Method for locating multiple defects of steel pipe based on acoustic emission signal modal decomposition

Abstract

The present application relates to the technical field of steel pipe defect detection and positioning technology, specifically to a steel pipe multiple defect positioning method based on acoustic emission signal modal decomposition, comprising the following steps: S1, system deployment, S2, signal analysis, S3, signal algorithm decomposition, S4, signal classification, S5, determining the optimal position defect. The steel pipe multiple defect positioning method based on acoustic emission signal modal decomposition can realize simultaneous detection of multiple defects in the steel pipe, and proposes an improved acoustic emission signal detection algorithm and an acoustic emission source arrival time difference positioning algorithm. Compared with existing defect detection methods, the present model can distinguish multiple defects detected in a time period and accurately obtain their defect positions. The positioning algorithm proposed by the present model can eliminate error factors such as sensor reaction time, system delay time uncertainty, system synchronous sensor clock error, and take the optimal value as the defect coordinates to improve the positioning accuracy.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of steel pipe defect detection and positioning technology, in particular to a steel pipe multi-defect positioning method based on acoustic emission signal modal decomposition. BACKGROUND

[0002] The common steel medium defect detection and positioning methods at present include ultrasonic flaw detection and positioning method, X-ray detection and positioning algorithm, magnetic flux leakage defect detection and positioning algorithm, etc. The ultrasonic flaw detection has the advantages of high flaw detection sensitivity. The X-ray industrial flaw detection is used more in large factories and assembly lines. Due to its high degree of automation, it is convenient and can be used with other automatic operations to increase industrial efficiency. The principle of magnetic flux leakage defect detection is that when the steel material is powered, the magnetic field curve amplitude of the place with defects often shows a downward or decreasing trend. When the magnetic field signal is converted into an electric signal, the strength of the electric signal at the defect will also decrease, so it can be determined that the steel pipe material has defects. The magnetic flux leakage detection is generally not suitable for flaw detection of objects with covering layers and coatings. When there is a defect in the steel workpiece, an interface between different media will be formed between the defect and the steel material. The acoustic impedance between the interfaces is different. When the reflected wave of the emitted ultrasonic wave encounters this interface, the amplitude and frequency of the reflected wave will be different from those of the original wave. Therefore, the acoustic emission wave signal can be used to achieve the purpose of low equipment requirement and simultaneous detection of multiple defect sources.

[0003] The common steel medium defect detection and positioning methods at present include ultrasonic flaw detection and positioning method, X-ray detection and positioning algorithm, magnetic flux leakage defect detection and positioning algorithm, etc. Although the ultrasonic flaw detection has high flaw detection sensitivity, it requires the surface of the workpiece to be smooth, and the ultrasonic energy attenuates seriously, so the receiving device for the signal has high requirements. The initial image effect of the X-ray is relatively important, so the environment for image acquisition also has high requirements, and the use cost is also high. Therefore, it is not suitable for steel pipe defect detection in the underground pipe gallery scene. The magnetic flux leakage defect detection is only suitable for the case where the defect is not large and the depth is not deep. When the defect is inside the detected object or the defect is large and the defect depth is deep, the detection accuracy will also be affected. SUMMARY

[0004] (1) Technical problems solved

[0005] In view of the deficiencies of the prior art, the present application provides a steel pipe multi-defect positioning method based on acoustic emission signal modal decomposition, which has the advantages of improving defect detection accuracy and positioning accuracy, simple production method, low detection cost, and unaffected by various environments, etc. The problems that the existing other positioning methods are greatly affected by environmental factors and the accuracy cannot be guaranteed in the use process are solved.

[0006] (2) Technical solutions

[0007] In order to achieve the above-mentioned advantages of improving defect detection accuracy and positioning accuracy, simple production method, low detection cost, and unaffected operation in various environments, the present application provides the following technical solutions: a steel pipe multiple defect positioning method based on acoustic emission signal modal decomposition, comprising the following steps:

[0008] S1, system deployment

[0009] The system composed of acoustic emission defect detection devices, acoustic emission signal receiving devices and central data processing servers is deployed in underground pipe gallery environments where manual detection of steel pipe defects is difficult, wherein the acoustic emission defect detection device is used for detecting steel pipe defects, so that acoustic emission signals generated at the steel pipe defect propagate along the steel pipe wall, the acoustic emission signal receiving device is used for collecting acoustic emission signals, the central data processing server includes a steel pipe body model, a sensor relative attachment position information, an acoustic emission signal processing classification module and a defect positioning analysis algorithm module, the acoustic emission signal is received by an external sensor module, and different position receiving sensors will receive acoustic emission signals emitted simultaneously by the same acoustic emission signal source.

[0010] S2, analyze the signal

[0011] The system model considers the case where the defect detection device detects multiple defects at the same time or with little time difference, analyzes the acoustic emission signals generated by the defects that arrive at the same time, and uses the acoustic emission source optimization positioning algorithm based on the arrival time difference to distinguish and accurately position the multiple defect coordinates, wherein the acoustic emission signal processing classification module is used to perceive, decompose and classify all signals collected by each sensor at each time, and obtain the optimal positioning accuracy.

[0012] S3, signal algorithm decomposition

[0013] The received acoustic emission signals are decomposed using an improved acoustic emission signal detection algorithm, and the improved adaptive noise modal decomposition algorithm (I-CEEMDAN) can convert the original acoustic emission signals into time-frequency spectrum signals (Hilbert-Huang spectrum) which can clearly show the time-frequency characteristics of each acoustic emission signal.

[0014] S4, signal classification

[0015] When the time-frequency spectrum signal is obtained, the singular value decomposition algorithm can effectively compress the data amount and extract the input features of the signal, so that the signals of each frequency can be classified, that is, the signals of the same acoustic emission signal source collected by different sensors at the same time can be classified, and the time when the signals reach each sensor is obtained. By repeatedly processing the above process, the data of the signals emitted by all defects detected by the detection equipment in the pipeline can be obtained, and these data can be stored in real time.

[0016] S5, determining the optimal position defect

[0017] In view of the problem that the positioning accuracy of acoustic emission events is low under a small sample, a VFOM (virtual field optimization method) positioning algorithm based on multi-scale grid search is proposed to determine the optimal position defect.

[0018] Preferably, in step S3, after the acoustic emission signal receiving devices attached to both sides of the steel pipe collect the acoustic emission signals, the signals are first processed, and a modal decomposition model based on an improved adaptive noise is used to decompose the acoustic emission signals to obtain modal components. It is assumed that the acoustic emission signal is x(n):

[0019] x (i) (n) = x(n) + β0E1(ω (i) (n)), (i = 1, 2,..., I)

[0020] where x (i) (n) represents the weight sum of the signal x(n) at i time and the Gaussian white noise ω (i) (n).

[0021] represents the complex number of the kth empirical mode component obtained by empirical mode decomposition (k = 1, 2,...), and the first residual component r1, where is calculated, and the first residual component r1is obtained. By combining the original signal x(n), the first empirical mode component IMF1can be calculated:

[0022] IMF1= x(n) - r1(n).

[0023] Preferably, after the first empirical mode component IMF1is calculated, the second empirical mode component IMF2is calculated:

[0024] where r2is the second residual component, and the kth residual component r k(k=3,...,K), the kth modal component IMF k :

[0025]

[0026] IMF k = r k-1 - r k

[0027] Finally, the original signal can be seen as the sum of a plurality of modal components and residual components, namely:

[0028]

[0029] Preferably, after obtaining the IMF components, the time-frequency spectrum signal of each modal component is obtained by using Hilbert transform, and for a signal x(t), the Hilbert transform is:

[0030]

[0031] Where P is the Cauchy principal value, the instantaneous amplitude of the signal is α(t), the instantaneous phase is θ(t), and the instantaneous frequency is ω(t):

[0032]

[0033]

[0034] Preferably, in step S3, the time-frequency spectrum can be represented as H(ω, t), and the time-frequency spectrum of a signal can be seen as a matrix, and according to the singular value decomposition principle, all key features of the matrix can be obtained in the form of a series of singular values in order, and according to the similarities and differences between these key features, the acoustic emission signals from different defects obtained by each sensor can be classified, so that the time of arrival of each acoustic emission signal to each sensor can be obtained, and according to the frequency amplitude, the signals from the same acoustic emission source to different sensors can be classified, thereby realizing simultaneous detection of multiple defects.

[0035] Preferably, in step S5, according to the singular value decomposition principle, the time of arrival of the acoustic emission signal to different sensors can be obtained, and by using parameters such as frequency and amplitude of the signal, it can be distinguished whether it is the same acoustic emission source signal, and then, according to the actual size and structure of the steel pipe, according to the sensor planning position coordinates and acoustic emission wave speed that have been determined, and considering the clock error when entering the system, the acoustic emission signal positioning model can be expressed as:

[0036]

[0037] Where, (x j , yj ,z j ) represents the physical position of the jth sensor, (x0, y0, z0) represents the defect position coordinates, which needs to be calculated, t0 represents the starting time of the acoustic emission signal emitted from the defect, t j represents the time when the signal reaches the physical position of the jth sensor, T j represents the system error time of the jth sensor, v ae represents the speed of the acoustic emission signal propagating in the steel pipe, the time of the acoustic emission signals emitted by different defects reaching different sensors is different, and the same acoustic emission signal has been classified, according to the virtual field optimization algorithm based on multi-scale search:

[0038] The above formula represents the simultaneous equations of the same acoustic emission source signal reaching the ith and jth sensors at different times, when the local coordinate system of the pipeline and the sensor is determined, according to the operation rule of the time difference hyperbolic equation, the above formula can be rewritten as:

[0039]

[0040] Wherein:

[0041]

[0042] Preferably, the local coordinate system coordinates of the sensor i and the sensor j in the sensor module are known, so c ij can be calculated. Since the energy of the acoustic emission signal decreases with the increase of the distance, an attenuation function f ij (X, Y, Z) is established for the acoustic emission signal emitted by the defect coordinates (X, Y, Z):

[0043]

[0044] Where d ij represents the distance from the acoustic emission signal source to the sensor coordinate surface:

[0045]

[0046] Then convert the local coordinate system f ij (X, Y, Z) to the global coordinate system f ij (x, y, z), which can be obtained:

[0047]

[0048] Preferably, the R ij is obtained by (x i , y i , z i ) and (x j , y jz j ) represents a constant matrix:

[0049]

[0050] wherein:

[0051]

[0052] wherein (x, y, z) represents global coordinate system coordinates, the total length of the close can be represented as:

[0053] n represents the total number of sensors receiving the acoustic emission signal, and the coordinates where the maximum value of the TCF in the model space is taken as the defect positioning result.

[0054] (Three) beneficial effects

[0055] Compared with the prior art, the present application provides a steel pipe multi-defect positioning method based on acoustic emission signal modal decomposition, which has the following beneficial effects:

[0056] 1. The steel pipe multi-defect positioning method based on acoustic emission signal modal decomposition can realize simultaneous detection of multiple defects in the steel pipe, and an improved acoustic emission signal detection algorithm and an acoustic emission source arrival time difference positioning algorithm are proposed. Compared with existing defect detection methods, the present model can distinguish multiple defects detected in a time period and accurately obtain the defect positions, and can not only detect defects on the surface of the steel pipe, but also detect defects inside the wall of the steel pipe.

[0057] 2. The steel pipe multi-defect positioning method based on acoustic emission signal modal decomposition, the traditional acoustic emission signal is a kind of vibration wave signal (hereinafter referred to as p wave signal) generated by defects when the pipe wall is subjected to structural stress under the pressure of the pipe wall of the steel pipe. Since the position of the defect is unknown, the starting time position of the acoustic emission wave occurs, and the system has many problems such as sensor reaction time, system delay time uncertainty, clock error between system synchronous sensors, etc. Therefore, the positioning algorithm proposed by the present model can eliminate these error factors, and the optimal value is taken as the defect coordinates to improve the positioning accuracy. This method is also applicable to the positioning of multiple defect positions detected simultaneously or with very small time difference. BRIEF DESCRIPTION OF DRAWINGS

[0058] Fig. 1 is a total equipment distribution diagram for acoustic emission signal detection and positioning;

[0059] Fig. 2 is a defect detection flowchart for acoustic emission signal. DETAILED DESCRIPTION

[0060] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0061] Please refer to Figs. 1-2 The present application provides a technical solution: a steel pipe multiple defect positioning method based on acoustic emission signal modal decomposition, comprising the following steps:

[0062] S1, system deployment

[0063] The system composed of the acoustic emission defect detection device, the acoustic emission signal receiving device and the central data processing server is deployed in the underground pipe gallery environment where manual detection of steel pipe defects is difficult. The acoustic emission defect detection device is used for detecting the defects of the steel pipe, so that the acoustic emission signals generated at the defects of the steel pipe propagate along the wall of the steel pipe. The acoustic emission signal receiving device is used for collecting the acoustic emission signals. The central data processing server includes a steel pipe body model, sensor relative attachment position information, an acoustic emission signal processing classification module and a defect positioning analysis algorithm module. The acoustic emission signals are received by the external sensor module. The receiving sensors at different positions receive the acoustic emission signals emitted simultaneously by the same acoustic emission signal source.

[0064] S2, analyze the signal

[0065] The system model considers the case that the defect detection device detects multiple defects at the same time or with a small time difference. The acoustic emission signals generated by the defects arriving at the same time are analyzed, and the acoustic emission source optimization positioning algorithm based on the arrival time difference is used to distinguish and accurately position the multiple defect coordinates. The acoustic emission signal processing classification module is used to perceive, decompose and classify all signals collected by each sensor at each time, so as to obtain the optimal positioning accuracy.

[0066] S3, signal algorithm decomposition

[0067] The received acoustic emission signals are decomposed by using an improved acoustic emission signal detection algorithm. The improved adaptive noise-based modal decomposition algorithm (I-CEEMDAN) can convert the original acoustic emission signals into time-frequency spectrum signals (Hilbert-Huang spectrum) which can clearly show the time-frequency characteristics of each acoustic emission signal. After the acoustic emission signal receiving devices attached to both sides of the steel pipe collect the acoustic emission signals, these signals need to be processed first. The acoustic emission signals are decomposed by using the improved adaptive noise-based modal decomposition model to obtain modal components. It is assumed that the acoustic emission signal is x(n):

[0068] x (i) (n) = x(n) + β0E1(ω (i) (n)), (i = 1, 2,..., I)

[0069] where x (i) (n) represents the weight sum of signal x(n) at i time and Gaussian white noise ω (i) (n),

[0070] represents the kth empirical mode component obtained by empirical mode decomposition (k = 1, 2,...), and then the first residual component r1, where represents the operation of averaging, by obtaining the first residual component r1, in combination with the original signal x(n), the first empirical mode component IMF1 can be calculated: IMF1 = x(n) - r1(n), after calculating the first empirical mode component IMF1, the second empirical mode component IMF2 is calculated:

[0071] where r2 is the second residual component, and so on, the kth residual component r k (k = 3,..., K) can be obtained, and the kth mode component IMF k can also be obtained:

[0072]

[0073] IMF k = r k-1 -r k

[0074] Finally, the original signal can be regarded as the sum of multiple mode components and residual components, that is:

[0075] After obtaining the IMF component, the time-frequency spectrum signal of each mode component is obtained by using Hilbert transform, for the signal x(t), its Hilbert transform is:

[0076]

[0077] where P is the Cauchy principal value, the instantaneous amplitude of the signal is α(t), the instantaneous phase is θ(t), and the instantaneous frequency is ω(t):

[0078]

[0079] The time-frequency spectrum can be expressed as H(ω, t), and the time-frequency spectrum of a signal can be regarded as a matrix, and according to the singular value decomposition principle, all key features of the matrix can be obtained in sequence in the form of a series of singular values, and according to the similarities and differences between the key features, the acoustic emission signals from different defects obtained by each sensor can be classified, so that the time of arrival of each acoustic emission signal at each sensor can be obtained, and according to the frequency amplitude, the signals from the same acoustic emission source arriving at different sensors can be classified, and thus the simultaneous detection of multiple defects can be realized.

[0080] S4, signal classification

[0081] After obtaining the time-frequency spectrum signal, the singular value decomposition algorithm can effectively compress the data amount and extract the input features of the signal, so that the signals of each frequency can be classified, that is, the signals of the same acoustic emission signal source collected by different sensors at the same time can be classified, and the time of arrival of these signals at each sensor can be obtained, and the above process is repeatedly processed to obtain the data of the signals emitted by all defects detected by the detection device in the pipeline, and these data are stored in real time.

[0082] S5, determining the optimal position defect

[0083] In view of the problem that the positioning accuracy of acoustic emission events is low under a small sample, a VFOM (virtual field optimization method) positioning algorithm based on multi-scale grid search is proposed to determine the optimal position defect. According to the singular value decomposition principle, the time of arrival of the acoustic emission signal at different sensors can be obtained, and through the frequency and amplitude parameters of the signal, it can be distinguished whether it is the same acoustic emission source signal. Then, according to the actual size and structure of the steel pipe, according to the sensor planning position coordinates and acoustic emission wave velocity that have been determined, and considering the system clock error, the acoustic emission signal positioning model can be expressed as:

[0084]

[0085] where (x j ,y j ,z j ) represents the physical position of the jth sensor, (x0, y0, z0) represents the defect position coordinates, which needs to be calculated, t0 represents the starting time of the acoustic emission signal emitted from the defect, t j represents the time of arrival of the signal at the physical position of the jth sensor, T j represents the system error time of the jth sensor, and v aeThe speed of propagation of the acoustic emission signal in the steel pipe is represented by c, which is based on the different times of arrival of acoustic emission signals emitted by different defects at different sensors, and the same acoustic emission signal has been classified, according to a virtual field optimization algorithm based on multi-scale search:

[0086] The above formula represents a simultaneous equation of the same acoustic emission source signal arriving at the i th and j th sensors at different times, respectively, when the pipe and sensor local coordinate system is determined, according to the time difference hyperbolic equation operation rule, the above formula can be rewritten as:

[0087]

[0088] Wherein:

[0089]

[0090] The local coordinate system coordinates of the sensor i and the sensor j in the sensor module are known, so c ij can be calculated. Since the acoustic emission signal energy decreases with increasing distance, an attenuation function f ij (X,Y,Z) is established for the acoustic emission signal emitted by the defect coordinate (X,Y,Z):

[0091]

[0092] Where d ij represents the distance from the acoustic emission signal source to the sensor coordinate surface:

[0093]

[0094] Then convert the local coordinate system f ij (X,Y,Z) to the global coordinate system f ij (x,y,z), which can be obtained: R ij is a constant matrix represented by (x i ,y i ,z i ) and (x j ,y j ,z j ):

[0095]

[0096] Wherein:

[0097]

[0098] Wherein, (x,y,z) represents the global coordinate system coordinates, and the total close length can be represented as:

[0099] n represents the total number of sensors receiving the acoustic emission signal, and the coordinates where the maximum value of the TCF in the model space is generally considered as the defect positioning result.

[0100] Although embodiments of the present application have been shown and described, it is to be understood that various modifications, substitutions, replacements and changes can be made to these embodiments without departing from the principles and spirit of the present application, and the scope of the present application is defined by the appended claims and their equivalents.

Claims

1. A method for locating multiple defects in a steel pipe based on acoustic emission signal modal decomposition, characterized in that, Comprise the following steps: S1, system deployment The system composed of acoustic emission defect detection device, acoustic emission signal receiving device and central data processing server is deployed in the underground pipe gallery environment where artificial detection of steel pipe defects is difficult, wherein the acoustic emission defect detection device is used for detection of steel pipe defects, so that the acoustic emission signal generated at the steel pipe defect propagates along the steel pipe wall, the acoustic emission signal receiving device is used for collection of the acoustic emission signal, the central data processing server includes a steel pipe body model, a sensor relative attachment position information, an acoustic emission signal processing classification module and a defect positioning analysis algorithm module, the acoustic emission signal is received by an external sensor module, and different position receiving sensors will receive the acoustic emission signals emitted simultaneously by the same acoustic emission signal source; S2, analyze the signal The system model considers the case that the defect detection device detects multiple defects at the same time or with little time difference, analyzes the acoustic emission signals generated by the defects arriving at the same time, and uses the acoustic emission source optimization positioning algorithm based on the arrival time difference to distinguish and accurately position the multiple defect coordinates, wherein the acoustic emission signal processing classification module is used to perceive, decompose and classify all signals collected by each sensor at each time, and obtain the optimal positioning accuracy; S3, signal algorithm decomposition The received acoustic emission signal is decomposed by using an improved acoustic emission signal detection algorithm, and the improved adaptive noise modal decomposition algorithm (I-CEEMDAN) can convert the original acoustic emission signal into a time-frequency spectrum signal (Hilbert-Huang spectrum) which can clearly show the time-frequency characteristics of each acoustic emission signal; S4, signal classification After obtaining the time-frequency spectrum signal, the singular value decomposition algorithm can effectively compress the data volume and extract the input features of the signal, so that each frequency signal can be classified, that is, the signals of the same acoustic emission signal source collected by different sensors at the same time can be classified, and the time when these signals reach each sensor is obtained. By processing the above process in a loop, the data of the signals emitted by all defects detected by the detection device in the pipeline can be obtained, and these data can be stored in real time; S5, determine the optimal position defect In view of the problem that the positioning accuracy of acoustic emission events is low under small sample, a VFOM (virtual field optimization method) positioning algorithm based on multi-scale grid search is proposed to determine the optimal position defect.

2. The method for locating multiple defects of steel pipe based on acoustic emission signal modal decomposition according to claim 1, characterized in that, In the step S3, after the acoustic emission signal receiving devices attached to both sides of the steel pipe collect the acoustic emission signals, the signals are first processed, the acoustic emission signals are decomposed to obtain modal components by using a modal decomposition model based on an improved adaptive noise, and it is first assumed that the acoustic emission signal is : ； wherein represents a signal at time instant and a Gaussian white noise with weight and represents a complex number of the first empirical mode component of the signal obtained by empirical mode decomposition, then the first residue component of the signal is calculated , wherein represents an averaging operation, by obtaining the first residue component , in combination with the original signal , the first empirical mode component is calculated 。 3. The method for locating multiple defects of steel pipe based on acoustic emission signal modal decomposition according to claim 2, characterized in that, After the computation of the first empirical mode component is completed, the computation of the second empirical mode component is started. where is the second residual component, and so on, to obtain the Kth residual component k = 3,..., K, and the Kth modal component : ； ； Finally, the original signal can be regarded as the sum of multiple modal components and residual components, that is: 。 4. The method for locating multiple defects of steel pipe based on acoustic emission signal modal decomposition according to claim 3, characterized in that, Obtained After obtaining the components, the time-frequency spectrum signal of each modal component is obtained by using the Hilbert transform, for the signal The Hilbert transform thereof is: ； wherein is the Cauchy principal value, the instantaneous amplitude of the signal is , the instantaneous phase is , and the instantaneous frequency is : ； ； 。 5. The method for locating multiple defects of steel pipe based on acoustic emission signal modal decomposition according to claim 4, characterized in that, The time-frequency spectrum in step S3 can be expressed as The time-frequency spectrum of a signal can be regarded as a matrix, and according to the singular value decomposition principle, all key features of the matrix can be obtained in the form of a series of singular values in sequence. According to the similarities and differences between the key features, the acoustic emission signals from different defects obtained by each sensor can be classified, so that the time of arrival of each acoustic emission signal at each sensor can be obtained, and according to the frequency amplitude, the signals from the same acoustic emission source arriving at different sensors can be classified, thereby realizing simultaneous detection of multiple defects.

6. The method for locating multiple defects of steel pipe based on acoustic emission signal modal decomposition according to claim 5, characterized in that, In the step S5, according to the singular value decomposition principle, the time when the acoustic emission signal reaches different sensors can be obtained, and by the frequency and amplitude of the signal, it can be distinguished whether it is the same acoustic emission source signal. Then, according to the actual size and structure of the steel pipe, according to the sensor planning position coordinates and acoustic emission wave velocity that have been determined, and considering the clock error of the system, the acoustic emission signal positioning model can be expressed as: ； wherein, represents the physical position of the th sensor, represents the defect position coordinate, which is to be calculated, represents the start time of the acoustic emission signal emitted from the defect, represents the time of the signal reaching the physical position of the th sensor, represents the system error time of the th sensor, represents the speed of the acoustic emission signal propagating in the steel pipe, the times of the acoustic emission signals emitted from different defects reaching different sensors are different, and the same acoustic emission signal has been classified, according to a virtual field optimization algorithm based on multi-scale search: The above equation represents simultaneous equations of the same acoustic emission source signal arriving at the first and the second sensors at different times, respectively. When the local coordinate system of the pipeline and the sensors is determined, the above equation can be rewritten according to the operation law of the hyperbolic equation of time difference as follows: ； Wherein: ； ； 。 7. The method for locating multiple defects of steel pipe based on acoustic emission signal modal decomposition according to claim 6, characterized in that, The transducer module And the transducer The local coordinate system coordinates are known, therefore Can be calculated; since the acoustic emission signal energy decreases with increasing distance, the defect coordinates The acoustic emission signal emitted by the defect establishes a decay function : ； wherein represents the distance of the acoustic emission signal source to the sensor coordinate surface: ； The local coordinate system is converted to the global coordinate system again The local coordinate system is converted to the global coordinate system again , we have 。 8. The method for locating multiple defects of steel pipe based on acoustic emission signal modal decomposition according to claim 7, characterized in that, The is a constant matrix represented by and ​ ， Wherein: ； ； ； ； ， wherein, represents the global coordinate system coordinates, the total length of the approach can be represented as: , This represents the total number of sensors that received acoustic emission signals, and is located in the model space. The coordinates of the maximum value are considered as the defect location result.

Images