise interference, greatly improves the detection accuracy of sound wave arrival time, and can locate structural damage sources in fatigue tests of complex structures in high-noise environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 This is a flow chart of a method for accurately locating acoustic emission sources of complex structures in a high-noise environment according to an embodiment of the present application;

[0048] Figure 2 This is a structural diagram of a deep residual neural network for identifying and denoising time-domain waveform features of acoustic emission signals according to an embodiment of the present application;

[0049] Figure 3 is a frequency domain bandpass filtering principle diagram of an implementation method of the present application;

[0050] Figure 4 This is a flow chart of the "two-step method" for picking up the arrival time of sound waves according to one embodiment of the present application;

[0051] Figure 5 This is a schematic diagram of the time window energy ratio arrival time picking principle of one embodiment of the present application;

[0052] Figure 6 This is a schematic diagram of the precise detection principle of the AIC value arrival time according to one implementation of the present application. DETAILED DESCRIPTION

[0053] In order to make the purpose, technical scheme and advantages of the implementation of this application clearer, the technical scheme in the embodiment of this application will be described in more detail below in conjunction with the drawings in the embodiment of this application. In the drawings, the same or similar reference numerals throughout represent the same or similar elements or elements with the same or similar functions. The described embodiments are part of the embodiments of this application, not all of them. The embodiments described below with reference to the drawings are exemplary and are intended to be used to explain this application, and should not be construed as limitations on this application. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application. The embodiments of this application are described in detail below in conjunction with the drawings.

[0054] In the description of the present application, it should be understood that the terms "center", "longitudinal", "lateral", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as limiting the scope of protection of the present application.

[0055] The following is combined with Figures 1 to 6 This application is described in further detail.

[0056] This application provides a method for accurately locating acoustic emission sources of complex structures in a high-noise environment. Figure 1 As shown, the following steps are included:

[0057] Step 1: Build a time difference mapping library for positioning;

[0058] Step 2: collecting acoustic emission signals and performing composite noise reduction on the acoustic emission signals;

[0059] Step 3: Process the acoustic emission signal by using the time window energy ratio method and the AIC method to obtain the arrival time of each acoustic emission signal;

[0060] Step 4: perform signal matching based on the similarity of the arrival time of each acoustic emission signal to obtain an event group for positioning;

[0061] Step 5: Perform positioning calculation based on the time difference mapping library and the event group to obtain the positioning coordinates of the damage signal source.

[0062] Among them, step one is the preparation work before the test, and steps two to five are the implementation phase of the acoustic emission damage signal source positioning test during the test.

[0063] The method for accurately locating acoustic emission sources of complex structures in a high-noise environment of the present application, in step 1, constructs a time difference mapping library for positioning, including:

[0064] S101, determining a detection area and a sensor layout plan, and dividing the detection area into grids;

[0065] Preferably, the sensor array should cover the detection area. When setting the grid size, the size of the detection area and the complexity of the structure should be comprehensively considered. The grid size is usually recommended to be 10mm×10mm or 20mm×20mm. The grid can be denser for more complex areas. The higher the resolution of the grid, the higher the positioning accuracy of the time difference mapping method, but the longer it takes, so comprehensive considerations should be taken when setting.

[0066] S102, using artificial lead breaking to simulate the acoustic emission source to excite each grid node, and collecting lead breaking waveform data at the same time;

[0067] Preferably, each grid node is stimulated 5-10 times to collect the lead-break waveform data of each channel of the positioning array;

[0068] S103, processing all lead-break waveform data to obtain the arrival time of each acoustic emission signal;

[0069] Preferably, all the lead-break waveform data are processed by the same "two-step method" as step three to obtain the precise arrival time of each acoustic emission signal;

[0070] S104, calculating the arrival time difference of all sensor pairs on each grid node according to the arrival time, and constructing an initial time difference mapping library of each sensor pair on all grid nodes;

[0071] A sensor pair refers to any two sensors in an array. For example, an array of four sensors can form six sensor pairs.

[0072] S105, eliminating abnormal data in the initial time difference mapping library based on the quartile method, and using a bilinear interpolation fitting algorithm to refine the data in the initial time difference mapping library and fill in missing data, so as to obtain a time difference mapping library for positioning.

[0073] Preferably, the quartile method (i.e., box plot statistics) can be used to eliminate abnormal values ​​(i.e., outliers) caused by human factors in the time difference data of each node. The bilinear interpolation algorithm is used to interpolate and refine the initial time difference mapping library, increase the density of the original time difference mapping library data, improve the positioning accuracy, and reduce the time cost of building the time difference mapping library. The advantage of bilinear interpolation is that it can more accurately enlarge or reduce the data matrix, and the calculation speed is faster. This method can also fill in the missing data in the time difference mapping library.

[0074] The precise positioning method of acoustic emission sources in complex structures in high-noise environments of the present application is to first perform noise reduction processing on the data collected by each channel of the positioning array in the formal positioning detection. The noise reduction method includes time domain waveform feature recognition noise reduction and frequency domain bandpass filtering noise reduction. In specific implementation, one noise reduction method can be used alone or two noise reduction methods can be used at the same time according to the noise reduction effect. When using them at the same time, the time domain waveform feature screening noise reduction is performed first and then the frequency domain filtering noise reduction is performed.

[0075] In a preferred embodiment of the present application, in step 2, composite noise reduction is performed on the acoustic emission signal, including:

[0076] S201, constructing a deep residual neural network model, inputting the material damage acoustic emission signal and the environmental noise signal obtained through the experiment into the deep residual neural network model for training, and performing time domain waveform feature recognition and noise reduction on the acoustic emission signal through the trained deep residual neural network model;

[0077] S202, performing frequency domain filtering and noise reduction on the acoustic emission signal through a Butterworth digital bandpass filter.

[0078] The method for accurately locating acoustic emission sources of complex structures in a high-noise environment of the present application aims to solve the problem that acoustic emission sensors are susceptible to environmental interference due to their high sensitivity. It proposes a composite noise reduction method for acoustic emission signals based on a combination of time domain characteristic waveform recognition and frequency domain filtering to solve the problem of removing frequency domain coherent and time domain correlated noise components in acoustic emission noise reduction.

[0079] a. Time domain characteristic waveform recognition of damage signal based on deep residual neural network

[0080] Design experiments to obtain and collect material damage signals and environmental noise signals, then design a deep residual neural network model, use the damage signal and noise signal time domain waveforms as input training models, and the trained model can automatically identify damage signals and noise signals. Figure 2 As shown in the figure, it consists of a convolutional layer (Conv1d), a pooling layer (Maxpool), a residual block (ResBlock) and a fully connected layer (Fully Connected). Finally, the trained deep residual neural network model is used to intelligently remove the acoustic emission noise signal, achieving the purpose of noise reduction from the time domain waveform feature recognition.

[0081] b. Frequency domain filtering

[0082] Frequency domain filtering is to selectively pass signals within a certain frequency range according to the frequency characteristics of the signal, and suppress signals in other frequency ranges. Since the Butterworth filter has a smooth frequency response curve, the passing signal maintains the original phase and amplitude characteristics, and does not introduce additional phase distortion or amplitude distortion. And it has a higher passband gain and a narrower bandwidth, which can maintain a higher signal-to-noise ratio and better frequency resolution after the signal passes. Therefore, in this method, the frequency domain filtering uses a Butterworth digital bandpass filter to perform frequency domain filtering and noise reduction on the acoustic emission signal. The filter consists of two digital filters: a low-pass filter and a high-pass filter. The low-pass filter is used to pass signals below a certain cutoff frequency, while the high-pass filter is used to pass signals above a certain cutoff frequency. The output signals of the two filters are added to form the output signal of a bandpass filter. The principle is as shown in the attached figure. Figure 3 shown.

[0083] The method for accurately locating the acoustic emission source of a complex structure in a high-noise environment of the present invention adopts a "two-step method" to extract the arrival time of the sound wave, processes the acoustic emission signal waveform data through the time window energy ratio method and the AIC method, and obtains the precise arrival time of all acoustic emission signals in each channel of the positioning array.

[0084] In a preferred embodiment of the present application, Figure 4 As shown, in step 3, the acoustic emission signal is processed by the time window energy ratio method and the AIC method to obtain the arrival time of each acoustic emission signal, including:

[0085] S301, processing the acoustic emission signal by a time window energy ratio method to obtain a damage characteristic time window position;

[0086] S302: Process the damage feature time window position by AIC method to obtain the arrival time of each acoustic emission signal.

[0087] The method for accurately locating the acoustic emission source of a complex structure in a high-noise environment of the present application aims to solve the problem of inaccurate extraction of the arrival time based on the "threshold" by the traditional method, and proposes an arrival time picking method based on the time window energy ratio to effectively identify the time period in which the damage feature in the acoustic emission signal is located, and combines the local AIC (Akaike Information Criterion) arrival time picking method to accurately extract the arrival time of the damage signal feature, so as to improve the accuracy of the extraction of the arrival time of the acoustic emission wave.

[0088] a. Time window energy ratio arrival time extraction

[0089] The time window energy ratio method uses two time windows of equal length to slide on the time axis and calculates the energy ratio of the two time windows. The principle is as shown in the attached Figure 5 L1 and L2 are the first time window and the second time window respectively. The energy ratio of the first time window and the second time window is shown in formula (1):

[0090]

[0091] Among them, x(t) is the time series of the acoustic emission signal, t is the sample point, T 1 is the starting point of the first time window, T 0 is the end point of the first time window, which is also the starting point of the second time window (assuming that L1 and L2 are two adjacent time windows), T2 is the end point of the second time window, ω is the relative energy of the acoustic emission signal waveform, and α is the stability coefficient, which is generally 0.5 to 2.0;

[0092]

[0093] The damage feature time window position is obtained according to the maximum value of the time window energy ratio.

[0094] b. Accurately extract arrival time based on AIC value

[0095] The signal arrival time detection principle based on AIC technology is shown in formula (2):

[0096] AIC(t)=tlog 10 (var({x(1:t)})+(Tt-1)log 10 (var({x(t:T})) (2)

[0097] Among them, var({x(1:t}) represents the variance of x, and T is the last sample point;

[0098] The arrival time of each acoustic emission signal is obtained according to the minimum value of the AIC value.

[0099] From formula (2), it can be seen that AIC divides the signal into two parts, x(1:t) and x(t:T), and calculates the similarity of the variance between the two parts of the signal. Through the above AIC calculation model, the AIC value can be calculated from the acoustic emission signal data. Since the statistical properties of the noise signal and the damage signal are quite different, the fit of the two signals is the worst under the minimum square error, and the corresponding AIC value is the smallest. The point with the smallest AIC value is the time point when the difference between the two parts of the signal is the largest, which is the starting point (arrival time) of the acoustic emission signal. An AIC minimum value will appear at each damage sample point in the acoustic emission signal. By calculating the time corresponding to these minimum sample points, the actual arrival time of the damage signal can be obtained. The principle of accurate detection of the sound wave arrival time based on AIC technology is as follows. Figure 6 shown.

[0100] Through the actual signal test of the project, it is found that the arrival time picking method based on the time window energy ratio method has the advantages of fast operation, simplicity and efficiency, and can effectively identify the time period of the AE signal damage characteristics, but its picking accuracy is greatly affected by the time window length; while the arrival time picking method based on AIC is sensitive to the change of signal stability, but cannot effectively identify the AE signal damage characteristics. Therefore, a method for accurately extracting the arrival time of acoustic emission damage signals that combines the advantages of both is proposed. The method includes two steps: detecting the time window position where the damage characteristics are located and accurately extracting the arrival time using local AIC.

[0101] In the method for accurately locating the acoustic emission source of a complex structure in a high-noise environment of the present application, the correspondence between the signals of each channel is still chaotic after composite noise reduction. In the sound source localization method, one-dimensional line positioning requires two sensors, and two-dimensional plane positioning requires at least three sensors. When the stress wave generated by the damage is transmitted to all sensors of the positioning array, the signal group that constitutes a positioning result is called an event. To facilitate subsequent positioning, the acoustic emission signals to be located in each channel of the positioning array are re-matched and combined based on the similarity of the arrival time of each signal of multiple sensors to form an effective event group that realizes the final location of the damage source. In step four, when the signals of each channel of the positioning array are matched, the acoustic emission signals of each channel are matched according to the similarity of the arrival time of the acoustic emission signals of each channel of the positioning array to form an event group that is ultimately used to calculate the positioning. Specifically including:

[0102] S401, converting the order of magnitude of the arrival time of each acoustic emission signal to the 0.1 ms level;

[0103] S402, extracting the integer part of the arrival time of each acoustic emission signal as the matching time;

[0104] S403: Match the acoustic emission signals with the same matching time one by one, and output an event group for positioning.

[0105] By comparing the matching time of the acoustic emission signals of each channel of the positioning array, the signals with the same matching time in each channel are matched one by one and the final valid positioning event group is output.

[0106] In the method for accurately locating acoustic emission sources of complex structures in high-noise environments of the present application, in step five, the time difference mapping library established in step one is used to perform positioning calculations on the event group signals using a weighted time difference mapping positioning algorithm to obtain the positioning coordinates of the damage signal source.

[0107] In order to solve the problem that the anisotropy and structural non-uniformity of composite materials lead to large differences in wave velocities in different propagation directions and thus large positioning errors, this application proposes the use of a weighted time-difference mapping (TDM) algorithm to achieve accurate positioning of the acoustic emission source of complex structures or materials. Specifically, for the acoustic emission signals to be located in each channel of the positioning array, the time difference matrices of each sensor pair are fused, the time difference data of the mapping library are compared, and the minimum value point in the fused time difference difference map is found to obtain the final positioning coordinates of the test point, thereby achieving accurate positioning of the acoustic emission source. The weighted time difference mapping positioning algorithm is mainly divided into the following four steps:

[0108] S501, calculating the time difference of the acoustic emission signal from the test point in the event group reaching each sensor pair

[0109] The time difference between the acoustic emission signal at the test point x and the arrival of each sensor pair is

[0110]

[0111] x(x=1,...,N x ) is the test point number, N x is the total number of test points, i and j are the sensor numbers.

[0112] When the number of sensors used is 2, the number of sensor pairs is 1, that is, sensor 1 and sensor 2 form a sensor pair, and the time difference of the acoustic emission signal reaching the sensor pair is expressed as Similarly, when the number of sensors used is 3, the number of sensor pairs is 3, that is, sensor 1 and sensor 2, sensor 1 and sensor 3, sensor 2 and sensor 3 constitute three sensor pairs respectively, and the time differences of the three sensor pairs are expressed as The calculation method of the number of sensor pairs SN is shown in formula (3), where A is the number of sensors.

[0113]

[0114] S502, calculate time difference The time difference of each grid node in the time difference mapping library with the same sensor pair The difference

[0115] Where y=1,2,...,N y is the node number in the time difference mapping library of the same sensor pair, N y is the total number of nodes;

[0116] S503, calculating the time difference weight f of the test point in each sensor pair K ;

[0117]

[0118] Wherein, K (K=1, 2, ..., SN) is the serial number of the sensor pair. is the time difference of the Kth sensor pair at test point x The time difference of each node in the pre-built time difference mapping database of the same sensor pair The difference between yK The total number of nodes in the database is mapped to the time difference of the Kth sensor pair.

[0119] S504: Fusion of the time difference of each sensor pair at the test point According to the time difference And the time difference weight f K The weighted time difference corresponding to each grid node is calculated, and the position coordinates of the grid node where the minimum weighted time difference is located are used as the damaged signal source positioning coordinates of the test point.

[0120] Fusion test point x each sensor pair time difference The weighted time difference value corresponding to each node in the fused time difference value database is calculated by formula (5): Finally, the position coordinates of the node with the minimum weighted time difference are used as the positioning coordinates of the test point.

[0121]

[0122] The method for accurately locating acoustic emission sources of complex structures in high-noise environments of the present application also includes step six, graphical visualization of positioning results: graphically visualizing the damaged parts of the structure according to the positioning coordinates of the damage signal source, evaluating the structural damage status, and outputting the damage positioning results.

[0123] The method for accurately locating the acoustic emission source of a complex structure in a high-noise environment of the present application realizes the accurate positioning of the acoustic emission source of a complex structure in a high-noise environment through the technologies of composite noise reduction of acoustic emission signals, "two-step" arrival time extraction, positioning array acoustic emission signal matching and weighted time difference mapping positioning. The present application has the following beneficial effects:

[0124] (1) Aiming at the problem that acoustic emission sensors are easily interfered by environmental noise due to their high sensitivity, a composite noise reduction method is proposed for single-channel monitoring signals before locating the acoustic emission source. The noise interference is effectively reduced by time domain waveform feature screening and frequency domain filtering.

[0125] (2) The proposed sound wave arrival time detection method based on time window energy ratio and AIC technology effectively solves the problem of inaccurate sound wave arrival time measurement in high noise environment by traditional acoustic emission source location technology, and greatly improves the accuracy of sound wave arrival time detection;

[0126] (3) The proposed weighted time difference mapping positioning technology effectively solves the problem of large positioning error caused by uneven wave velocity when facing complex materials or structures using traditional acoustic emission positioning technology.

[0127] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.

Claims

1. A method for accurately locating acoustic emission sources of complex structures in a high noise environment, characterized in that: include: Step 1: Build a time difference mapping library for positioning; Step 2: collecting acoustic emission signals and performing composite noise reduction on the acoustic emission signals; Step 3: Processing the acoustic emission signal by using the time window energy ratio method and the AIC method to obtain the arrival time of each acoustic emission signal; Step 4: perform signal matching based on the similarity of the arrival time of each acoustic emission signal to obtain an event group for positioning; Step 5: Perform positioning calculation based on the time difference mapping library and the event group to obtain the positioning coordinates of the damage signal source.

2. The method for accurately locating acoustic emission sources of complex structures in high-noise environments according to claim 1 is characterized in that: In step 1, a time difference mapping library for positioning is constructed, including: S101, determining a detection area and a sensor layout plan, and dividing the detection area into grids; S102, using artificial lead breaking to simulate the acoustic emission source to excite each grid node, and collecting lead breaking waveform data at the same time; S103, processing all lead-break waveform data to obtain the arrival time of each acoustic emission signal; S104, calculating the arrival time difference of all sensor pairs on each grid node according to the arrival time, and constructing an initial time difference mapping library of each sensor pair on all grid nodes; S105, eliminating abnormal data in the initial time difference mapping library based on the quartile method, and using a bilinear interpolation fitting algorithm to refine the data in the initial time difference mapping library and fill in missing data, so as to obtain a time difference mapping library for positioning.

3. The method for accurately locating acoustic emission sources of complex structures in high-noise environments according to claim 2 is characterized in that: In step 2, composite noise reduction is performed on the acoustic emission signal, including: S201, constructing a deep residual neural network model, inputting the material damage acoustic emission signal and the environmental noise signal obtained through the experiment into the deep residual neural network model for training, and performing time domain waveform feature recognition and noise reduction on the acoustic emission signal through the trained deep residual neural network model; S202: Perform frequency domain filtering and noise reduction on the acoustic emission signal using a Butterworth digital bandpass filter.

4. The method for accurately locating acoustic emission sources of complex structures in high-noise environments according to claim 3 is characterized in that: In step three, the acoustic emission signal is processed by the time window energy ratio method and the AIC method to obtain the arrival time of each acoustic emission signal, including: S301, processing the acoustic emission signal by a time window energy ratio method to obtain a damage characteristic time window position; S302: Process the damage characteristic time window position by AIC method to obtain the arrival time of each acoustic emission signal.

5. The method for accurately locating acoustic emission sources of complex structures in high-noise environments according to claim 4 is characterized in that: In S301, the acoustic emission signal is processed by a time window energy ratio method to obtain a damage feature time window position, including: Assuming that L1 and L2 are two adjacent time windows, the time window energy ratio of the acoustic emission signal is calculated; Where x(t) is the time series of the acoustic emission signal, t is the sample point, T1 is the starting point of the first time window, T0 is the end point of the first time window, which is also the starting point of the second time window, T2 is the end point of the second time window, ω is the relative energy of the acoustic emission signal waveform, and α is the stability coefficient; The damage feature time window position is obtained according to the maximum value of the time window energy ratio.

6. The method for accurately locating acoustic emission sources of complex structures in high-noise environments according to claim 1 is characterized in that: In S302, the damage characteristic time window position is processed by the AIC method to obtain the arrival time of each acoustic emission signal, including: Calculate the AIC value of the damage feature time window position: AIC(t)=tlog 10 (var({x(1:t)})+(T-t-1)log 10 (var({x(t:T})) Among them, var({x(1:t}) represents the variance of x, and T is the last sample point; The arrival time of each acoustic emission signal is obtained according to the minimum value of the AIC value.

7. The method for accurately locating acoustic emission sources of complex structures in high-noise environments according to claim 6 is characterized in that: Step 4: Perform signal matching based on the similarity of the arrival time of each acoustic emission signal to obtain an event group for positioning, including: S401, converting the order of magnitude of the arrival time of each acoustic emission signal to the 0.1 ms level; S402, extracting the integer part of the arrival time of each acoustic emission signal as the matching time; S403: Match the acoustic emission signals with the same matching time one by one, and output an event group for positioning.

8. The method for accurately locating acoustic emission sources of complex structures in high-noise environments according to claim 7 is characterized in that: In step 5, positioning calculation is performed according to the time difference mapping library and the event group to obtain the positioning coordinates of the damage signal source, including: S501, calculating the time difference of the acoustic emission signal from the test point in the event group reaching each sensor pair S502, calculate time difference The time difference of each grid node in the time difference mapping library with the same sensor pair The difference S503, calculating the time difference weight f of the test point at each sensor pair K ; S504: Fusion of the time difference of each sensor pair at the test point According to the time difference And the time difference weight f K The weighted time difference corresponding to each grid node is calculated, and the position coordinates of the grid node where the minimum weighted time difference is located are used as the damaged signal source positioning coordinates of the test point.

9. The method for accurately locating acoustic emission sources of complex structures in high-noise environments according to claim 8 is characterized in that: The method also includes step 6, graphically visualizing the structural damage location according to the damage signal source positioning coordinates, evaluating the structural damage state, and outputting the damage positioning result.

Images