Leak detection method, device and system for pressure vessel

Abstract

The invention relates to the technical field of sealing performance detection, in particular to a pressure vessel-oriented leakage detection method, device and system, and the method comprises the steps: respectively obtaining the acoustic emission signals of each detection sensor on the outer wall surface of a pressure vessel and a guard sensor right above the detection sensor, and the liquid level height at each moment; judging the appearing acoustic emission event; obtaining each event set and a noise interference degree corresponding to the event set, and dividing the event sets into a noise event set and a non-noise event set; and determining a sound source estimation value and a liquid drop estimation value to obtain a modal decomposition layer number, performing signal reconstruction on all modal components corresponding to non-noise through modal decomposition, and positioning the leakage position of the pressure vessel. According to the method, the number of decomposition layers of the modal decomposition algorithm is dynamically adjusted, insufficient decomposition or excessive decomposition of the mode of the acoustic emission signal is avoided, the filtering effect on the acoustic emission signal is improved, and the leakage point of the pressure vessel can be positioned more accurately.

Description

Technical Field

[0001] The present application relates to the technical field of sealing detection, and in particular to a leak detection method, equipment and system for pressure vessels. Background Art

[0002] Pressure vessels are widely used in many industries such as chemical, energy, and food, and their safety is of vital importance. Once a leak occurs, it will not only cause medium loss and resource waste, but may also cause serious accidents such as fire, explosion, and poisoning, threatening people's lives and damaging the ecological environment.

[0003] In the process of leak detection of pressure vessels by traditional methods using acoustic emission detection, in large-volume pressure vessels, the acoustic emission signal will be attenuated on the propagation path due to medium absorption, scattering and the presence of interference sources, resulting in an increase in the influence of noise on the delay estimation of the waveform of the acoustic emission signal. Secondly, for complex, variable, nonlinear and non-stationary acoustic emission signals, when filtering and denoising through the modal decomposition algorithm, the number of decomposition layers of the modal decomposition algorithm needs to be set in advance, which is prone to modal aliasing or over-decomposition, affecting the accuracy of the final extraction of feature information in the acoustic emission signal, resulting in low accuracy in locating the leakage point of the pressure vessel. Summary of the invention

[0004] In order to solve the above technical problems, a leak detection method, device and system for pressure vessels are provided to solve the existing problems.

[0005] The solution to the technical problem of the present application is to provide a leak detection method, device and system for pressure vessels, including the following steps: In a first aspect, an embodiment of the present application provides a leak detection method for a pressure vessel, the method comprising the following steps: Acquire the acoustic emission signals of each detection sensor on the outer wall of the pressure vessel and the guard sensor directly above it, as well as the liquid level height at each moment; The acoustic emission signal is divided into frames, and the upper envelope of each frame signal under each detection sensor is extracted; the acoustic emission event that occurs is judged according to the deviation of the liquid level height at all times, the size of the signal amplitude in the upper envelope and the distribution of the extreme value points; Analyze the energy intensity of the signal corresponding to each acoustic emission event and the duration of the event for each frame of the signal of each detection sensor, and combine the frequency distribution of the signal in the frequency domain to form a feature vector. Classify the acoustic emission events of the same frame signal under each detection sensor and its adjacent detection sensors, and obtain each event set corresponding to each frame of signal under each detection sensor; Determine the noise interference degree of each event set according to the difference in the feature vectors between different acoustic emission events in each event set and the number of acoustic emission events, and divide the event set into a noise event set and a non-noise event set; Based on the number of the noise event set, the number of the non-noise event set, and the noise interference degree, determine the sound source estimation value of each frame of signal under each detection sensor; Analyze the energy intensity of the signal corresponding to each acoustic emission event in each frame of the signal of the guard sensor, calculate the droplet estimation value, and combine the sound source estimation value to obtain the modal decomposition layer number of each frame of signal under each detection sensor, and perform modal decomposition on each frame of signal under each detection sensor to obtain all modal components; Obtain the modal components corresponding to the non-noise through the correlation between each modal component and the other modal components, and after signal reconstruction of all the modal components corresponding to the non-noise, locate the leakage position of the pressure vessel.

[0006] Preferably, the judgment of the occurring acoustic emission events includes: Record the ratio of the average value of the liquid level heights at all times to the preset rated operating height as the liquid level ratio; Effective threshold value The calculation formula of is: , where is the preset reference value, is the liquid level ratio, is the preset reference ratio, is the exponential function with the natural constant as the base; Obtain the maximum points and minimum points of the upper envelope; Starting from any maximum point and extending backward, record the moment when the signal amplitude after the starting point in the upper envelope is less than the effective threshold value and the first minimum point after the starting point as the critical points; Record the critical point closest to the starting point as the termination point; For each frame of signal under each detection sensor, record the time period between the starting point corresponding to each maximum point and the termination point as an acoustic emission event.

[0007] Preferably, the method for obtaining the feature vector is: For each signal corresponding to each acoustic emission event within each frame of signals from each detection sensor, calculate the duration of the signal corresponding to each acoustic emission event, denoted as the duration; calculate the product of the total energy of all signal amplitudes within the signal corresponding to each acoustic emission event and the duration, as the evaluation value for each acoustic emission event; Conduct frequency-domain analysis on the signal corresponding to each acoustic emission event to obtain a spectrogram. Denote the frequency corresponding to the maximum energy in the spectrogram as the main frequency; denote the frequency corresponding to the central value after integrating all energies in the spectrogram as the central frequency; Statistically calculate the ratio of the number of signal amplitudes less than the effective threshold value within the signal corresponding to each acoustic emission event to the duration, denoted as the average frequency; Perform normalization processing on the main frequency, the central frequency, the average frequency, and the evaluation value respectively, and form a feature vector.

[0008] Preferably, the obtaining of each event set corresponding to each frame of signals from each detection sensor includes: clustering the feature vectors corresponding to all acoustic emission events of the same frame of signals from each detection sensor and all its adjacent detection sensors to obtain multiple clustering clusters, and denoting all acoustic emission events belonging to the same clustering cluster as each event set corresponding to each frame of signals from each detection sensor.

[0009] Preferably, the determining of the noise interference degree of each event set and the partitioning of the event set into a noise event set and a non-noise event set includes: For each event set corresponding to each frame of signals of any detection sensor, denote the mean of the distances between the feature vectors of any two acoustic emission events belonging to the any detection sensor within each event set as the dispersion degree of each event set; Statistically calculate the number of all acoustic emission events within each event set, denoted as the total number; calculate the ratio of the number of acoustic emission events belonging to the any detection sensor within each event set to the total number, denoted as the relative ratio; The noise interference degree is the normalized result of the product of the relative ratio and the dispersion degree; Obtain the segmentation threshold of the noise interference degree of all event sets corresponding to each frame of signals of the any detection sensor, denoted as the first segmentation threshold; denote the event sets with a noise interference degree greater than the first segmentation threshold as the noise event set, and vice versa as the non-noise event set.

[0010] Preferably, the sound source estimation value of the th frame of signals of the th detection sensor is calculated by the formula: , where is the th detection sensor and The number of all non-noise event sets corresponding to the frame signal, is the number of all noise event sets corresponding to the frame signal of the th detection sensor, and is the accumulative sum of the noise interference degrees of all noise event sets corresponding to the frame signal under the th detection sensor, where

[0011] is the ceiling function. Preferably, calculating the droplet estimation value and combining it with the sound source estimation value to obtain the modal decomposition layer number of each frame signal under each detection sensor includes: Based on the occurrence period of each acoustic emission event in each frame signal under each detection sensor, the total energy of all signal amplitudes of the signal corresponding to the occurrence period in the same frame signal under the guard sensor directly above is denoted as the energy eigenvalue; Obtaining the segmentation threshold of the energy eigenvalues corresponding to all acoustic emission events in each frame signal under the guard sensor, denoted as the second segmentation threshold; counting the number of acoustic emission events with energy eigenvalues greater than the second segmentation threshold as the droplet estimation value of each frame signal under the guard sensor;

[0012] The modal decomposition layer number is the sum of the sound source estimation value of each frame signal under each detection sensor and the droplet estimation value of the same frame signal of the guard sensor directly above.

[0013] In a second aspect, an embodiment of the present application further provides a leak detection device for a pressure vessel. The device includes a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps of any one of the above-mentioned leak detection methods for a pressure vessel.

[0014] In a third aspect, an embodiment of the present application further provides a leak detection system for a pressure vessel. There is a signal analysis module in the system, and the sub-modules included in the module are: The filtering processing sub-module is used to obtain the reconstructed acoustic emission signal, including: framing the acoustic emission signal, and extracting the upper envelope of each frame of signal under each detection sensor; judging the acoustic emission events that meet the acoustic emission event judgment conditions according to the deviation of the liquid level height at all times, as well as the magnitude of the signal amplitude and the distribution of extreme points in the upper envelope; analyzing the energy intensity and the duration of the event corresponding to each acoustic emission event in each frame of signal of each detection sensor, and combining the frequency distribution of the signal in the frequency domain to form a feature vector, classifying the acoustic emission events of the same frame of signal under each detection sensor and its adjacent detection sensors, and obtaining each event set corresponding to each frame of signal under each detection sensor; determining the noise interference degree of each event set through the difference of the feature vectors between different acoustic emission events in each event set and the number of acoustic emission events, and obtaining the noise event set and the non-noise event set; determining the sound source estimation value of each frame of signal under each detection sensor based on the number of the noise event set, the number of the non-noise event set and the noise interference degree; analyzing the energy intensity of the signal corresponding to each acoustic emission event in each frame of signal under the guard sensor, calculating the droplet estimation value, and combining the sound source estimation value to obtain the modal decomposition layer number of each frame of signal under each detection sensor, and obtaining all modal components by performing modal decomposition on each frame of signal under each detection sensor; obtaining the modal components corresponding to the non-noise through the correlation between each modal component and the other modal components, and performing signal reconstruction on all the modal components corresponding to the non-noise; The time delay estimation sub-module is used to estimate the time delay of the reconstructed acoustic emission signal; The positioning algorithm sub-module is used to locate the leakage position of the container bottom plate through the reconstructed acoustic emission signal and the result of the time delay estimation.

[0015] This application has at least the following beneficial effects: In this application, by extracting the upper envelope of each frame of signal under each detection sensor and analyzing the change of the signal amplitude after the maximum point in the upper envelope, the acoustic emission events that meet the determination conditions of acoustic emission events are determined. Its beneficial effect lies in considering the dynamic setting of the effective threshold value of the acoustic emission signal according to the liquid level height. When the liquid level height is low, the detection accuracy of the weak acoustic emission signal during the leakage of the container bottom plate is improved. Furthermore, a feature vector is formed to classify the acoustic emission events of the same frame of signal under each detection sensor and its adjacent detection sensors, and each event set corresponding to each frame of signal under each detection sensor is obtained. Its beneficial effect lies in considering that the probability of the adjacent detection sensors detecting the same acoustic emission source signal is relatively high. By classifying the acoustic emission events in the same frame of signal under the adjacent detection sensors, the events that may be of the same type are grouped together. Then, the noise interference degree of each event set is determined, and the noise event set and the non-noise event set are obtained. Its beneficial effect lies in explaining the difference situation of the signals between each detection sensor and its adjacent sensors through the number of acoustic emission events belonging to each detection sensor in the event set, so as to reflect the significant situation of the signals of each detection sensor being interfered by noise aliasing; the sound source estimation value of each frame of signal under each detection sensor is determined, the droplet estimation value is calculated, and the modal decomposition layer number of each frame of signal under each detection sensor is obtained. Its beneficial effect lies in estimating the acoustic emission source appearing in each frame of signal under each detection sensor by analyzing the number of the noise event set and the non-noise event set. In addition, according to the significant situation of the energy of the signal corresponding to each acoustic emission event in each frame of signal under the guard sensor, the acoustic emission events of the high-energy signals appearing in each frame of signal under the guard sensor are estimated to evaluate the interference effect of the high-energy signals in the guard sensor on the lower detection sensors, so as to dynamically adjust the decomposition layer number of the modal decomposition algorithm according to the signal hybrid superposition situation of the acoustic emission sources of the bottom plate and the top in the acoustic emission signal of the detection sensor, so as to avoid under-decomposition or over-decomposition of the mode of the acoustic emission signal; through the correlation between each modal component and the other modal components, the modal components corresponding to non-noise are obtained, and signal reconstruction is performed on all the modal components corresponding to non-noise. Through the reconstructed acoustic emission signal, the leakage position of the pressure vessel is located. Its beneficial effect lies in eliminating the modes corresponding to the noise with weak correlation by analyzing the correlation between the modal components and performing signal reconstruction, so as to effectively exclude the noise signals generated by the droplet fall, reduce the interference effect of the detection sensor for bottom plate leakage detection, improve the filtering effect of the acoustic emission signal of the detection sensor, and improve the positioning accuracy of the leakage point of the pressure vessel through the acoustic emission signal of the detection sensor. Description of the Drawings

[0016] The following further elaborates in detail on a pressure vessel leakage detection method according to this application with reference to the drawings.

[0017] Figure 1 The flowchart of steps of a leak detection method for pressure vessels provided by an embodiment of the present application; Figure 2 The deployment schematic diagram of acoustic emission sensors on the outer wall of the pressure vessel provided by an embodiment of the present application; Figure 3 The flowchart of steps of a method for obtaining the number of modal decomposition layers provided by an embodiment of the present application; Figure 4 The block diagram of a leak detection system for pressure vessels provided by an embodiment of the present application. Detailed implementation manners

[0018] In order to make the objectives, technical solutions and advantages of the present application clearer, a leak detection method, device and system for pressure vessels proposed by the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0019] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present application belongs.

[0020] Please refer to Figure 1 , which shows the flowchart of steps of a leak detection method for pressure vessels provided by an embodiment of the present application. The method includes the following steps: Step 1: Obtain the acoustic emission signals of each detection sensor on the outer wall of the pressure vessel and its guard sensor directly above, as well as the liquid level height at each moment.

[0021] A pressure vessel generally refers to a sealed device that can withstand the pressure of fluid media, such as various pressure cylinders, glove boxes, and storage tanks. With the acceleration of the industrialization process, pressure vessels play an indispensable role in key industries such as petroleum, chemical industry, and electric power. Due to long-term use, the bottom plate of the pressure vessel will be corroded, rusted, impacted, etc., causing damage to the container and triggering dangers such as leakage.

[0022] When detecting the sealing performance of a pressure vessel, the traditional method is to detect by opening the tank, but this traditional detection method of shutting down and opening the tank will cause a waste of a large amount of manpower, material resources and financial resources. The online detection technology can detect the corrosion condition of the bottom of the container without opening the tank. The online detection methods mainly include acoustic emission, ultrasonic guided wave and robot technology, etc. Among them, the acoustic emission detection technology is a non-destructive testing method for leak detection based on the weak acoustic wave signals generated by tiny displacements inside the material. In the leak detection of sealed pressure vessels, the acoustic emission detection technology is widely used.

[0023] When there are corrosion defects on the bottom plate of a pressure vessel, the material strength decreases, and under the action of the liquid level, local micro-deformations occur, resulting in the peeling and falling off of corrosion products, generating acoustic emission signals. Therefore, when leakage occurs, the flow of the medium will generate continuous acoustic emission signals.

[0024] Due to the temperature difference between the inside and outside of the pressure vessel, liquid droplets will form at the top of the vessel. When the liquid droplets fall onto the liquid surface, acoustic emission signals will also be generated, and these signals are very similar to the signals generated by the leakage of the bottom plate of the vessel. General signal acquisition methods cannot distinguish the acoustic emission signals from the defects of the bottom plate of the vessel and the liquid droplets at the top of the vessel. To solve this problem, guard sensors and detection sensors are respectively deployed on the outer wall of the vessel, and the guard sensors are used to filter out the acoustic emission noise signals from the liquid droplets at the top of the vessel.

[0025] Two groups of acoustic emission sensors are arranged on the outer wall surface of the vessel. One group receives the acoustic emission signals generated by the impact of the liquid droplets at the top of the vessel on the liquid surface, that is, the guard sensors, and the other group receives the acoustic emission signals from the bottom plate of the vessel, denoted as the detection sensors; In this embodiment, at a height of 0.75 m from the bottom plate on the outer wall surface of the vessel, a group of acoustic emission sensors are evenly deployed annularly along the outer wall surface. Among them, the number of a group of acoustic emission sensors is 8, and this group of acoustic emission sensors is used as the detection sensors to receive the acoustic emission signals of the bottom plate of the vessel; Secondly, at a position where the height from the bottom plate of the vessel is 0.85 times the preset rated operating height of the vessel, another group of acoustic emission sensors are evenly deployed annularly along the outer wall surface, and this group of acoustic emission sensors is used as the guard sensors. Among them, the number of guard sensors is also 8. As other implementation manners, the implementer can deploy the acoustic emission sensors according to the actual situation. Therefore, by converting the analog signals collected by the acoustic emission sensors into digital signals, the acoustic emission signals collected by each detection sensor and each guard sensor are obtained; In this embodiment, the guard sensors are directly above the detection sensors. The center frequency of the selected sensors is 140 kHz, the frequency range is 50 - 200 kHz, and the sensitivity of the selected acoustic emission sensors is >65, and the acquisition frequency is 1 MHz. As other implementation manners, the implementer can set according to the actual situation; Secondly, the rated operating height refers to the maximum liquid level height that the vessel is allowed to reach under the design and normal operating conditions. The maximum liquid level height is 0.8 or 0.9 times the height of the vessel. In this embodiment, the height of the vessel is 21.8, and 0.8 times the height of the vessel is used as the rated operating height.

[0026] In this embodiment, the deployment schematic diagram of the acoustic emission sensors on the outer wall surface of the pressure vessel provided in this embodiment is as Figure 2 shown, where 201 are the detection sensors, 202 are the guard sensors, 203 is the liquid level of the vessel, and 204 is the outer wall surface of the vessel.

[0027] When a pressure vessel leaks, the liquid medium flows out from the leakage point and generates an acoustic emission signal. Due to the huge area of the container bottom plate, the acoustic emission signal at the leakage position attenuates. If the leakage position is recorded as the acoustic emission source, when the acoustic emission sensor receives the signal at the acoustic emission source, interference signals will inevitably appear, resulting in a complex waveform of the acoustic emission sensor, masking the signal of the acoustic emission source, and seriously affecting the accuracy of the acoustic emission signal at the leakage point and the time-delay estimation for leakage localization. Therefore, it is necessary to denoise the acoustic emission signal and remove the interference signals.

[0028] Since the acoustic emission detection of the container depends on the liquid level, a radar level gauge is installed at the central position of the container top to ensure that the radar wave can propagate vertically downward, reducing the interference of the reflected wave. The liquid level height of the container is monitored in real time through the radar level gauge, and the liquid level height at each moment is collected. Among them, the sampling frequency of the radar level gauge is 10Hz. As other implementation manners, the implementer can set it according to the actual situation.

[0029] Thus, according to the signal acquisition module, the acoustic emission signals of each detection sensor and the guard sensor directly above it, as well as the liquid level height at each moment, are obtained.

[0030] Step 2: Frame the acoustic emission signal and extract the upper envelope of each frame of the signal under each detection sensor; judge the acoustic emission events that occur according to the deviation of the liquid level height at all moments, as well as the magnitude of the signal amplitude and the distribution of the extreme points in the upper envelope.

[0031] Secondly, to facilitate subsequent feature analysis of the signal at the acoustic emission source of the container bottom plate, the signals collected by the detection sensor and the guard sensor are respectively framed. The acoustic emission signal is divided into multiple frame segments, and each frame of the signal of each detection sensor and each frame of the signal of the guard sensor directly above each detection sensor are obtained respectively. In this embodiment, the frame length of each frame during the framing process is 100ms, that is, the time length of each frame of the signal is 100ms. As other implementation manners, the implementer can set it according to the actual situation.

[0032] During the acoustic emission detection process, in addition to the signals generated by phenomena such as container bottom leakage, substrate cracking, and stress corrosion cracking, there will also be various background noises and other interference signals. For example, interference generated by liquid droplets dripping from the top of the container. Among them, the change in liquid level height will affect the propagation and amplitude of the acoustic emission signals collected by the detection sensor. The pressure of the liquid on the bottom and side walls of the container increases with the increase in liquid level height. When the liquid level is low, the pressure intensity of the liquid on the container bottom is low; while when the liquid level is high, the pressure intensity of the liquid on the container bottom is high. During the acoustic emission detection process, the valve of the pressure vessel will be closed, so the change in liquid level height is extremely small. By setting an effective threshold value based on the liquid level height, interference signals can be preliminarily filtered and acoustic emission events can be identified. Specifically: The ratio of the mean value of the liquid level height at all times to the preset rated operating height is denoted as the liquid level ratio; In this embodiment, the preset rated operating height refers to the maximum liquid level height of the pressure vessel. Then the preset rated operating height is 0.8 times the height of the container. Among them, the height of the container is 21.8 meters, and the height of the container is obtained from the container production materials. As other implementation methods, the implementer can determine it according to the actual situation.

[0033] The method for determining the effective threshold value is: Among them, is the effective threshold value, is the preset reference value, is the liquid level ratio, is the preset reference ratio, is the exponential function with the natural constant as the base.

[0034] In this embodiment, the preset reference value is set to 15 dB, and the preset reference ratio is set to 0.85. Among them, the selection of the preset reference ratio is based on industry standards. In the "Monitoring Methods and Implementation Guidelines for Corrosion of Atmospheric Storage Tanks", it is stipulated that the liquid level for on-line acoustic emission detection of the storage tank bottom plate should be located at 85%-105% of the highest operating liquid level. Among them, in this embodiment, the pressure vessel is taken as an example of a storage tank, and 0.85 is selected as the preset reference ratio. As other implementation methods, the implementer can set it according to the actual situation.

[0035] It should be noted that if , it means that the pressure intensity of the liquid in the pressure vessel is low and the acoustic emission signal weakens. To more sensitively detect these weaker acoustic emission signals, the effective threshold value needs to be reduced.

[0036] Further, when the bottom plate of the pressure vessel is under leak detection, leakage is a continuous process that generates continuous acoustic emission signals. Acoustic emission signals such as substrate cracking and stress corrosion cracking of the bottom plate are mutant signals, presenting as obvious spikes or pulses. By analyzing the change trend of each frame of the signal and combining with the effective threshold value, acoustic emission events are determined. Specifically: Adopt the envelope extraction algorithm to extract the upper envelope of each frame of the signal under each detection sensor; In this embodiment, the Hilbert transform algorithm is used to extract the upper envelope. Among them, the Hilbert transform algorithm is a well-known technology and will not be elaborated here. As other implementation manners, implementers can adopt other methods of existing technologies, such as the cepstrum method, etc. This embodiment does not make special restrictions on this.

[0037] Obtain the maximum points and minimum points of the upper envelope; In this embodiment, the AMPD (Automatic multiscale-based peak detection) algorithm is used to obtain the extreme points. Among them, the AMPD algorithm is a well-known technology and will not be elaborated here.

[0038] Starting from any of the maximum points and extending backward, the moment when the signal amplitude after the starting point in the upper envelope is less than the effective threshold value, and the first minimum point after the starting point are both recorded as critical points; Record the critical point closest to the starting point as the termination point; For each frame of the signal under each detection sensor, the time period between the starting point corresponding to each maximum point and the termination point is recorded as one acoustic emission event; It should be noted that taking the maximum point as the starting point reflects the start of a potential acoustic emission event. The acoustic emission event may include the acoustic emission signal of the container bottom plate leakage, or may also include mutant acoustic emission signals such as substrate cracking and stress corrosion cracking.

[0039] Thus, the acoustic emission events are obtained.

[0040] Step 3: Analyze the energy intensity of the signal corresponding to each acoustic emission event and the duration of the event for each frame of the signal of each detection sensor, and combine with the frequency distribution of the signal in the frequency domain to form a feature vector. Classify the acoustic emission events of the same frame of the signal under each detection sensor and its adjacent detection sensors, and obtain each event set corresponding to each frame of the signal of each detection sensor; Determine the noise interference degree of each event set through the difference situation of the feature vectors between different acoustic emission events in each event set and the number situation of the acoustic emission events, and obtain the noise event set and the non-noise event set.

[0041] Secondly, the continuous acoustic emission signals caused by leakage of the container bottom plate have a longer duration than the abrupt acoustic emission signals such as substrate cracking and stress corrosion cracking, and the overall intensity of the continuous signals is relatively high during the duration. Therefore, analyze the duration of the signals under each acoustic emission event and their signal energy, and calculate the evaluation value, specifically as follows: Calculate the duration of the signals corresponding to each acoustic emission event within each frame of signals under each detection sensor, denoted as the duration; It should be noted that for the convenience of understanding, for the r-th frame of signals under the m-th detection sensor, if an acoustic emission event occurs during this time period, the signals of the r-th frame of signals under the m-th detection sensor during this time period are denoted as the signals corresponding to the acoustic emission event within the r-th frame of signals under the m-th detection sensor; for the signals within the r-th frame of signals under the guard sensor directly above the m-th detection sensor during this time period, they are denoted as the signals corresponding to the acoustic emission event within the r-th frame of signals under the guard sensor directly above.

[0042] It should be noted that the duration is the time interval between the starting point and the ending point under each acoustic emission event.

[0043] Calculate the product of the energy of all signal amplitudes within the signals corresponding to each acoustic emission event within each frame of signals under each detection sensor and the duration as the evaluation value of each acoustic emission event; In this embodiment, the calculation of the energy of all signal amplitudes within the signals corresponding to each acoustic emission event is a well-known technology, and the energy is calculated by summing the squares of all signal amplitudes within the signals corresponding to each acoustic emission event.

[0044] It should be noted that the larger the duration, the longer the event lasts, indicating stronger signal continuity. Secondly, the greater the total energy, the higher the overall intensity of the signal during the duration, and the higher the possibility of belonging to continuous acoustic emission signals. The larger the obtained evaluation value, the more likely it is that the acoustic emission event is a continuous signal caused by bottom plate leakage.

[0045] Furthermore, classify the acoustic emission events by analyzing the characteristic differences of the signals corresponding to the acoustic emission events in the same frame of signals between adjacent detection sensors, specifically as follows: Conduct frequency domain analysis on the signals corresponding to each acoustic emission event within each frame of signals under each detection sensor to obtain a spectrogram, and denote the frequency corresponding to the maximum energy in the spectrogram as the main frequency; In this embodiment, the Fourier transform algorithm is used to obtain the spectrogram. Among them, the Fourier transform algorithm is a well-known technology and will not be elaborated here.

[0046] The frequency corresponding to the center value after integrating all the energies in the spectrogram is denoted as the center frequency; Count the number of times the signal amplitude is less than the effective threshold value for each acoustic emission event in each frame of the signal under each detection sensor, and denote it as the ringing count. Denote the ratio of the ringing count to the duration as the average frequency; It should be noted that the ringing count reflects the activity of the signal in this event and may correspond to more intense acoustic emission activities. Among them, continuous signals such as leakage events may have a relatively large number of ringing counts because their signal duration is long, the energy is relatively stable, and they fluctuate frequently near the effective threshold value; while mutant signals such as substrate cracking events may have relatively few ringing counts, but the signal intensity may be high and the amplitude may be large.

[0047] Normalize the main frequency, the center frequency, the average frequency, and the evaluation value respectively, and form a feature vector; In this embodiment, the maximum-minimum normalization method is used for normalization. Among them, the maximum-minimum normalization method is a well-known technology and will not be elaborated here. As other implementation manners, implementers can adopt other methods of existing technologies, for example, the Z-score normalization method. This embodiment does not make special restrictions on this.

[0048] Denote the remaining detection sensors adjacent to any one detection sensor as adjacent sensors; Cluster the feature vectors corresponding to all acoustic emission events of the same frame signal under the any one detection sensor and all adjacent sensors to obtain a plurality of clustering clusters; Denote all the acoustic emission events belonging to the same clustering cluster as each event set corresponding to each frame of the signal under the any one detection sensor; In this embodiment, the Density Peak Clustering (DPC) algorithm is used for clustering. Among them, the DPC algorithm is a well-known technology and will not be elaborated here. As other implementation manners, implementers can adopt other methods of existing technologies, for example, the hierarchical clustering algorithm, etc. This embodiment does not make special restrictions on this.

[0049] It should be noted that the acoustic emission events in the clustering cluster have similar feature vectors, so it is more likely to correspond to the acoustic emission sources of the container bottom plates at the same position and of the same type. Therefore, each event set corresponds to the same type of acoustic emission events.

[0050] Furthermore, as a dynamic non-destructive testing method, acoustic emission detection technology can be masked or distorted by background noise and environmental interference during propagation, affecting the accuracy of event detection and thus the positioning result. Since the distance between adjacent detection sensors is relatively close, the acoustic emission attenuation degree is low, and the acoustic emission event correlation is high. If the noise interference degree of acoustic emission events is low, they are likely to be classified into the same category. By analyzing the number of sensors to which the acoustic emission events in each event set corresponding to any one of the detection sensors belong, the noise interference degree is calculated as follows: The mean of the distances between the feature vectors of all any two acoustic emission events belonging to any one of the detection sensors in each event set is used as the dispersion degree of each event set corresponding to each frame of signal under any one of the detection sensors; In this embodiment, the distance is measured by the DTW distance between the feature vectors of any two acoustic emission events belonging to any one of the detection sensors in each event set. As other implementation manners, implementers can adopt other methods in the prior art, such as Euclidean distance, etc. This embodiment does not make special restrictions on this.

[0051] Count the number of all acoustic emission events in each event set, denoted as the total number; calculate the ratio of the number of acoustic emission events belonging to any one of the detection sensors in each event set to the total number, denoted as the relative ratio; The normalized result of the product of the relative ratio and the dispersion degree is used as the noise interference degree of each event set corresponding to each frame of signal under any one of the detection sensors; In this embodiment, taking the th detection sensor, the th frame of signal, and the th event set as an example, the calculation formula for its noise interference degree is: where, is the noise interference degree of the th detection sensor, the th frame of signal, and the th event set, is the number of acoustic emission events belonging to the th frame of signal and the th event set and belonging to the th detection sensor, is the number of all acoustic emission events in the th frame of signal and the th event set, is the dispersion degree of the th detection sensor, the th frame of signal, and the th event set, is a normalization function. In this embodiment, the softmax function is used for normalization processing. Among them, the softmax function is a well-known technology and will not be elaborated here. As other implementation manners, implementers can adopt other methods in the prior art. For example, the sigmoid function, etc. This embodiment does not make special restrictions on this. Among them, is a relative ratio.

[0052] It should be noted that the larger the relative ratio, it indicates that the number of acoustic emission events belonging to the th detection sensor in the th event set is larger. Then, the signal feature differences between the adjacent sensor and the th detection sensor for the same acoustic emission source may be larger. Ideally, for the same type of acoustic emission events, the number of acoustic emission events belonging to the th detection sensor and the adjacent sensor in the event set should be relatively close. If the number of acoustic emission events belonging to the th detection sensor is larger, it reflects that there is a greater possibility of large noise aliasing in the th event set corresponding to the th type of event; the larger the dispersion degree, it indicates that the correlation between the

[0053] Next, based on the noise interference degree, the event sets are screened. Specifically: Obtain the segmentation threshold of the noise interference degree of all event sets corresponding to each frame of signal under any detection sensor, denoted as the first segmentation threshold; In this embodiment, the maximum inter-class variance method is used to obtain the segmentation threshold. Among them, the maximum inter-class variance method is a well-known technology and will not be elaborated here. As other implementation manners, implementers can adopt the cross-validation method, etc. This embodiment does not make special restrictions on this.

[0054] The event sets with the noise interference degree of each frame of signal under any detection sensor greater than the first segmentation threshold are denoted as noise event sets, and vice versa, denoted as non-noise event sets; It should be noted that the acoustic emission events in the noise event sets are more likely to have large noise aliasing, and the degree of signal interference by noise is greater.

[0055] So far, all noise event sets and non-noise event sets corresponding to each frame of signal under each detection sensor are obtained.

[0056] Step 4: Determine the sound source estimation value of each frame of signal under each detection sensor based on the number of noise event sets, the number of non-noise event sets, and the noise interference degree; analyze the energy intensity of the signal corresponding to each acoustic emission event in each frame of signal under the guard sensor, calculate the droplet estimation value, and combine the sound source estimation value to obtain the modal decomposition layer number of each frame of signal under each detection sensor.

[0057] Furthermore, determine the sound source estimation value based on the number of events in the noise event set and the non-noise event set, specifically: where, is the sound source estimation value of the th frame of signal under the th detection sensor, is the number of all non-noise event sets corresponding to the th frame of signal under the th detection sensor, is the number of all noise event sets corresponding to the th frame of signal under the th detection sensor, is the cumulative sum of the noise interference degrees of all noise event sets corresponding to the th frame of signal under the th detection sensor, is the ceiling function.

[0058] It should be noted that for the non-noise event set, the acoustic emission events of the acoustic emission source have less noise mixing and correspond to a bottom plate acoustic emission source; for the noise event set, the acoustic emission events of the acoustic emission source have more noise mixing and there may be more noise interference sources. To avoid insufficient variational mode decomposition, the larger the sound source estimation value at the bottom plate of the container, the more reasonable the number of decomposed modes during subsequent modal decomposition.

[0059] Secondly, in the leakage detection of the container, when the volatilization characteristic of the liquid phase medium is strong, there will be a gas phase region at the top of the container, which will condense on the inner wall of the top of the container. As the condensation point continuously increases, a droplet falling effect will be formed and drip back onto the liquid phase medium. The signal energy generated by the droplet falling is high and has a great influence on acoustic emission detection. Once these noise signals are received by the detection sensor, it is easy to cause mispositioning of the leakage situation of the container bottom plate, seriously affecting the accuracy of the bottom plate inspection result. It is necessary to exclude these noise signals that have nothing to do with the container bottom plate.

[0060] Therefore, analyze the energy significance of the signal corresponding to each acoustic emission event in each frame of signal under the guard sensor, and calculate the droplet estimation value, specifically: According to the occurrence period of each acoustic emission event in each frame of signal under each detection sensor, the signal corresponding to the same occurrence period in the same frame of signal under the guard sensor directly above is recorded as the signal corresponding to each acoustic emission event in each frame of signal under the guard sensor; Calculate the energy of all signal amplitudes of the signal corresponding to each acoustic emission event in each frame of signal under the guard sensor directly above each detection sensor, and record it as the energy eigenvalue; Obtain the segmentation threshold of the energy eigenvalues corresponding to all acoustic emission events in each frame of signal under the guard sensor, and record it as the second segmentation threshold; In this embodiment, the Otsu method is used to obtain the segmentation threshold. The Otsu method is a well-known technology and will not be elaborated here. As other implementation manners, the implementer can adopt the cross-validation method, etc. This embodiment does not make special restrictions on this.

[0061] Count the number of acoustic emission events with energy eigenvalues greater than the second segmentation threshold in each frame of signal under the guard sensor directly above each detection sensor, and use it as the droplet estimation value of each frame of signal under the guard sensor; It should be noted that the larger the energy eigenvalue, the more likely the acoustic emission event is a high-energy signal generated by droplet falling. Then the droplet estimation value reflects the number of significant high-energy signals when droplet falling appears at the top of the container in each frame of signal. Only these high-energy signals may be received by the lower detection sensor and interfere with the lower detection sensor.

[0062] Furthermore, based on the sound source estimation value and the droplet estimation value, the number of variational mode decomposition layers is obtained. Specifically: Take the sum of the sound source estimation value of each frame of signal under each detection sensor and the droplet estimation value of the same frame of signal under the guard sensor directly above as the number of variational mode decomposition layers of each frame of signal under each detection sensor; It should be noted that the signals corresponding to the acoustic emission events with energy eigenvalues greater than the second segmentation threshold have higher energy. Only these high-energy signals may be received by the lower detection sensor. The larger the number of variational mode decomposition layers, the more acoustic emission sources there are at the bottom and top of the container. In order to avoid mixing the acoustic emission signal modes between acoustic emission sources or missing effective modes and causing under-decomposition, when decomposing the acoustic emission signal of the detection sensor by the variational mode decomposition algorithm, the decomposition layer number of the variational mode decomposition algorithm should be larger. Among them, the flow chart of the method for obtaining the number of variational mode decomposition layers provided in the embodiment of the present application is as Figure 3 shown.

[0063] So far, the number of variational mode decomposition layers of each frame of signal under each detection sensor is obtained.

[0064] Step 5: Based on the number of modal decomposition layers, perform modal decomposition on each frame of signal under each detection sensor to obtain all modal components; based on the correlation between each modal component and the remaining modal components, obtain the modal components corresponding to non-noise, perform signal reconstruction on all modal components corresponding to non-noise, and locate the leakage position of the pressure vessel through the reconstructed acoustic emission signal.

[0065] Take the number of modal decomposition layers as the decomposition layer number of the modal decomposition algorithm, and perform modal decomposition on each frame of signal under each detection sensor to obtain multiple modal components; where the number of modal components is the same as the number of modal decomposition layers. In this embodiment, the variational mode decomposition algorithm is used for modal decomposition. The variational mode decomposition algorithm is a well-known technology and will not be elaborated here.

[0066] Secondly, according to the characteristics that the correlation between the acoustic emission signals received by the detection sensors from the same acoustic emission source is strong, and the correlation between noises is weak, analyze the correlation between each modal component and all the remaining modal components, screen the modal components, and eliminate the modes corresponding to the noise. Specifically: Calculate the correlation degree between each modal component and the remaining modal components. In this embodiment, the correlation degree is calculated by taking the reciprocal of the DTW distance between each modal component and the remaining modal components. As other implementation manners, implementers can adopt other methods of the prior art, such as cosine similarity, etc. This embodiment does not make special limitations on this.

[0067] Count the number of cases where the correlation degree between each modal component and all the remaining modal components is less than a preset threshold, and record it as the judgment coefficient. Select the modal components with the judgment coefficient less than the preset number threshold, and record them as the modal components corresponding to non-noise. In this embodiment, the preset threshold is set to 1, and the preset number threshold is set to 5. As other implementation manners, implementers can set them according to the actual situation.

[0068] Perform signal reconstruction on all modal components corresponding to non-noise of each frame of signal under each detection sensor respectively to obtain the reconstructed frame of signal, and splice all the reconstructed frames of signal under each detection sensor to obtain the reconstructed acoustic emission signal of each detection sensor. Among them, the reconstructed acoustic emission signal eliminates the interference signal of the noise. Therefore, the reconstructed acoustic emission signal is the acoustic emission signal after filtering is completed. It should be noted that the process of signal reconstruction is a well-known technology and will not be elaborated here.

[0069] Based on the reconstructed acoustic emission signal, time delay estimation is carried out through an improved quadratic correlation algorithm, and the leakage position of the container bottom plate is located through an overdetermined positioning algorithm based on the sequential quadratic programming algorithm to complete the leak detection of the container.

[0070] It should be noted that the improved quadratic correlation algorithm and the overdetermined positioning algorithm based on the sequential quadratic programming algorithm are well-known technologies and will not be elaborated here.

[0071] The embodiment of the present application also provides a leak detection device for a pressure vessel, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps of any one of the above-mentioned leak detection methods for a pressure vessel.

[0072] Based on the same inventive concept as the above method, the embodiment of the present application also provides a leak detection system for a pressure vessel. The system includes: An acoustic emission sensor array for deploying acoustic emission sensors on the outer wall surface of the pressure vessel; A signal conditioning module for preprocessing the collected signals. Specifically: The acoustic emission sensor has no built-in signal conditioning circuit and outputs a weak high-impedance charge signal, which is difficult to collect. By connecting the signal conditioning module to the acoustic emission sensor array, it is used to amplify and process the weak electrical signals collected by the acoustic emission sensors to improve the signal-to-noise ratio of the signals for subsequent analysis.

[0073] In this embodiment, AD8606 is selected as the chip of the charge amplifier. At the same time, a compensation circuit composed of capacitor C1 is connected in the operational amplifier to eliminate self-oscillation. The filtering process uses a second-order band-pass filter circuit. The bandwidth B of the band-pass filter is set to 10, and the center frequency is 140 kHz. As other implementation manners, the implementer can set it according to the actual situation.

[0074] A signal acquisition module for acquiring the acoustic emission signals of each detection sensor on the outer wall surface of the pressure vessel and the guard sensors directly above them, as well as the liquid level height at each moment; A signal transmission module. The signal transmission module is connected to the signal acquisition module and supports real-time signal upload to the host computer through the USB2.0 high-speed transmission protocol; A signal analysis module for processing the received acoustic emission signals to detect the leakage position of the container, specifically including the following sub-modules: The filtering processing sub-module is used to obtain the reconstructed acoustic emission signal, including: framing the acoustic emission signal, extracting the upper envelope of each frame of signal under each detection sensor; judging the acoustic emission events that meet the acoustic emission event judgment conditions according to the deviation of the liquid level height at all times, as well as the magnitude of the signal amplitude and the distribution of extreme points in the upper envelope; analyzing the energy intensity and the duration of occurrence of each acoustic emission event corresponding to the signal in each frame of signal of each detection sensor, combining with the frequency distribution of the signal in the frequency domain, forming a feature vector, classifying the acoustic emission events of the same frame of signal under each detection sensor and its adjacent detection sensors, and obtaining each event set corresponding to each frame of signal under each detection sensor; determining the noise interference degree of each event set through the difference of the feature vectors between different acoustic emission events in each event set and the number of acoustic emission events, and obtaining the noise event set and the non-noise event set; determining the sound source estimation value of each frame of signal under each detection sensor based on the number of the noise event set, the number of the non-noise event set and the noise interference degree; analyzing the energy intensity of each acoustic emission event corresponding to the signal in each frame of signal of the guard sensor, calculating the droplet estimation value, and combining with the sound source estimation value to obtain the modal decomposition layer number of each frame of signal under each detection sensor, and obtaining all modal components by performing modal decomposition on each frame of signal under each detection sensor; obtaining the modal components corresponding to the non-noise through the correlation between each modal component and the remaining modal components, and performing signal reconstruction on all modal components corresponding to the non-noise; The time delay estimation sub-module is used to estimate the time delay of the reconstructed acoustic emission signal; In this embodiment, the improved quadratic correlation algorithm is used to estimate the time delay of the acoustic emission signal processed by the filtering processing sub-module, so as to solve the problems of poor anti-noise ability, large calculation amount and low estimation accuracy for non-linear and non-periodic signals in time delay estimation in acoustic emission source localization.

[0075] The positioning algorithm sub-module is used to locate the leakage position of the container bottom plate through the reconstructed acoustic emission signal and the result of time delay estimation.

[0076] In this embodiment, based on the overdetermined positioning algorithm (Overdetermined Positioning Algorithm Based on Sequential Quadratic Algorithm) of the sequential quadratic programming algorithm, the leakage position of the container bottom plate is located through the acoustic emission signal processed by the filtering processing sub-module and the result of time delay estimation.

[0077] It should be noted that the improved quadratic correlation algorithm and the overdetermined positioning algorithm based on the sequential quadratic programming algorithm are well-known technologies and will not be elaborated here.

[0078] Among them, a block diagram of a leak detection system for pressure vessels provided by an embodiment of the present application is as shown in Figure 4 the figure.

[0079] It should be understood that although Figure 1 each step in the flowchart of Figure 1 is shown in sequence according to the indication of the arrow, these steps are not necessarily executed in the order indicated by the arrow. Unless there is a clear indication in this article, the execution of these steps has no strict order limit, and these steps can be executed in other orders. Moreover,

[0080] The technical features of the above embodiments can be combined arbitrarily. For the sake of concise description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope recorded in this specification. The above embodiments only represent several implementation manners of the present application, and their descriptions are relatively specific and detailed, but they should not be construed as a limitation to the present application. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can still be made. Therefore, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present application without departing from the content of the technical solution of the present application all belong to the protection scope of the technical solution of the present application.

Claims

1. A leak detection method for pressure vessels, characterized in that, The method includes the following steps: Acquire the acoustic emission signals of each detection sensor on the outer wall surface of the pressure vessel and the guard sensor at the position directly above it, respectively, as well as the liquid level height at each moment; Frame the acoustic emission signals, and extract the upper envelope of each frame of signals under each detection sensor; judge the occurring acoustic emission events according to the deviation of the liquid level height at all moments, as well as the magnitude of the signal amplitude and the distribution of extreme points in the upper envelope; Analyze the energy intensity of the signals corresponding to each acoustic emission event and the duration of the event under each frame of signals of each detection sensor, and combine with the frequency distribution of the signals in the frequency domain to form a feature vector. Classify the acoustic emission events of the same frame of signals under each detection sensor and its adjacent detection sensors, and obtain each event set corresponding to each frame of signals under each detection sensor; determine the noise interference degree of each event set according to the difference of the feature vectors between different acoustic emission events in each event set and the number of acoustic emission events, and divide the event set into a noise event set and a non-noise event set; Based on the number of the noise event set, the number of the non-noise event set and the noise interference degree, determine the sound source estimation value of each frame of signals under each detection sensor; analyze the energy intensity of the signals corresponding to each acoustic emission event in each frame of signals under the guard sensor, calculate the droplet estimation value, and combine with the sound source estimation value to obtain the modal decomposition layer number of each frame of signals under each detection sensor, and perform modal decomposition on each frame of signals under each detection sensor to obtain all modal components; obtain the modal components corresponding to the non-noise through the correlation between each modal component and the remaining modal components, and after signal reconstruction of all the modal components corresponding to the non-noise, locate the leakage position of the pressure vessel.

2. The leak detection method for pressure vessels according to claim 1, characterized in that, The judgment of the occurring acoustic emission events includes: Denote the ratio of the average value of the liquid level height at all moments to the preset rated operation height as the liquid level ratio; Effective threshold value The calculation formula is as follows: , where is a preset reference value, is the liquid level ratio, is the preset reference ratio, is the exponential function with the natural constant as the base; Obtain the maximum value points and minimum value points of the upper envelope; starting from any maximum value point and extending backward, denote the moment corresponding to the first signal amplitude less than the effective threshold value after the starting point in the upper envelope and the first minimum value point after the starting point as the critical points; denote the critical point closest to the starting point as the termination point; For each frame of signals under each detection sensor, denote the time period between the starting point corresponding to each maximum value point and the termination point as an acoustic emission event.

3. The leak detection method for pressure vessels according to claim 2, characterized in that, The method for obtaining the feature vector is: For the signals corresponding to each acoustic emission event in each frame of signals under each detection sensor, calculate the duration of the signals corresponding to each acoustic emission event, denoted as the duration; Calculate the product of the total energy of all signal amplitudes in the signals corresponding to each acoustic emission event and the duration as the evaluation value of each acoustic emission event; Perform frequency domain analysis on the signals corresponding to each acoustic emission event, obtain the frequency spectrum diagram, denote the frequency corresponding to the maximum energy in the frequency spectrum diagram as the main frequency; denote the frequency corresponding to the central value after integrating all the energies in the frequency spectrum diagram as the central frequency; Statistically calculate the ratio of the number of signal amplitudes within the signal corresponding to each acoustic emission event that are less than the effective threshold value to the duration, and denote it as the average frequency; Normalize the main frequency, the center frequency, the average frequency, and the evaluation value respectively, and form a feature vector.

4. The leak detection method for pressure vessels according to claim 1, characterized in that, The obtaining of each event set corresponding to each frame of signal under each detection sensor includes: clustering the feature vectors corresponding to all acoustic emission events of the same frame signal under each detection sensor and all its adjacent detection sensors, obtaining a plurality of clustering clusters, and denoting all acoustic emission events belonging to the same clustering cluster as each event set corresponding to each frame of signal under each detection sensor.

5. The leak detection method for pressure vessels according to claim 1, characterized in that, The determining of the noise interference degree of each event set and dividing the event set into a noise event set and a non-noise event set includes: For each event set corresponding to each frame of signal of any detection sensor, denote the mean of the distances of the feature vectors of all any two acoustic emission events belonging to the any detection sensor within each event set as the dispersion degree of each event set; Statistically calculate the number of all acoustic emission events within each event set, and denote it as the total number; calculate the ratio of the number of acoustic emission events belonging to the any detection sensor within each event set to the total number, and denote it as the relative ratio; The noise interference degree is the normalized result of the product of the relative ratio and the dispersion degree; Obtain the segmentation threshold of the noise interference degree of all event sets corresponding to each frame of signal of the any detection sensor, and denote it as the first segmentation threshold; denote the event sets with the noise interference degree greater than the first segmentation threshold as the noise event set, and vice versa, denote them as the non-noise event set.

6. The leak detection method for pressure vessels according to claim 1, characterized in that, The estimated sound source value of the frame signal under the th detection sensor is calculated as follows: , where is the number of all non-noise event sets corresponding to the th frame signal under the th detection sensor, is the number of all noise event sets corresponding to the th detection sensor at the th frame signal, is the sum of the noise interference degrees of all noise event sets corresponding to the th frame signal under the th detection sensor, and is the ceiling function.

7. The leak detection method for pressure vessels according to claim 1, characterized in that, The calculating of the droplet estimation value and combining the sound source estimation value to obtain the modal decomposition layer number of each frame of signal under each detection sensor includes: Based on the occurrence time period of each acoustic emission event within each frame of signal under each detection sensor, denote the total energy of all signal amplitudes corresponding to the signal within the same frame signal of the guard sensor directly above during the occurrence time period as the energy eigenvalue; Obtain the segmentation threshold of the energy eigenvalue corresponding to all acoustic emission events within each frame of signal of the guard sensor, and denote it as the second segmentation threshold; statistically calculate the number of acoustic emission events with the energy eigenvalue greater than the second segmentation threshold as the droplet estimation value of each frame of signal of the guard sensor; The modal decomposition layer number is the sum of the sound source estimation value of each frame of signal under each detection sensor and the droplet estimation value of the same frame signal of the guard sensor directly above.

8. A leak detection method for pressure vessels according to claim 1, characterized in that The obtaining process of the modal component corresponding to the non-noise is: calculate the correlation degree between each modal component and the rest of the modal components; Statistically calculate the number of the correlation degrees between each modal component and the rest of all modal components that are less than the preset threshold, and denote it as the judgment coefficient of each modal component; Select the modal components with the judgment coefficient less than the preset number threshold, and denote them as the modal components corresponding to the non-noise.

9. A leak detection device for a pressure vessel, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the steps of a leak detection method for a pressure vessel as described in any one of claims 1-8.

10. A leak detection system for pressure vessels, which is applied to a leak detection method for pressure vessels in claim 1. There is a signal analysis module in the system, and the sub-modules of this module include: A filtering processing sub-module, which is used to obtain the reconstructed acoustic emission signal, including: framing the acoustic emission signal, extracting the upper envelope of each frame of signal under each detection sensor; judging the acoustic emission events that meet the acoustic emission event judgment conditions according to the deviation of the liquid level height at all times, as well as the magnitude of the signal amplitude and the distribution of extreme points in the upper envelope; analyzing the energy intensity of the signal corresponding to each acoustic emission event and the duration of the event occurrence in each frame of signal of each detection sensor, and combining the frequency distribution of the signal in the frequency domain, forming a feature vector, classifying the acoustic emission events of the same frame of signal under each detection sensor and its adjacent detection sensors, and obtaining each event set corresponding to each frame of signal under each detection sensor; determining the noise interference degree of each event set through the difference of the feature vectors between different acoustic emission events in each event set and the number of acoustic emission events, and obtaining the noise event set and the non-noise event set; determining the sound source estimation value of each frame of signal under each detection sensor based on the number of the noise event set, the number of the non-noise event set and the noise interference degree; analyzing the energy intensity of the signal corresponding to each acoustic emission event in each frame of signal under the guard sensor, calculating the droplet estimation value, and combining the sound source estimation value to obtain the modal decomposition layer number of each frame of signal under each detection sensor, and obtaining all modal components by performing modal decomposition on each frame of signal under each detection sensor; obtaining the modal components corresponding to non-noise through the correlation between each modal component and the remaining modal components, and performing signal reconstruction on all modal components corresponding to non-noise; A time delay estimation sub-module, which is used to estimate the time delay of the reconstructed acoustic emission signal; A positioning algorithm sub-module, which is used to locate the leakage position of the container bottom plate through the reconstructed acoustic emission signal and the result of time delay estimation.

Images