A detection alarm method and system for detecting surface defects of a pressure vessel

Abstract

The application discloses a detection alarm method and system for detecting surface defects of a pressure container, and relates to the field of surface defects of pressure containers.The method comprises the following steps: obtaining acoustic wave reflection information of the surface of the pressure container, identifying defect areas existing on the surface of the pressure container based on the acoustic wave reflection information, and judging the characteristic state of the surface defects of the pressure container in combination with the acoustic wave reflection intensity; combining the characteristic state with stress elastoplastic analysis technology, obtaining damage modes of different position points in the defect area on the surface of the pressure container, and determining damage characterization parameters and damage evolution rules according to the damage modes; evaluating the risk value of leakage of the pressure container according to the damage characterization parameters and the damage evolution rules, and generating an alarm prompt based on the risk value and sending the alarm prompt to a management terminal for risk control of the pressure container.The application can accurately depict the damage state of the container, is helpful for predicting the future performance degradation of the container, and is convenient for early warning.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of surface defects of pressure vessels, in particular to a detection alarm method and system for detecting surface defects of pressure vessels. BACKGROUND

[0002] A pressure vessel is a device for storing or transporting high-pressure fluid (gas or liquid), which is usually designed to withstand higher internal pressure than the external environment, so the structure of the pressure vessel must have sufficient strength and durability.

[0003] Pressure vessels withstand extremely high internal and external pressure in many industrial fields, and any defects can cause equipment failure and even major accidents, so pressure vessel defect detection is extremely necessary. Failure of the pressure vessel can not only cause damage to the equipment itself, but also pose a serious threat to the operator and the surrounding environment, and thus regular detection and maintenance of the pressure vessel is of great significance to ensure the safety of industrial production.

[0004] However, the prior art cannot analyze the deformation of the defect of the pressure vessel during the detection of the surface defect of the pressure vessel, and thus cannot identify the initiation and accumulation of fatigue damage at the defect of the pressure vessel, making it difficult to predict the fatigue life of the pressure vessel and the crack propagation process in the surface defect area of the pressure vessel, and unable to provide a basis for the residual life prediction and risk control of the pressure vessel. SUMMARY

[0005] To solve the above problems, the present application provides a detection alarm method and system for detecting surface defects of pressure vessels, which can accurately depict the damage state of the container and help predict its future performance degradation, and facilitate early warning.

[0006] To achieve the above purpose, the present application adopts the following technical solutions:

[0007] In a first aspect, the present application provides a detection alarm method for detecting surface defects of pressure vessels, which comprises:

[0008] Obtaining acoustic wave reflection information of the surface of the pressure vessel, identifying the defect area existing on the surface of the pressure vessel based on the acoustic wave reflection information, and determining the characteristic state of the surface defect of the pressure vessel in combination with the acoustic wave reflection intensity;

[0009] Combining the characteristic state with stress elastoplastic analysis technology to obtain the damage mode of different position points in the surface defect area of the pressure vessel, and determining the damage characterization parameter and damage evolution law according to the damage mode;

[0010] According to the damage characterization parameter and the damage evolution law, the risk value of the pressure container appearing leakage is evaluated, and an alarm prompt is generated based on the risk value and sent to a management terminal to control the risk of the pressure container.

[0011] Preferably, the acoustic wave reflection information of the surface of the pressure container is acquired, the defect area existing on the surface of the pressure container is identified based on the acoustic wave reflection information, and the characteristic state of the defect on the surface of the pressure container is judged in combination with the acoustic wave reflection intensity, including:

[0012] The ultrasonic probe is contacted with the surface of the pressure container to send the ultrasonic signal, and the acoustic wave reflection information sensed by the surface of the pressure container along the way is analyzed according to the ultrasonic signal;

[0013] The backscattering light signal of the acoustic wave reflection information is analyzed based on the optical time domain technology, the defect area existing on the surface of the pressure container is acquired according to the analysis result, and the characteristic recognition of the backscattering light signal is performed;

[0014] The defect type is determined by matching the characteristic recognition result with the known defect mode, and the defect characteristic state including the defect depth and the defect shape is acquired in combination with the time delay of the ultrasonic signal.

[0015] Preferably, the backscattering light signal of the acoustic wave reflection information is analyzed based on the optical time domain technology, the defect area existing on the surface of the pressure container is acquired according to the analysis result, and the characteristic recognition of the backscattering light signal is performed, including:

[0016] The vibration change of the acoustic wave reflection signal in the propagation process with the surface of the pressure container is analyzed based on the optical time domain technology, and the backscattering light signal reflected from the surface of the pressure container is detected in combination with the optical sensor;

[0017] The fractional order transformation of the backscattering light signal is performed based on the state function superposition technology to obtain the fractional order time-frequency feature, and the symmetry law of the time-frequency distribution corresponding to the fractional order time-frequency feature is analyzed;

[0018] The backscattering light signal is analyzed according to the symmetry law of the time-frequency distribution, the average refractive index of the maximum time of the time-frequency distribution is acquired, and the defect area existing on the surface of the pressure container is judged based on the average refractive index;

[0019] After the backscattering light signal is removed by the filter, the frequency spectrum feature of the backscattering light signal is extracted, and the peak value and the frequency in the frequency spectrum feature are recognized as the feature vector of the backscattering light signal.

[0020] Preferably, the fractional order transformation of the backscattering light signal is performed based on the state function superposition technology to obtain the fractional order time-frequency feature, and the symmetry law of the time-frequency distribution corresponding to the fractional order time-frequency feature is analyzed, including:

[0021] The backscattering light signal is preliminarily processed by using fractional Fourier transform to obtain the frequency distribution and time variation trend of the backscattering light signal and determine the base characteristic parameters of the state function superposition technology;

[0022] According to the base characteristic function, the peak value of the backscattering light signal at the fractional order domain position after the preliminary processing is analyzed, and the backscattering light signal component is determined based on the peak value and the fractional order domain.

[0023] The backscattering light signal component is discretized based on the pseudo time-frequency distribution technology to generate a discrete matrix, and the discrete matrix is mapped into a unit circle after being binarized.

[0024] The fractional order time-frequency features are extracted according to the position information of the discrete matrix in the unit circle, and the entropy feature vector analysis of the fractional order time-frequency features is performed by using the mean clustering technology to obtain the time-frequency distribution symmetry rule.

[0025] Preferably, the fractional order time-frequency features are extracted according to the position information of the discrete matrix in the unit circle, and the entropy feature vector analysis of the fractional order time-frequency features is performed by using the mean clustering technology to obtain the time-frequency distribution symmetry rule, which includes:

[0026] The center of the discrete matrix is selected as the origin of the polar coordinate, and the complete orthogonal complex function set of the discrete matrix in the unit circle is determined, and the fractional order time-frequency features are extracted according to the complete orthogonal complex function set and the origin with the set order and multiplicity;

[0027] The number of mean clustering categories is set, and a plurality of samples are randomly selected from the fractional order time-frequency features as the entropy feature vector centers, and the Euclidean distances from the fractional order time-frequency features to the entropy feature vector centers are calculated.

[0028] The fractional order time-frequency features are divided into the subcategories corresponding to the nearest entropy feature vector centers according to the Euclidean distance calculation results, and the corresponding time-frequency feature modes are analyzed based on the subcategory division results.

[0029] The time axis symmetry of each time-frequency in the fractional order time-frequency features is analyzed based on the time-frequency feature mode analysis, and the time-frequency distribution symmetry rule is analyzed according to the time axis symmetry of each time-frequency feature mode.

[0030] Preferably, the characteristic state and the stress elastoplastic analysis technology are combined to obtain the damage mode of different position points in the defect area of the pressure vessel surface, and the damage characterization parameter and the damage evolution rule are determined according to the damage mode, which includes:

[0031] A stress elastoplastic analysis model is constructed according to the material and mechanical property parameters of the pressure vessel, and the stress and strain distribution of the defect area of the pressure vessel when the defect exists is analyzed by combining the defect characteristic state.

[0032] The stress and strain distribution reflects strain accumulation and fatigue damage accumulation of the pressure vessel, and identifies fatigue crack propagation processes of different position points in the surface defect area of the pressure vessel.

[0033] According to the fatigue crack propagation process, the damage mode of different position points in the surface defect area of the pressure vessel is determined, and the damage characterization parameter is analyzed in combination with the stress concentration degree of the fatigue crack end.

[0034] The relationship between the crack propagation speed and the stress intensity factor is simulated by combining the cyclic load condition, the damage characterization parameter and the incremental damage technology, and the damage evolution law is determined based on the relationship result.

[0035] According to the fatigue crack propagation process, the damage mode of different position points in the surface defect area of the pressure vessel is determined, and the damage characterization parameter is analyzed in combination with the stress concentration degree of the fatigue crack end.

[0036] According to the fatigue crack propagation process, a damage mode database is established, and a matching factor matrix is constructed based on the stress analysis local transfer rate function of the position points in the surface defect area of the pressure vessel.

[0037] The extension parameters in the damage mode database are called to construct a damage mode matrix, the mapping relationship between the damage mode matrix and the matching factor matrix is analyzed, and the damage mode of different position points in the defect area is matched according to the mapping relationship.

[0038] The stress concentration degree of the fatigue crack end is obtained, a characterization matrix of the stress concentration degree and the damage mode is constructed, and a loss function is set to analyze the fitting score data corresponding to the characterization matrix.

[0039] The plastic zone of the fatigue crack end is analyzed by analyzing the fitting score data, the damage degree of the pressure vessel is described according to the analysis result and the stress concentration degree, and the damage characterization parameter in the surface defect area of the pressure vessel is determined based on the damage degree.

[0040] According to the fatigue crack propagation process, a damage mode database is established, and a matching factor matrix is constructed based on the stress analysis local transfer rate function of the position points in the surface defect area of the pressure vessel.

[0041] According to the fatigue crack propagation process, a damage mode database is established, and a matching factor matrix is constructed based on the stress analysis local transfer rate function of the position points in the surface defect area of the pressure vessel.

[0042] The transmission relationship between the extension form, the extension rate and the extension direction is determined by establishing a damage mode database based on the fatigue crack extension form, the extension rate and the extension direction.

[0043] The Duhamel integral technique is used to solve the displacement response vector to analyze the relationship between the crack propagation rate and the stress of the position point in the defect area on the surface of the pressure container, to generate a local transfer rate function reflecting the influence of the defect position point on the crack propagation;

[0044] The displacement response vector of the single freedom propagation equation of all groups of uncoupled groups is obtained, and a transfer rate function is generated based on the obtained results to construct a matching factor matrix.

[0045] Preferably, the expression of the local transfer rate function is:

[0046]

[0047] In the formula, X(x) represents the influence of the xth defect position point on the crack propagation, i represents the total number of defect position points, j represents the number of fatigue crack propagation, β j represents the crack propagation length corresponding to the jth fatigue crack propagation process, σ j represents the crack propagation rate corresponding to the jth fatigue crack propagation process, γ j represents the modal damping ratio corresponding to the jth fatigue crack propagation process, δ j represents the crack propagation mode corresponding to the jth fatigue crack propagation process, υ j represents the ratio between the crack propagation mode corresponding to the jth fatigue crack propagation process and the modal damping ratio corresponding to the jth fatigue crack propagation process, e represents a constant, F(α) represents the displacement response vector corresponding to the single freedom propagation equation of the αth group, dα represents the approximate consideration value of the single freedom propagation equation of the αth group, represents the crack propagation length corresponding to the T time in the jth fatigue crack propagation process.

[0048] In a second aspect, the present application also provides a detection alarm system for detecting defects on the surface of a pressure container, which comprises:

[0049] A defect detection and judgment module is configured to obtain acoustic wave reflection information of the surface of the pressure container, identify a defect area existing on the surface of the pressure container based on the acoustic wave reflection information, and determine the characteristic state of the defect on the surface of the pressure container in combination with the acoustic wave reflection intensity;

[0050] A defect damage analysis module is configured to combine the characteristic state with the stress elastoplastic analysis technology, obtain damage modes of different position points in the defect area on the surface of the pressure container, and determine damage characterization parameters and damage evolution rules according to the damage modes;

[0051] An alarm prompt control module is configured to evaluate a risk value of leakage of the pressure container according to the damage characterization parameters and the damage evolution rules, generate an alarm prompt based on the risk value, and send the alarm prompt to a management terminal to perform risk control of the pressure container.

[0052] The beneficial effects of the present application are:

[0053] 1、The present application reflects the nature of defects by obtaining the sound wave reflection information of the surface of the pressure vessel, and the sound wave reflection detection method does not need to directly contact the surface of the container, can be detected without interfering with the normal operation of the equipment, and at the same time, by combining the sound wave reflection data with the stress elastic-plastic analysis, the stress distribution, plastic deformation and its performance under different working conditions of the defect area can be accurately evaluated, the potential impact of the defect on the stability of the pressure vessel can be further analyzed, and by analyzing the damage mode of the defect area, the damage characterization parameter can be determined, the damage state of the container can be accurately described, which is helpful for predicting the future performance degradation and facilitating early warning.

[0054] 2、The present application contacts the ultrasonic probe with the surface of the pressure vessel, and sends ultrasonic signals to detect defects without damaging the structural integrity of the pressure vessel, which can effectively detect surface defects of the pressure vessel, and analyzing the sound wave reflection information can comprehensively understand the defect characteristics of the container surface, and through automatic feature recognition and matching with known defect modes, the automation degree of detection and the intelligence of analysis are improved, and human error is reduced.

[0055] 3、The present application constructs a stress elastic-plastic analysis model by the material properties and mechanical properties of the pressure vessel, which can accurately calculate the stress and strain distribution of the pressure vessel, and further helps to understand the stress condition of the defect area under working load, avoids excessive simplification of assumptions, and improves the accuracy of analysis, and based on the stress and strain distribution, the deformation process of the container material under cyclic load can be reflected, which further helps to identify the initiation and accumulation of fatigue damage, and helps to predict the fatigue life of the pressure vessel and the crack propagation process of the surface defect area, and provides a scientific basis for the residual life prediction and risk control of the pressure vessel. BRIEF DESCRIPTION OF DRAWINGS

[0056] The drawings accompanying the specification of the present application form a part thereof, serve to provide further understanding of the present application, and together with the description of the exemplary embodiments of the present application and their description serve the purpose of explaining the present application, and do not constitute an improper limitation of the present application.

[0057] Figure 1 is a flow chart of a detection and warning method for detecting surface defects of a pressure vessel according to an embodiment of the present application;

[0058] Figure 2 is a principle block diagram of a detection and warning system for detecting surface defects of a pressure vessel according to an embodiment of the present application.

[0059] In the drawings:

[0060] 1, defect detection judgment module; 2, defect damage analysis module; 3, alarm prompt control module. DETAILED DESCRIPTION

[0061] The application will be further described below in conjunction with the drawings and embodiments.

[0062] It should be noted that the following detailed description is exemplary and is intended to provide further explanation of the application. Unless otherwise indicated, all technical and scientific terms used herein have the same meaning as those commonly understood by one of ordinary skill in the art to which this application belongs.

[0063] It should be noted that the terms used herein are only for the purpose of describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, the singular form is intended to include the plural form unless the context clearly indicates otherwise, and it should be understood that the terms "comprise" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device comprising a series of steps or units does not have to be limited to those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to the process, method, product or device.

[0064] The embodiments in the present application and the features in the embodiments can be combined with each other without conflict.

[0065] Please refer to Figure 1 The present application provides a detection and alarm method for detecting surface defects of a pressure vessel, which comprises:

[0066] Step S1, obtaining acoustic wave reflection information of the surface of the pressure vessel, identifying defect areas existing on the surface of the pressure vessel based on the acoustic wave reflection information, and judging the characteristic state of the surface defects of the pressure vessel in combination with the acoustic wave reflection intensity.

[0067] In one embodiment, obtaining acoustic wave reflection information of the surface of the pressure vessel, identifying defect areas existing on the surface of the pressure vessel based on the acoustic wave reflection information, and judging the characteristic state of the surface defects of the pressure vessel in combination with the acoustic wave reflection intensity comprises:

[0068] The ultrasonic probe is in contact with the surface of the pressure vessel to send ultrasonic signals, and the acoustic wave reflection information sensed by the pressure vessel along the way is analyzed according to the ultrasonic signals;

[0069] Based on the optical time domain technology, the backscattering light signal of the acoustic wave reflection information is analyzed, the defect areas existing on the surface of the pressure vessel are obtained according to the analysis results, and the backscattering light signal is characterized and identified;

[0070] The feature recognition result is matched with a known defect mode to determine a defect type, and a time delay condition of the ultrasonic signal is combined to obtain a defect feature state including a defect depth and a defect shape.

[0071] It should be explained that in the process of contacting the ultrasonic probe with the surface of the pressure container to send the ultrasonic signal, and analyzing the sound wave reflection information sensed by the surface of the pressure container along the way according to the ultrasonic signal, a suitable ultrasonic probe is selected in advance, a probe with different frequency and diameter is selected according to the detection requirement, the probe is calibrated to ensure its stable performance and can accurately send and receive the ultrasonic signal, a coupling agent is applied between the probe and the surface of the pressure container to reduce the interference of the air layer and improve the transmission efficiency of the sound wave, the ultrasonic probe is contacted with the surface of the pressure container to ensure that the contact surface is flat and well coupled, the ultrasonic signal is sent through the probe, the signal frequency and intensity should be adjusted according to the detection requirement, the probe receives the sound wave signal reflected from the surface and internal structure of the pressure container, and the sound wave signal is recorded.

[0072] Specifically, in the process of analyzing the backscattered light signal of the sound wave reflection information based on the optical time domain technology, obtaining the defect area existing on the surface of the pressure container according to the analysis result, and performing feature recognition on the backscattered light signal, the vibration change of the sound wave reflection signal in the propagation process with the surface of the pressure container can be analyzed based on the optical time domain technology, and the backscattered light signal reflected from the surface of the pressure container is detected by combining the optical sensor; the fractional order transformation of the backscattered light signal is obtained based on the state function superposition technology to obtain the fractional order time-frequency feature, and the time-frequency distribution symmetry law corresponding to the fractional order time-frequency feature is analyzed; the backscattered light signal is analyzed according to the time-frequency distribution symmetry law, the maximum time of the time-frequency distribution is obtained, the average refractive index is obtained, and the defect area existing on the surface of the pressure container is judged based on the average refractive index; after the backscattered light signal is removed by the filter, the frequency spectrum feature of the backscattered light signal is extracted, and the peak value and frequency in the frequency spectrum feature are identified as the feature vector of the backscattered light signal.

[0073] It needs to be explained that in the process of detecting the backscattered light signal reflected by the surface of the pressure vessel, a suitable optical sensor (such as a laser displacement sensor, an optical fiber sensor, etc.) and a light source (such as a laser beam, an LED, etc.) need to be selected to ensure that the light signal can effectively interact with the surface of the pressure vessel and be reflected back to the sensor. The optical sensor is kept in stable contact with the surface of the pressure vessel to ensure that it can accurately capture the surface vibration and the reflected light signal. When the ultrasonic probe sends ultrasonic signals to the surface of the pressure vessel, the ultrasonic waves interact with the surface of the vessel during propagation, causing slight vibrations on the surface. These vibrations will affect the light signal reflected back from the surface of the vessel. The surface vibration caused by the ultrasonic waves causes slight deformation of the vessel surface, which causes scattering of the light beam, especially backscattered light. The backscattered light is the light reflected from the surface of the vessel and finally captured by the optical sensor. The optical sensor captures the backscattered light signal and converts it into an electronic signal, recording the intensity, frequency, phase, and other information of the signal.

[0074] In the application of optical time domain technology, the key is to analyze the interaction between the ultrasonic wave and the vibration of the pressure vessel surface. The ultrasonic wave causes slight vibration of the vessel surface during propagation, and the vibration has a significant impact on the scattering of light. Specifically, the frequency, amplitude, and direction of the vibration affect the intensity and phase of the scattered light, thereby providing important clues for defect identification.

[0075] In the process of obtaining the fractional order time-frequency feature of the backscattered light signal based on the state function superposition technology and analyzing the symmetry rule of the time-frequency distribution corresponding to the fractional order time-frequency feature, the fractional order Fourier transform can be used to preliminarily process the backscattered light signal to obtain the frequency distribution and time variation trend of the backscattered light signal, and to determine the base characteristic parameters of the state function superposition technology. According to the base characteristic function, the sharp peak value of the backscattered light signal at the position in the fractional order domain after preliminary processing is analyzed, and the backscattered light signal component is determined based on the sharp peak value and the fractional order domain. The backscattered light signal component is discretized based on the pseudo time-frequency distribution technology (pseudo Wigner-Vile distribution) to generate a discrete matrix, which is mapped into a unit circle after binary processing. The fractional order time-frequency feature is extracted according to the position information of the discrete matrix in the unit circle, and the entropy feature vector analysis of the fractional order time-frequency feature is performed using the mean clustering technology to obtain the symmetry rule of the time-frequency distribution.

[0076] Wherein, when the fractional order time-frequency features are extracted according to the position information of the discrete matrix in the unit circle, and the entropy feature vector analysis is performed on the fractional order time-frequency features by using the mean clustering technology, the center of the discrete matrix can be selected as the origin of the polar coordinates, and the complete orthogonal complex function set of the discrete matrix in the unit circle is judged, and the fractional order time-frequency features are extracted according to the complete orthogonal complex function set and the origin according to the set order and multiplicity; the number of mean clustering categories is set, and a plurality of samples are randomly selected from the fractional order time-frequency features as the entropy feature vector centers, and the Euclidean distance from the fractional order time-frequency features to the entropy feature vector centers is calculated; the fractional order time-frequency features are divided into the subcategory corresponding to the nearest entropy feature vector center according to the Euclidean distance calculation result, and the time-frequency feature mode is analyzed based on the subcategory division result; the time axis symmetry of each time-frequency in the time-frequency graph is analyzed based on the time-frequency feature mode, and the time-frequency distribution symmetry rule is analyzed according to the time axis symmetry of each time-frequency feature mode.

[0077] It needs to be explained that in the process of obtaining the feature vector of the backscattering light signal, the peak value of the backscattering light signal at the fractional order domain position is analyzed according to the base feature function, the peak value reflects the concentration degree of the signal in the specific fractional order domain, which is helpful to identify the principal component of the signal, and the components of the backscattering light signal are determined based on the peak value and the fractional order domain, which represent the main features of the signal at different frequencies and times. The backscattering light signal components are discretized to generate a discrete matrix, and each element of the discrete matrix represents the intensity of the signal at a specific time and frequency. The discrete matrix is binarized, and elements with signal intensity greater than a certain threshold are marked as 1, and the rest are marked as 0.

[0078] The center of the discrete matrix is selected as the origin of the polar coordinates, and the complete orthogonal complex function set of the discrete matrix in the unit circle is judged, which is used to extract the fractional order time-frequency features. According to the complete orthogonal complex function set and the origin, the fractional order time-frequency features are extracted according to the set order and multiplicity, which reflects the distribution of the signal at different times and frequencies. According to the analysis requirement, the number of mean clustering categories is set, and the selection of the number of categories depends on the expected defect type and quantity. A plurality of samples are randomly selected from the fractional order time-frequency features as the entropy feature vector centers, and the Euclidean distance from each fractional order time-frequency feature to each entropy feature vector center is calculated. According to the Euclidean distance, the fractional order time-frequency features are divided into the subcategory corresponding to the nearest entropy feature vector center, and the time-frequency feature mode of each subcategory is analyzed based on the subcategory division result. These modes reflect the distribution rule of the signal at different times and frequencies.

[0079] According to the time axis symmetry of each time-frequency feature mode, the symmetry rule of the time-frequency distribution is analyzed, the symmetry analysis is helpful to identify the periodicity and symmetry characteristics in the signal, so as to judge the existence and position of the defect, according to the symmetry rule of the time-frequency distribution, the maximum time of the time-frequency distribution (i.e. the maximum signal intensity) is determined, the average refractive index at this time is calculated, the average refractive index reflects the physical properties of the container surface material and its influence on signal propagation, according to the change of the average refractive index, the defect area existing on the container surface is judged, the significant change of the refractive index usually indicates the existence of defects, such as cracks, corrosion, etc.

[0080] Suitable filters (such as low-pass filters, high-pass filters, band-pass filters) are selected to remove noise from the backscattered light signal, and the denoised signal is subjected to frequency spectrum analysis to extract the frequency spectrum characteristics of the signal, such as frequency peak value, frequency distribution, etc. These frequency spectrum characteristics reflect the frequency components and their intensities of the signal. The peak values in the frequency spectrum characteristics are combined with the frequencies to form feature vectors, which can be used for subsequent defect identification and classification.

[0081] In order to facilitate the understanding of the above technical solutions of the present application, the operation mode of the defect area detection of the present application in the actual process will be described in detail below.

[0082] Step one, backscattered light signal analysis;

[0083] Suppose a pressure vessel surface is being detected, which may have cracks, corrosion or other defects, according to the detection requirements, an ultrasonic probe and related parameters are selected: an array ultrasonic probe is selected, which can emit and receive ultrasonic signals, suitable for surface and internal detection, a probe with a frequency of 5MHz is selected, a probe with a higher frequency (such as 5MHz) is suitable for detecting smaller surface defects, while a probe with a lower frequency is suitable for deep defect detection.

[0084] At the same time, a probe diameter of 1.5cm is selected, which is suitable for smaller detection areas, ensuring that the signal covers the surface of the container, and a water-based coupling agent is used, which helps to reduce the interference of the air layer and ensure the transmission efficiency of the signal.

[0085] The ultrasonic probe is brought into contact with the surface of the pressure vessel, 5MHz ultrasonic signals (such as sinusoidal waves) are sent through the probe and made to propagate through the surface and interior of the container, the probe receives the reflected acoustic signals from the surface and internal structure of the container, and records the time delay and intensity of the reflected signals.

[0086] Suppose in a defect-free area, the intensity of the reflected signal is 70dB, if there is a defect such as a crack or corrosion, the intensity of the reflected signal may be significantly reduced or changed in shape (for example, the intensity is reduced to 50dB or lower).

[0087] In analyzing the sound wave reflection information, the optical sensor and the light time domain technology are combined to detect the vibration changes on the container surface. A laser displacement sensor with a wavelength of 670 nm is used to provide high-precision displacement measurement. The laser sensor is kept in stable contact with the surface of the pressure container to ensure that it can accurately capture the backscattered light caused by the small vibrations of the surface. During the propagation of the ultrasonic signal, the sound wave interacts with the container surface, causing small vibrations that affect the scattering of the light beam. The frequency, amplitude, and direction of the vibrations affect the intensity and phase of the backscattered light. The laser sensor captures the backscattered light reflected from the surface of the pressure container and converts it into an electronic signal. Assuming that the intensity of the reflected light signal is 500 μW and the frequency is 10 kHz.

[0088] Step two, time-frequency feature pattern analysis;

[0089] The captured backscattered light signal is subjected to fractional Fourier transform, assuming that the selected order is 0.5, the frequency distribution and time variation trend of the signal are obtained. In the fractional order domain, the signal's peak value may appear at 10 kHz, indicating the main frequency component of the signal. The pseudo-time-frequency distribution technique (such as pseudo-Wigner-Vile distribution) is used to discretely process the components of the backscattered light signal, as follows:

[0090] The signal components are discretized, assuming that the generated discrete matrix contains 200x200 elements, each element representing the signal intensity at a specific time and frequency. Assuming that the threshold value is set to 0.4 μW, all elements with intensity higher than this value are marked as 1, and others are marked as 0. At this time, the size of the discrete matrix is 200x200, and it is mapped into the unit circle.

[0091] The number of clusters is set to 3 to distinguish different types of defects. 50 samples are randomly selected from the fractional order time-frequency features as entropy feature vector centers. The Euclidean distance between each fractional order time-frequency feature and the entropy feature vector center of each class is calculated. Assuming that the Euclidean distance between a certain feature and the feature vector center of a certain class is 2.5, then this feature is classified into this class. By analyzing the time-frequency feature pattern, the periodic and symmetric features are identified. Assuming that at a frequency of 10 kHz, a feature with obvious symmetry is found, which may be related to a crack on the surface of the container.

[0092] Step three, final defect identification;

[0093] According to the symmetry rule of time-frequency distribution, the maximum time of time-frequency distribution (i.e. the time of maximum signal strength) is obtained, and the average refractive index at this time is calculated: assuming that the maximum signal strength occurs at 5 μs, the average refractive index at this time is 1.45, which is slightly higher than the refractive index of the normal region (1.4), indicating that this region may have defects. If the average refractive index is significantly different from that of the normal region, it may indicate that there are cracks, corrosion or other defects on the surface of the container, for example, a refractive index of 1.55 may indicate a corrosion area.

[0094] Using a low-pass filter, assuming a 1 MHz cutoff frequency is selected to remove high-frequency noise in the signal, the denoised signal is subjected to spectral analysis, and assuming the resulting spectral features are 15 kHz and 30 kHz peaks, reflecting the vibration mode of the container surface, the peaks in the spectral features are combined with the frequency to form a feature vector, and assuming the resulting feature vector is 15 kHz, 0.5 μW and 30 kHz, 0.3 μW.

[0095] Combining the results of acoustic wave reflection intensity and optical time domain technology analysis, assuming that by matching with known defect patterns, it is determined that the feature vector indicates a crack defect, and through the time delay of the ultrasonic wave signal, it is assumed that the depth of the crack is 3 mm and the shape of the crack is a long strip.

[0096] Therefore, through this series of processes, defects present on the surface of the pressure vessel can be accurately identified and analyzed for depth and shape.

[0097] Step S2, combining the feature state with stress elastoplastic analysis technology, obtaining the damage mode of different position points in the defect area on the surface of the pressure vessel, and determining the damage characterization parameter and damage evolution law according to the damage mode.

[0098] In one embodiment, combining the feature state with stress elastoplastic analysis technology, obtaining the damage mode of different position points in the defect area on the surface of the pressure vessel, and determining the damage characterization parameter and damage evolution law according to the damage mode comprises:

[0099] A stress elastoplastic analysis model is constructed according to the material and mechanical properties of the pressure vessel, and the stress and strain distribution of the pressure vessel in the defect area is analyzed in combination with the defect feature state;

[0100] Based on the stress and strain distribution, the strain accumulation and fatigue damage accumulation of the pressure vessel are reflected, and the fatigue crack propagation process of different position points in the defect area on the surface of the pressure vessel is identified;

[0101] According to the fatigue crack propagation process, the damage mode of different position points in the defect area on the surface of the pressure vessel is determined, and the damage characterization parameter is analyzed in combination with the stress concentration degree at the end of the fatigue crack.

[0102] The relationship between crack propagation rate and stress intensity factor is simulated by combining cyclic loading conditions, damage characterization parameters, and incremental damage techniques, and the damage evolution law is determined based on the relationship results.

[0103] It should be noted that when analyzing the stress and strain distribution in the defect area of the pressure vessel in the presence of defects, the material used for the pressure vessel and its mechanical property parameters need to be determined first. These parameters are crucial for accurately simulating the stress, strain, and plastic behavior of the vessel. Common pressure vessel materials include steel, aluminum alloy, composite materials, etc. Selecting the appropriate material type, assuming carbon steel or stainless steel as the vessel material, based on experimental data of the material, the following parameters are determined: yield strength: for carbon steel, the yield strength is 250 MPa; elastic modulus: the elastic modulus of carbon steel is 210 GPa; Poisson's ratio: the Poisson's ratio of carbon steel is 0.3; hardening modulus: describes the hardening behavior of the material in the plastic deformation stage, usually obtained from experimental data.

[0104] The stress-strain relationship can determine the plastic behavior of the material according to the von Mises yield criterion or the Tresca yield criterion. When performing stress analysis, the geometry of the vessel and the loading conditions also need to be determined. Assuming the pressure vessel is a thick-walled cylinder (the most common shape), the cross-sectional shape of the pressure vessel is circular or elliptical, and modeling is performed as needed.

[0105] A stress elastoplastic analysis model of the pressure vessel is established using the finite element method (FEM), and the following are the modeling steps:

[0106] According to the actual size and shape of the pressure vessel, a geometric model of the vessel is established; the model is divided into small finite element units, such as tetrahedral elements or hexahedral elements, ensuring that there are enough fine grids near the defect area; appropriate boundary conditions (such as fixed ends) and loading conditions (such as internal pressure) are applied; assign material properties (such as elastic modulus, yield strength, etc.) to each element, and modify the local properties of the material in the defect area.

[0107] Nonlinear elastoplastic analysis is performed on the model to consider plastic deformation and material hardening, and the incremental iteration method (such as the Newton-Raphson method) is used to solve the elastoplastic problem. Assuming that the defect exists on the surface of the vessel, stress will concentrate near the crack or corrosion area, showing stress peaks. In the crack or corrosion area, local stress concentration may occur, such as significant increase in principal stress and shear stress. Von Mises stress analysis is performed on the stress to identify whether the yield limit is reached.

[0108] According to the stress distribution, the strain of the container at different positions is calculated, and the strain in the plastic region is usually larger, while the strain in the elastic region is smaller. For the defect region, especially near the crack, the strain may be larger, indicating that the material is undergoing plastic deformation. For the corrosion region, the strain distribution may be uniform, but the overall deformation is larger.

[0109] It should be noted that when reflecting the strain accumulation and fatigue damage accumulation of the pressure vessel based on the stress and strain distribution, and identifying the fatigue crack propagation process at different positions in the defect region on the surface of the pressure vessel, modeling can be based on the stress-strain curve of the material (such as bilinear hardening, ideal elastic material, etc.), the elastic and plastic regions of the material are determined according to the stress distribution, and the plastic deformation part will have an impact on the fatigue damage. Since the pressure vessel is repeatedly loaded during operation, the strain distribution of the container will change, and the strain accumulation can be calculated through the time history of the strain.

[0110] Fatigue damage is generally represented by the following standards: Poisson's ratio, elastic modulus, and hardening modulus have important influence on fatigue damage accumulation; stress amplitude and average stress play a key role in fatigue life and damage accumulation; strain energy density is an energy conversion model based on strain distribution, which evaluates the energy accumulation of the material under fatigue loading, and S-N curve (stress amplitude-life curve) is used to describe the fatigue life of the material at different stress levels.

[0111] First, the finite element method (FEM) is used to establish the stress and strain distribution model of the pressure vessel, based on the geometry of the container, loading conditions and material parameters, the stress and strain field of the container under external force is solved, which requires fine meshing, especially around the defect area, to ensure accurate calculation of stress concentration and strain accumulation. For defect areas (such as cracks, corrosion areas, etc.), the extended finite element method (XFEM) can be used to simulate the existence and evolution of cracks.

[0112] Defect areas will cause stress concentration, which usually occurs at the crack tip, corrosion area, etc. The strain distribution at each position is calculated through the stress and constitutive relationship, and the strain in the plastic region is larger, reflecting that the material is undergoing plastic deformation. For the crack region, the strain concentration will cause significant plastic deformation at the crack tip, which helps the crack propagation.

[0113] Before fatigue analysis, the initial position, shape, direction and depth of the crack need to be determined, based on the results of acoustic wave reflection or optical sensor detection to obtain preliminary information about the crack. Assuming the crack depth is 2mm, the crack shape may be sharp or curved, and the crack is assumed to be located on the surface of the container with a certain directionality.

[0114] The calculation of fatigue damage in Miner's Rule typically uses Miner's fatigue damage accumulation method, where damage can be calculated by the accumulation of damage per cycle, with the damage of each loading cycle being G = n / k, where n represents the number of cycles and k represents the fatigue life under that stress amplitude, and the accumulated fatigue damage is the sum of the damage per cycle.

[0115] Crack propagation generally follows the Paris law, which describes the relationship between crack propagation rate and stress intensity factor range (ΔK): p / h = C(ΔK), where p / h represents the crack propagation rate, C represents the number of loadings, and ΔK represents the change in stress intensity factor. According to the initial position, shape and loading conditions of the crack, the propagation path of the crack is simulated, and the process of crack propagation is closely related to the fatigue life of the material, the number of loadings and the stress distribution.

[0116] Under different stress cycles, the crack will gradually expand with the increase of loading cycles. When simulating crack propagation, the following factors should be considered:

[0117] Crack propagation direction: cracks usually propagate along the direction of maximum principal stress, and may exhibit oblique propagation; Crack depth variation: as the crack propagates, the crack depth and length increase, and the crack shape may change (such as from sharp crack to curved crack); Stress concentration effect: the propagation of the crack may cause greater stress concentration, thus accelerating the propagation process of the crack.

[0118] The fatigue life is calculated by the accumulated damage, and the failure time of the container is predicted after the crack propagates to a certain size. At the crack tip, damage will accelerate accumulation, leading to rapid crack propagation and eventually rupture.

[0119] Specifically, in determining the damage mode of different position points in the surface defect area of the pressure vessel according to the fatigue crack propagation process, and analyzing the damage characterization parameters in combination with the stress concentration degree at the fatigue crack tip, a damage mode database can be established according to the fatigue crack propagation process, and a matching factor matrix can be constructed based on the stress analysis local transfer rate function of the position points in the surface defect area of the pressure vessel; the extension parameters in the damage mode database are called to construct a damage mode matrix, the mapping relationship between the damage mode matrix and the matching factor matrix is analyzed, and the damage mode of different position points in the defect area is matched according to the mapping relationship; the stress concentration degree at the fatigue crack tip is obtained, a characterization matrix of stress concentration degree and damage mode is constructed, and a loss function is set to analyze the fitting score data corresponding to the characterization matrix; the plastic zone at the fatigue crack tip is analyzed by fitting score data, the damage degree of the pressure vessel is described according to the analysis result and the stress concentration degree, and the damage characterization parameters in the surface defect area of the pressure vessel are determined based on the damage degree.

[0120] Wherein, in the process of establishing the damage mode database according to the fatigue crack propagation process, and constructing the matching factor matrix based on the stress analysis local transfer rate function of the position points in the surface defect area of the pressure vessel, the damage mode database can be established based on the fatigue crack propagation morphology, propagation rate and propagation direction in the process of fatigue crack propagation, and the transfer relationship between the propagation morphology, propagation rate and propagation direction is judged; the transfer relationship is decoupled into a plurality of groups of single free expansion equations which are not coupled with each other, and the displacement response vector in the process of fatigue crack propagation is obtained by combining the approximate consideration of the stress and strain distribution of the defect area of the pressure vessel; the relationship between the crack propagation rate and the stress of the position points in the surface defect area of the pressure vessel is analyzed by using the Duhamel integral technology to solve the displacement response vector, and the local transfer rate function is generated to reflect the influence of the defect position points on the crack propagation; the displacement response vectors of all groups of single free expansion equations which are not coupled with each other are obtained, and the transfer rate function is generated according to the obtained results to construct the matching factor matrix.

[0121] Wherein, the expression of the local transfer rate function is:

[0122]

[0123] In the formula, X(x) represents the influence of the xth defect position point on the crack propagation, i represents the total number of defect position points, j represents the number of fatigue crack propagation, β j represents the crack propagation length corresponding to the jth fatigue crack propagation process, σ j represents the crack propagation rate corresponding to the jth fatigue crack propagation process, γ j represents the modal damping ratio corresponding to the jth fatigue crack propagation process, δ j represents the crack propagation morphology corresponding to the jth fatigue crack propagation process, υ j represents the ratio between the crack propagation morphology corresponding to the jth fatigue crack propagation process and the modal damping ratio corresponding to the jth fatigue crack propagation process, e represents a constant, F(α) represents the displacement response vector corresponding to the αth group of single free expansion equations, dα represents the approximate consideration value corresponding to the αth group of single free expansion equations, represents the crack propagation length corresponding to the Tth time in the jth fatigue crack propagation process.

[0124] In order to facilitate the understanding of the above technical solutions of the present application, the operation mode of the damage characterization parameters and the damage evolution law of the present application in the actual process will be described in detail below.

[0125] Step one, finite element model establishment;

[0126] Assuming that carbon steel is selected as the material of the pressure vessel, the mechanical property parameters of the carbon steel are as follows:

[0127] Yield strength: 250 MPa; Elastic modulus: 210 GPa; Poisson's ratio: 0.3; Hardening modulus: 25 MPa;

[0128] Assume the container is a thick-walled cylinder, its size and loading conditions are as follows:

[0129] Inner radius: 200 mm; Outer radius: 300 mm; Length: 500 mm; Internal pressure: 10 MPa; External load: no external load;

[0130] The stress elastoplastic analysis model of the pressure vessel is established using the finite element method (FEM). First, a thick-walled cylindrical model is established according to the size of the container, and tetrahedral elements and hexahedral elements are used. In the defect area, finer mesh division is used to improve accuracy. Fixed boundary conditions are applied to both ends of the container, and the internal pressure is 10 MPa.

[0131] Step two, fatigue crack propagation process analysis;

[0132] Assume that there is an initial crack on the surface of the container, with a crack depth of 2 mm, a sharp crack shape, and a location on the side wall of the container, and the crack propagates along the direction of the maximum principal stress. This crack is a surface defect. The stress and strain distribution in the defect area is assumed as follows: at the crack tip, the stress intensity factor increases significantly to 50, and the strain at the crack tip is mainly plastic deformation, with a calculated strain value of 0.015.

[0133] Based on the S-N curve of the pressure vessel and the elastic modulus, yield strength and other parameters of the material, fatigue damage analysis is performed. Assuming that Miner's Rule is used for damage accumulation analysis, considering the cyclic loading conditions: the container withstands repeated changes in internal and external pressure, assuming 1000 cycles per second, the damage is calculated based on the stress amplitude and average stress of each cycle.

[0134] The S-N curve is used to determine the fatigue life under different stress amplitudes, assuming the S-N curve is as follows:

[0135]

[0136] where l represents the stress amplitude (unit: MPa), U represents the fatigue life (number of cycles), and the fatigue damage of each position point is calculated based on the working conditions of the pressure vessel.

[0137] Based on the Paris rule, the crack propagation process is simulated, and the crack propagation rate and stress intensity factor are calculated based on the stress and strain distribution. The crack propagation process from a depth of 2 mm to 10 mm is simulated.

[0138] Step three, characteristic state judgment;

[0139] According to the fatigue crack propagation process, combined with stress distribution, the damage mode of different position points is determined, mainly including the following:

[0140] Surface crack mode: stress concentration is serious, and the crack propagates along the direction of the maximum principal stress;

[0141] Corrosion damage mode: the strain in the corrosion area is relatively uniform, but the overall deformation is large, and micro-cracks are easy to occur on the surface;

[0142] Plastic zone damage mode: the plastic zone at the crack tip is large, and the local deformation is significant;

[0143] According to the stress concentration degree of the fatigue crack tip, the relationship between the damage characterization parameter and the damage evolution is analyzed, and the damage characterization matrix is established. Based on the fatigue crack propagation process, a damage mode database is established. Assuming that the database includes crack propagation rate, shape, propagation direction, etc. Based on the stress and strain distribution calculation of different position points in the surface defect area of the pressure vessel, the local transfer rate function is obtained to reflect the influence of the defect position point on the crack propagation, and the matching factor matrix is constructed to reflect the damage mode of the defect area.

[0144] Through the incremental damage technology and the relationship between stress intensity factor and crack propagation rate, the damage evolution law is analyzed. Based on the analytical results, the crack propagation rate and the fatigue life of the pressure vessel can be predicted, and finally:

[0145] Crack propagation rate: according to the initial state of the crack and the loading condition, the crack propagation rate is predicted;

[0146] Damage evolution law: through the simulation results, the damage evolution law of the surface defect area of the pressure vessel is obtained, and the failure time of the container is determined;

[0147] Assuming that the surface defects of the container are detected by acoustic reflection technology, the type and size of the defects are judged by the acoustic reflection intensity: the reflection intensity is proportional to the size and depth of the crack, and larger cracks will result in stronger reflection.

[0148] Assuming that the acoustic reflection intensity is 0.8, which indicates that the depth of the crack is about 2mm, then combined with the reflection intensity and the crack propagation model, the state of the crack under different loading conditions can be further determined, and the further development of the crack can be analyzed combined with the stress and strain distribution and the damage evolution law.

[0149] Therefore, by combining stress elastic-plastic analysis with characteristic state, using finite element method and fatigue damage analysis technology, the damage mode of the surface defect area of the pressure vessel is analyzed in detail, and based on the stress and strain distribution, crack propagation process and incremental damage technology, the damage characterization parameter and damage evolution law are determined, and the characteristic state of the crack is judged by using the intensity of acoustic wave reflection, and the process of crack propagation and the fatigue life of the vessel are predicted according to the analysis results in the model.

[0150] In step S3, the risk value of the pressure vessel leaking is evaluated according to the damage characterization parameter and the damage evolution law, and an alarm prompt is generated based on the risk value and sent to the management terminal for risk control of the pressure vessel.

[0151] In one embodiment, in the process of evaluating the risk value of the pressure vessel leaking according to the damage characterization parameter and the damage evolution law, and generating an alarm prompt based on the risk value and sending it to the management terminal for risk control of the pressure vessel, the leakage risk evaluation index (leakage risk coefficient R leak ) can be defined according to the damage characterization parameter and the damage evolution law:

[0152]

[0153] In the formula, K represents the stress intensity factor, η represents the strain, p / v represents the crack propagation rate, v represents the stress amplitude, and f represents the function;

[0154] According to the evaluated risk value, the risk of leakage can be divided into several levels, for example:

[0155] Low risk (R leak ≤0.2): The container has no obvious damage, the crack propagation is slow, and the risk of leakage is low; Medium risk (0.2<R leak ≤0.5): There is a certain crack or damage, the crack propagation speed is moderate, and the risk of leakage is medium; High risk (R leak >0.5): The crack depth is large or the crack propagation speed is fast, the fatigue damage is serious, and the risk of leakage is high.

[0156] A risk threshold is set, when the risk value exceeds a certain critical value, an alarm should be triggered, for example, when the leakage risk coefficient R leak reaches 0.5 or higher, the system determines that it is high risk and enters the warning state.

[0157] According to the evaluated leakage risk value, combined with the set triggering condition, when the risk value reaches the predetermined threshold, an alarm information is generated, and the alarm information should include the following contents:

[0158] Risk level: such as low, medium, high risk; crack information: crack location, depth, expansion rate, etc.; injury degree: stress, strain and damage degree and other parameters; recommended operation: such as continue to monitor, repair or stop using, etc.

[0159] Through the automatic monitoring system or the Internet of Things equipment, the alarm prompt information is sent to the management terminal, and the management personnel need to quickly analyze the current state of the container, the crack morphology and the damage degree after receiving the risk alarm, and decide whether to take emergency measures. The management personnel can view the real-time monitoring data, historical state, crack expansion trend and other information of the container through the management platform to make reasonable decisions.

[0160] According to the leakage risk level, different response measures are taken, as follows:

[0161] Low risk: continue to monitor, set periodic inspection;

[0162] Medium risk: arrange regular inspection and intensive monitoring, check the crack expansion trend;

[0163] High risk: repair or replace immediately to avoid the container continuing to be used.

[0164] Further, the state of the pressure container can be effectively monitored, and potential leakage risks can be found in time, and appropriate measures can be taken to prevent accidents.

[0165] Please refer to Figure 2 The application also provides a detection alarm system for detecting surface defects of a pressure container, which comprises:

[0166] A defect detection and judgment module 1 is used to acquire acoustic wave reflection information of the surface of the pressure container, identify a defect area existing on the surface of the pressure container based on the acoustic wave reflection information, and judge a characteristic state of the surface defect of the pressure container in combination with acoustic wave reflection intensity;

[0167] A defect damage analysis module 2 is used to combine the characteristic state with a stress elastic-plastic analysis technology, acquire damage modes of different position points in the defect area on the surface of the pressure container, and determine damage representation parameters and damage evolution rules according to the damage modes;

[0168] An alarm prompt management and control module 3 is used to evaluate a risk value of leakage of the pressure container according to the damage representation parameters and the damage evolution rules, generate an alarm prompt based on the risk value and send the alarm prompt to a management terminal, and perform risk management and control of the pressure container.

[0169] Those skilled in the art can understand that the units and algorithm steps of each example described in combination with the present embodiment can be realized in electronic hardware or a combination of computer software and electronic hardware. Whether the functions are realized in hardware or software manner depends on the specific application and design constraints of the technical solution. The skilled person can use different methods to realize the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0170] Although the specific embodiments of the present application are described above in combination with the drawings, it is not a limitation on the scope of protection of the present application. Those skilled in the art should understand that various modifications or variations made by those skilled in the art on the basis of the technical solutions of the present application without creative labor are still within the scope of protection of the present application.

Claims

1. A detection alarm method for detecting surface defects of a pressure vessel, characterized by, The method comprises: acquiring acoustic wave reflection information of the surface of the pressure container, identifying a defect area existing on the surface of the pressure container based on the acoustic wave reflection information, and judging a characteristic state of the defect on the surface of the pressure container in combination with acoustic wave reflection intensity; constructing a stress elastic-plastic analysis model according to material and mechanical property parameters of the pressure container, and analyzing stress and strain distribution of the defect area of the pressure container when the defect exists in combination with the characteristic state of the defect; reflecting strain accumulation and fatigue damage accumulation of the pressure container based on the stress and strain distribution, and identifying a fatigue crack propagation process of different position points in the defect area on the surface of the pressure container; determining a damage mode of the different position points in the defect area on the surface of the pressure container according to the fatigue crack propagation process, and analyzing a damage characterization parameter in combination with stress concentration degree of a fatigue crack end; simulating a relationship between a crack propagation speed and a stress intensity factor in combination with the cyclic load condition, the damage characterization parameter and the incremental damage technology, determining a damage evolution law based on a relationship result; evaluating a risk value of the pressure container in terms of leakage according to the damage characterization parameter and the damage evolution law, and generating an alarm prompt based on the risk value and sending the alarm prompt to a management terminal to perform risk control of the pressure container.

2. The method of claim 1, wherein the method further comprises: The method of acquiring acoustic wave reflection information of the surface of the pressure container, identifying a defect area existing on the surface of the pressure container based on the acoustic wave reflection information, and judging a characteristic state of the defect on the surface of the pressure container comprises: contacting an ultrasonic probe with the surface of the pressure container to send an ultrasonic signal, and analyzing acoustic wave reflection information sensed by the surface of the pressure container along the way according to the ultrasonic signal; analyzing backscattered light signals of the acoustic wave reflection information based on optical time domain technology, acquiring the defect area existing on the surface of the pressure container according to an analysis result, and performing feature recognition on the backscattered light signals; matching a feature recognition result with a known defect mode to determine a defect type, and acquiring the characteristic state of the defect including defect depth and defect shape in combination with time delay of the ultrasonic signal.

3. The method of claim 2, wherein the method further comprises: The method of analyzing backscattered light signals of the acoustic wave reflection information based on optical time domain technology, acquiring the defect area existing on the surface of the pressure container according to an analysis result, and performing feature recognition on the backscattered light signals comprises: analyzing vibration changes of the acoustic wave reflection signals occurring with the surface of the pressure container in a propagation process based on optical time domain technology, and detecting the backscattered light signals reflected from the surface of the pressure container in combination with an optical sensor; performing fractional order transformation on the backscattered light signals based on state function superposition technology to obtain fractional order time-frequency features, and analyzing time-frequency distribution symmetry rules corresponding to the fractional order time-frequency features; analyzing the backscattered light signals according to the time-frequency distribution symmetry rules, acquiring a maximum time average refractive index of the time-frequency distribution, and judging the defect area existing on the surface of the pressure container based on the average refractive index; performing noise removal on the backscattered light signals by applying a filter, extracting frequency spectrum features of the backscattered light signals, and identifying peak values and frequencies in the frequency spectrum features as feature vectors of the backscattered light signals.

4. The method of claim 3, wherein the method further comprises: The method of performing fractional order transformation on the backscattered light signals based on state function superposition technology to obtain fractional order time-frequency features, and analyzing time-frequency distribution symmetry rules corresponding to the fractional order time-frequency features comprises: The backscattering light signal is preliminarily processed by using fractional Fourier transform to obtain the frequency distribution and time variation trend of the backscattering light signal and determine the base characteristic parameters of the state function superposition technology; According to the sharp peak value of the backscattering light signal after the preliminary processing in the fractional order domain, the backscattering light signal component is determined based on the sharp peak value and the fractional order domain; The backscattering light signal component is discretized based on the pseudo time-frequency distribution technology to generate a discrete matrix, and the discrete matrix is mapped into a unit circle after being binarized; The fractional order time-frequency features are extracted according to the position information of the discrete matrix in the unit circle, and the entropy feature vector analysis of the fractional order time-frequency features is performed by using the mean clustering technology to obtain the time-frequency distribution symmetry rule.

5. The method of claim 4, wherein the method further comprises: The fractional order time-frequency features are extracted according to the position information of the discrete matrix in the unit circle, and the entropy feature vector analysis of the fractional order time-frequency features is performed by using the mean clustering technology to obtain the time-frequency distribution symmetry rule. The center of the discrete matrix is selected as the origin of the polar coordinates, and the complete orthogonal complex function set of the discrete matrix in the unit circle is determined, and the fractional order time-frequency features are extracted according to the complete orthogonal complex function set and the origin with the set order and multiplicity; The number of mean clustering categories is set, and a plurality of samples are randomly selected from the fractional order time-frequency features as the entropy feature vector centers, and the Euclidean distances from the fractional order time-frequency features to the entropy feature vector centers are calculated; The fractional order time-frequency features are divided into the subcategories corresponding to the entropy feature vector centers with the nearest Euclidean distances according to the Euclidean distance calculation results, and the corresponding time-frequency feature modes are analyzed based on the subcategory division results; The time axis symmetry of each time-frequency in the time-frequency graph is analyzed based on the time-frequency feature mode analysis of the fractional order time-frequency features, and the time-frequency distribution symmetry rule is analyzed according to the time axis symmetry of each time-frequency feature mode.

6. The method of claim 1, wherein the method further comprises: The damage modes of different position points in the surface defect region of the pressure vessel are determined according to the fatigue crack propagation process, and the damage characterization parameters are analyzed in combination with the stress concentration degree of the fatigue crack end. A damage mode database is established according to the fatigue crack propagation process, and a matching factor matrix is constructed based on the stress analysis local transfer rate function of the position points in the surface defect region of the pressure vessel. The extension parameters in the damage mode database are called to construct a damage mode matrix, the mapping relationship between the damage mode matrix and the matching factor matrix is analyzed, and the damage modes of different position points in the defect region are matched according to the mapping relationship. The stress concentration degree of the fatigue crack end is obtained, a characterization matrix of the stress concentration degree and the damage mode is constructed, and a loss function is set to analyze the fitting score data corresponding to the characterization matrix. The plastic zone of the fatigue crack end is analyzed according to the fitting score data, the damage degree of the pressure vessel is described according to the analysis result and the stress concentration degree, and the damage characterization parameters of the surface defect region of the pressure vessel are determined based on the damage degree.

7. The method of claim 6, wherein the method further comprises: The damage mode database is established according to the fatigue crack propagation process, and a matching factor matrix is constructed based on the stress analysis local transfer rate function of the position points in the surface defect region of the pressure vessel. A damage mode database is established based on fatigue crack propagation process, fatigue crack propagation morphology, propagation rate and propagation direction, and a transmission relationship among the propagation morphology, the propagation rate and the propagation direction is determined; The transmission relationship is decoupled into a plurality of groups of single free propagation equations which are not coupled with each other, and a displacement response vector during fatigue crack propagation is obtained by combining an approximate consideration of stress and strain distribution in a defect area of the pressure vessel; A Duhamel integral technique is used to solve the displacement response vector to analyze a relationship between the crack propagation rate and stress at a position point in the defect area on the surface of the pressure vessel, to generate a local transmission rate function reflecting an influence of the defect position point on the crack propagation, and to reflect the influence of the defect position point on the crack propagation; Displacement response vectors of all groups of single free propagation equations which are not coupled with each other are obtained, a transmission rate function is generated according to an obtained result, and a matching factor matrix is constructed.

8. The method of claim 7, wherein the method further comprises: An expression of the local transmission rate function is as follows: ； In the formula, X(x) represents the influence of the xth defect position point on crack propagation, i represents the total number of defect position points, j represents the number of fatigue crack propagation, β j represents the crack propagation length corresponding to the jth fatigue crack propagation process, σ j represents the crack propagation rate corresponding to the jth fatigue crack propagation process, γ j represents the modal damping ratio corresponding to the jth fatigue crack propagation process, δ j represents the crack propagation mode corresponding to the jth fatigue crack propagation process, υ j represents the ratio between the crack propagation mode corresponding to the jth fatigue crack propagation process and the modal damping ratio corresponding to the jth fatigue crack propagation process, e represents a constant, F(α) represents the displacement response vector corresponding to the αth group of single freedom propagation equations, dα represents the approximate consideration value corresponding to the αth group of single freedom propagation equations, represents the crack propagation length corresponding to the Tth time in the jth fatigue crack propagation process.

9. A detection alarm system for detecting surface defects of a pressure vessel, for implementing the detection alarm method for detecting surface defects of a pressure vessel according to any one of claims 1 to 8, characterized by, The system comprises: A defect detection and determination module is configured to obtain acoustic wave reflection information of the surface of the pressure vessel, identify a defect area existing on the surface of the pressure vessel based on the acoustic wave reflection information, and determine a characteristic state of the defect on the surface of the pressure vessel in combination with acoustic wave reflection intensity; A defect damage analysis module is configured to combine the characteristic state with a stress elastic-plastic analysis technique, obtain damage modes of different position points in the defect area on the surface of the pressure vessel, and determine damage representation parameters and damage evolution rules according to the damage modes; An alarm prompt control module is configured to evaluate a risk value of leakage of the pressure vessel according to the damage representation parameters and the damage evolution rules, generate an alarm prompt based on the risk value, and send the alarm prompt to a management terminal to perform risk control of the pressure vessel.

Images