Pressure vessel monitoring system and method

Abstract

The present invention discloses a pressure vessel monitoring system, comprising: a sensing device for collecting acoustic signals from the surface of the pressure vessel; a data transmission device for sequentially performing noise filtering, defect type analysis, and signal feature screening on the acoustic signals, and transmitting the processed acoustic emission data to a data processing device; and a data processing device for analyzing the current damage state of the pressure vessel based on the acoustic emission data, thereby generating corresponding analysis results and realizing online monitoring of the damage state of the pressure vessel during operation. The present invention can provide real-time feedback on the damage status of the pressure vessel and avoid electromagnetic interference factors.

Description

Technical Field

[0001] The present invention relates to the field of pressure vessel monitoring, and in particular to a pressure vessel monitoring system and method. Background Art

[0002] A pressure vessel is a sealed container capable of withstanding pressure. Pressure vessels have a wide range of uses, playing a vital role in numerous sectors, including industry, civilian use, and military applications, as well as in numerous fields of scientific research. The chemical and petrochemical industries are the most heavily used, with pressure vessels in the petrochemical industry alone accounting for approximately 50% of all pressure vessels. In the chemical and petrochemical industries, pressure vessels are primarily used for heat transfer, mass transfer, reaction processes, and the storage and transportation of pressurized gases or liquefied gases. The media stored in pressure vessels are often flammable and explosive. Once damaged or defective, they can lead to serious accidents and environmental pollution, posing a threat to human life and property. Therefore, online health monitoring of operating pressure vessels to provide early warning of the formation and spread of damage and defects is crucial for preventing sudden structural failure and ensuring the safe operation of equipment.

[0003] Current pressure vessel monitoring methods, tailored to different monitoring requirements, primarily include ultrasonic guided wave monitoring, embedded strain sensor in-situ monitoring, and fiber Bragg grating (FBG) monitoring. Ultrasonic guided wave monitoring significantly reduces the monitoring range for pressure vessels with numerous defects, and monitoring primarily relies on echo signals. Weld unevenness directly impacts the accuracy of test results. Embedded strain sensor in-situ monitoring, due to its stringent sensor operating requirements, is difficult to apply to online pressure vessel monitoring. Fiber Bragg grating (FBG) monitoring is a point-based monitoring method and cannot monitor the entire pressure vessel's condition.

[0004] Because acoustic emission methods can detect overall structural defects in pressure vessels, they have been widely used to detect and determine defect status during regular offline hydrostatic testing of pressure vessels. However, due to the complex noise levels associated with on-site monitoring and the lack of effective damage mechanism signal processing and analysis models, defect signal extraction and identification are difficult, resulting in low reliability for system monitoring and early warning. Consequently, a real-time online acoustic emission monitoring and early warning method and system for pressure vessels has yet to be developed.

[0005] Therefore, in the prior art, there is a need to provide a solution that can perform real-time online health monitoring on pressure vessels during operation, so as to solve one or more of the above-mentioned technical problems. Summary of the Invention

[0006] In order to solve the above technical problems, an embodiment of the present invention provides a pressure vessel monitoring system, comprising: a sensing device, which is used to collect acoustic signals from the surface of the pressure vessel; a data transmission device, which is connected to the sensing device and is used to perform noise filtering processing, defect type analysis and signal feature screening processing on the acoustic signals in sequence, and transmit the processed acoustic emission data to a data processing device; the data processing device is used to analyze the damage status of the current pressure vessel based on the acoustic emission data, thereby generating corresponding analysis results, and realizing online monitoring of the damage status of the pressure vessel during operation.

[0007] Preferably, the sensing device comprises a plurality of acoustic emission sensors, and the plurality of acoustic emission sensors are evenly distributed at different positions on the outer surface of the pressure vessel.

[0008] Preferably, the data processing device is further used to receive acoustic emission data for different acoustic emission sensors, and extract the characteristics of the acoustic emission data per unit time, calculate the corresponding state evaluation value according to the acoustic emission data characteristics corresponding to the unit time, and use the state evaluation value to characterize the damage state analysis result.

[0009] Preferably, the pressure vessel monitoring system also includes: the alarm device, which is used to respond to the alarm instruction, wherein the data processing device is also used to determine whether the real-time damage status of the current pressure vessel meets the alarm condition based on the damage status analysis result, and if so, generate the alarm instruction.

[0010] Preferably, the acoustic emission data features include at least time domain features, frequency features, and entropy features including signal energy, impact intensity, ring count, and signal amplitude, wherein the data processing device calculates the state evaluation value according to the following steps: based on the acoustic emission data corresponding to each acoustic emission sensor, extracting the evaluation features of different acoustic emission data per unit time and calculating the average value of each feature; based on the average value of each evaluation feature, weighting the mean values ​​of different evaluation features to obtain the state evaluation value corresponding to the current unit time.

[0011] Preferably, the data processing device is further used to collect acoustic emission data that does not meet the alarm conditions, and use the acoustic emission data that does not meet the alarm conditions to calibrate the container defect recognition model required for the defect type analysis and processing process.

[0012] Preferably, the data transmission device is wirelessly connected to the data processing device and is used to perform noise filtering on the acoustic signal, and then use a preset container defect recognition model to determine whether the current pressure vessel has a crack defect and / or whether the crack defect has expanded based on the noise-filtered acoustic signal, and if so, extract abnormal acoustic signal segments, thereby transmitting the abnormal acoustic signal segments that meet the specified time-frequency entropy conditions to the data processing device.

[0013] On the other hand, an embodiment of the present invention also provides a pressure vessel monitoring method, which uses the pressure vessel detection system as described above to achieve long-term non-destructive monitoring of the pressure vessel, wherein the pressure vessel monitoring method includes the following steps: step 1, collecting acoustic signals from the surface of the pressure vessel; step 2, performing noise filtering processing, defect type analysis and signal feature screening processing on the acoustic signals in sequence, and transmitting the processed acoustic emission data to a data processing device; step 3, the data processing device analyzes the damage status of the current pressure vessel based on the acoustic emission data, thereby generating corresponding analysis results, and realizing online monitoring of the damage status of the pressure vessel during operation.

[0014] Preferably, in step three, acoustic emission data from different acoustic emission sensors are received, and features of the acoustic emission data per unit time are extracted, and corresponding state evaluation values ​​are calculated based on the features of the acoustic emission data corresponding to the unit time, so as to use the state evaluation values ​​to characterize the damage state analysis results.

[0015] Preferably, the pressure vessel monitoring method further includes: the data processing device determines whether the real-time damage status of the current pressure vessel meets the alarm condition based on the damage status analysis result, and if so, generates an alarm instruction; and the alarm device responds to the alarm instruction.

[0016] Compared with the prior art, one or more embodiments of the above solutions may have the following advantages or beneficial effects:

[0017] The present invention proposes a pressure vessel monitoring system and method. Based on acoustic emission technology, the system and method sequentially processes collected acoustic signals through a series of preliminary screening processes, including noise filtering, defect type analysis, and valid / invalid signal identification. This system and method generates acoustic emission data, which is transmitted to a data processing terminal via a wireless network. The system also determines the damage status of the pressure vessel in real time and issues an alarm when the damage status reaches a warning threshold, thereby forming a relatively complete monitoring system. Based on acoustic emission technology, the present invention provides a monitoring method specifically for pressure vessel integrity research. This method, specifically for pressure vessel integrity research, can perform online monitoring and analysis of defects, newly formed defects, or defect expansion at the sensor end, providing real-time feedback on the pressure vessel's damage status. This method avoids electromagnetic or electrical interference and structural influences, suppresses noise interference, accurately identifies defect status, effectively improves data transmission efficiency, remotely processes data, and provides real-time warnings based on the processing results. This reduces the inconvenience caused to operators by the pressure vessel itself and the surrounding environment, is easy to use, and has high reliability. Furthermore, the present invention is applicable to a variety of harsh environments, improves monitoring efficiency, and has broad application prospects. It provides effective technical support for the intelligent management of safe pressure vessel operation and lays the foundation for the development of intelligent, networked, and real-time pressure vessel management.

[0018] Other features and advantages of the present invention will be described in the following description, and in part will become apparent from the description, or will be understood by practicing the present invention. The purposes and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the description, claims and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0020] Figure 1 This is a schematic diagram of the overall structure of the pressure vessel monitoring system according to an embodiment of the present application.

[0021] Figure 2 This is a schematic diagram of the specific structure of the pressure vessel monitoring system according to an embodiment of the present application.

[0022] Figure 3 This is a step diagram of a pressure vessel monitoring method according to an embodiment of the present application. DETAILED DESCRIPTION

[0023] The following describes the embodiments of the present invention in detail with reference to the accompanying drawings and examples, so that the present invention can fully understand how to apply technical means to solve technical problems and achieve technical effects, and thus implement the invention accordingly. It should be noted that, as long as no conflict exists, the various embodiments of the present invention and the various features of the embodiments can be combined with each other, and the resulting technical solutions are all within the scope of protection of the present invention.

[0024] In addition, the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions. Also, although a logical order is shown in the flowchart, in some cases, the steps shown or described can be performed in a different order than here.

[0025] A pressure vessel is a sealed container capable of withstanding pressure. Pressure vessels have a wide range of uses, playing a vital role in numerous sectors, including industry, civilian use, and military applications, as well as in numerous fields of scientific research. The chemical and petrochemical industries are the most heavily used, with pressure vessels in the petrochemical industry alone accounting for approximately 50% of all pressure vessels. In the chemical and petrochemical industries, pressure vessels are primarily used for heat transfer, mass transfer, reaction processes, and the storage and transportation of pressurized gases or liquefied gases. The media stored in pressure vessels are often flammable and explosive. Once damaged or defective, they can lead to serious accidents and environmental pollution, posing a threat to human life and property. Therefore, online health monitoring of operating pressure vessels to provide early warning of the formation and spread of damage and defects is crucial for preventing sudden structural failure and ensuring the safe operation of equipment.

[0026] Current pressure vessel monitoring methods, tailored to different monitoring requirements, primarily include ultrasonic guided wave monitoring, embedded strain sensor in-situ monitoring, and fiber Bragg grating (FBG) monitoring. Ultrasonic guided wave monitoring significantly reduces the monitoring range for pressure vessels with numerous defects, and monitoring primarily relies on echo signals. Weld unevenness directly impacts the accuracy of test results. Embedded strain sensor in-situ monitoring, due to its stringent sensor operating requirements, is difficult to apply to online pressure vessel monitoring. Fiber Bragg grating (FBG) monitoring is a point-based monitoring method and cannot monitor the entire pressure vessel's condition.

[0027] Because acoustic emission methods can detect overall structural defects in pressure vessels, they have been widely used to detect and determine defect status during regular offline hydrostatic testing of pressure vessels. However, due to the complex noise levels associated with on-site monitoring and the lack of effective damage mechanism signal processing and analysis models, defect signal extraction and identification are difficult, resulting in low reliability for system monitoring and early warning. Consequently, a real-time online acoustic emission monitoring and early warning method and system for pressure vessels has yet to be developed.

[0028] Therefore, in order to solve one or more of the above-mentioned technical problems, an embodiment of the present invention proposes a pressure vessel monitoring system and method. The system and method are based on acoustic emission technology and include at least a sensing device, a data transmission device, and a data processing device. Specifically, the acoustic emission data obtained after local screening and processing of the collected acoustic emission signals at the sensing end is transmitted to the data processing terminal via a wireless network, thereby judging the real-time damage condition of the pressure vessel and determining whether the damage condition of the pressure vessel during the current use process has reached the alarm level. In addition, the system also includes a device for alarming when the current damage state reaches a dangerous level. In this way, the pressure vessel monitoring solution proposed in the embodiment of the present invention can perform long-term real-time health monitoring of pressure vessels during use, and can also work normally in harsh working environments.

[0029] Example 1

[0030] Figure 1 FIG. 1 is a schematic diagram of the structure of the pressure vessel monitoring system according to an embodiment of the present application. Figure 1 As shown, the pressure vessel monitoring system described in the embodiment of the present invention comprises at least: a sensor device 1, a data transmission device 2 and a data processing device 3. The sensor device 1 is used to collect acoustic signals on the surface of the pressure vessel. The data transmission device 2 is connected to the sensor device 1, and is used to receive the acoustic signals collected by the sensor device 1, perform noise filtering processing, defect type analysis (especially defect formation and / or expansion analysis) and signal feature screening processing on the acoustic signals in sequence, and transmit the processed acoustic emission data to the data processing device 3. The data processing device 3 is connected to the data transmission device 2, and is used to analyze the damage state of the current pressure vessel based on the acoustic emission data sent from the data transmission device 2, thereby generating corresponding analysis results, and realizing online monitoring of the damage state of the pressure vessel during operation. It should be noted that the pressure vessel as the monitoring object described in the embodiment of the present invention is a high-parameter pressure vessel that can withstand high temperature and high pressure.

[0031] In actual application, since acoustic emission detection is a dynamic detection method, it can provide real-time and continuous information on defects as the environment changes, and can quickly detect large components, as well as active defects that endanger structural safety, and provide dynamic information on defects. Therefore, the present invention uses this method as an important technology for pressure vessel monitoring, realizing long-term online real-time status monitoring of operating pressure vessels. In addition, the present invention also uses acoustic emission sensing technology to effectively avoid the influence of electromagnetic interference and structural factors, and collects, analyzes, and transmits acoustic emission signals of pressure vessel damage and defect status in real time and issues early warnings, with this as the goal to form an online acoustic emission monitoring solution for pressure vessels.

[0032] Figure 2This is a schematic diagram of the specific structure of the pressure vessel monitoring system of the embodiment of the present application. Figure 1 and Figure 2 The structure and functions of the pressure vessel monitoring system according to the embodiment of the present invention are described in detail.

[0033] In this embodiment of the present invention, the sensing device 1 includes a plurality of acoustic emission sensors. These acoustic emission sensors are evenly distributed at different locations on the outer surface of the pressure vessel and collect acoustic signals indicative of the damage state of the pressure vessel at the corresponding locations. It should be noted that this embodiment of the present invention does not specifically limit the number and distribution of acoustic emission sensors. Those skilled in the art may configure these sensors based on factors such as monitoring accuracy, the volume and shape of the pressure vessel to be monitored, and so on.

[0034] The acoustic emission sensor 1 can work in harsh environments and can withstand high temperatures and high pressures. It is made of stainless steel and can also withstand high temperatures and is not affected by electrical interference. In this way, the embodiment of the present invention can use multiple acoustic emission sensors to collect acoustic emission signals at different positions on the outer surface of the pressure vessel to be monitored as the data basis for realizing pressure vessel monitoring, so that the pressure vessel monitoring system can also perform normal damage status information collection work in harsh environments such as high temperature, high pressure and / or electromagnetic interference. For high-temperature pressure vessels, the acoustic emission sensor 1 contacts the pressure vessel through a waveguide rod, and a boss for arranging the acoustic emission sensor is provided on the waveguide rod. Two sensors can be arranged on the boss of each waveguide rod. The signals collected and analyzed by the two sensors can verify each other and play a redundant role.

[0035] Furthermore, in an embodiment of the present invention, the data transmission device 2 is connected to multiple acoustic emission sensors. The data transmission device 2 receives acoustic signals transmitted from different acoustic emission sensors, integrates each acoustic signal after source numbering and data screening (including noise filtering, defect type analysis, and signal feature screening), and generates acoustic emission data for the current monitoring moment after clock marking. The current acoustic emission data can then be used to determine and characterize the damage state of the pressure vessel.

[0036] Furthermore, during the data screening process of the data transmission device 2, the data transmission device 2 is configured to perform corresponding data screening processing on the acoustic signal of each acoustic emission sensor received in real time, clock-stamp the acoustic emission data after data screening processing, and transmit it to the data processing device 3. In the embodiment of the present invention, since the data screening processing method for the acoustic signal of each acoustic emission sensor is similar, the present invention uses the acoustic signal of one acoustic emission sensor as an example to illustrate the data screening processing flow.

[0037] Specifically, the data transmission device 2 is used to first perform noise filtering on the acoustic signal, and then use a preset container defect recognition model to determine whether a crack defect has occurred in the current pressure vessel and / or whether the crack defect has expanded based on the noise-filtered acoustic signal. If a crack defect has occurred, the data transmission device 2 extracts abnormal acoustic signal segments, and then wirelessly transmits the abnormal acoustic signal segments that meet the specified time-frequency entropy conditions to the data processing device 3 after time-stamping. The data transmission device 2 and the data processing device 3 are connected wirelessly. Therefore, in this embodiment of the present invention, the data transmission device 2 is also used to send the real-time generated acoustic emission data to the data processing device 3 via a wireless transmission link (in a wireless transmission manner).

[0038] Specifically, in this embodiment of the present invention, the data transmission device 2 performs noise filtering on the acoustic signal using a preset noise distribution pattern. This noise distribution pattern is a feature formed by pre-collecting the distribution patterns of various acoustic emission parameters corresponding to ambient sound signals (background noise and equipment noise during operation) in a stopped state.

[0039] Furthermore, in an embodiment of the present invention, the data transmission device 2 will also use a preset container defect recognition model to perform feature extraction processing on the current noise-filtered sound signal based on the noise-filtered sound signal to determine whether the current pressure vessel has a crack defect or whether the crack defect has expanded, and if so, extract abnormal sound signal segments.

[0040] In an embodiment of the present invention, a container defect recognition model is pre-built. The model is constructed according to the following steps: stress corrosion acoustic signals and corresponding stress corrosion types (stress corrosion types are defect types, including no cracks, cracks, and crack propagation, where crack propagation refers to the state in which the size of cracks generated per unit time continuously increases) collected simultaneously from historical stress corrosion evolution events that occurred in different pressure vessels are collected; signal features of the historical stress corrosion acoustic signals are extracted based on the acoustic damage mechanism of high-temperature and high-pressure vessel materials; and a neural network model is used to train the mapping relationship between the signal features of the stress corrosion acoustic signals and the stress corrosion types using a deep learning method, thereby obtaining a container defect recognition model.

[0041] Furthermore, the data transmission device 2 of the embodiment of the present invention is used to use the trained container defect recognition model to identify different situations of the current pressure vessel, such as no defects, crack defect generation, and crack propagation. When the stress corrosion type of the current pressure vessel is identified as crack defect generation or crack defect propagation, the acoustic signal segment (after noise filtering) corresponding to the current crack defect generation or crack defect propagation is determined to be an abnormal state, thereby extracting the corresponding abnormal acoustic signal segment. In addition, when the data transmission device 2 identifies the defect type of the current pressure vessel as no crack generation, there is no need to extract signal segments and the stress corrosion type of the continuously transmitted acoustic signals can be diagnosed.

[0042] Furthermore, the data transmission device 2 will also perform invalid signal filtering on the acoustic signal that has completed the noise filtering and defect type analysis processing to generate an acoustic signal that meets the specified amplitude and frequency conditions. Specifically, the data transmission device 2 is further used to determine the real-time time domain parameters (selected from one or more of signal energy, impact intensity, ring count and signal amplitude), real-time entropy value and real-time frequency of the current signal (current abnormal acoustic signal segment) based on the acoustic signal after completing the noise filtering and defect type analysis processing, and compare these three real-time data with the threshold threshold of the type (the time domain feature data of the corresponding category in the real-time time domain parameter is compared with the time domain parameter threshold threshold of the corresponding category, the real-time frequency is compared with the frequency threshold threshold, and the real-time entropy value is compared with the entropy threshold). Based on the comparison results, it is determined whether the current acoustic signal is an invalid signal. If it is an invalid signal, then at this time, the data transmission device 2 directly deletes the current acoustic signal (current abnormal acoustic signal segment); if it is a valid signal, then at this time, the data transmission device 2 directly integrates the current acoustic signal (current abnormal acoustic signal segment) with the valid acoustic signals generated by other sensors and transmits them to the data processing device 3 after time stamping. It should be noted that, in the invalid signal judgment process, the data transmission device 2 is further used to compare the real-time amplitude, real-time energy and real-time frequency corresponding to the current acoustic signal (after noise filtering and defect type analysis) with the amplitude threshold, energy threshold and frequency threshold respectively. If each type of real-time data reaches or exceeds the corresponding threshold (the time domain feature data of the corresponding category in the real-time time domain parameter reaches or exceeds the time domain parameter threshold of the corresponding category, and the real-time frequency reaches or exceeds the frequency threshold, and the real-time entropy value reaches or exceeds the entropy threshold), then the current acoustic signal (the current abnormal acoustic signal segment) is a valid signal. In addition, if any type of real-time data does not reach the corresponding threshold (the time domain feature data of the corresponding category in the real-time time domain parameter is lower than the time domain parameter threshold of the corresponding category, or the real-time frequency is lower than the frequency threshold, or the real-time entropy value is lower than the entropy threshold), then the current acoustic signal (the current abnormal acoustic signal segment) is an invalid signal.

[0043] Thus, the data transmission device 2 can first perform a corresponding preliminary screening process locally on the acoustic signal received from the sensor in real time, specifically screening out the effective acoustic signal that can be used to diagnose the damage state of the pressure vessel when a defect occurs or the defect expands, and after performing analog-to-digital conversion processing on the effective acoustic signal and marking the clock, generate acoustic emission data (at least including: the effective acoustic signal of the digital quantity and the current defect type), thereby utilizing its remote wireless transmission function to efficiently transmit the information sent by multiple acoustic emission sensors 1 to the data processing device 3, making data transmission more real-time and convenient, and making the subsequent container monitoring process more efficient. In this way, the embodiment of the present invention can perform online monitoring of the development of defects or newly formed defects or defects in pressure vessels, and transmit signal segments with key monitoring needs to the background for analysis, thereby taking into account the monitoring accuracy and monitoring efficiency of the online monitoring system.

[0044] The data processing device 3 is used to receive acoustic emission data from different acoustic emission sensors, extract acoustic emission data features per unit time from the acoustic emission data, and then perform calculations based on the extracted acoustic emission data features corresponding to the unit time.

[0045] Furthermore, the data processing device 3 is implemented using a remote server. In this embodiment of the present invention, the data processing device 3 is configured to receive acoustic emission data from different acoustic emission sensors, extract the characteristics of the acoustic emission data per unit time, and calculate a state evaluation value for the corresponding time period based on the corresponding acoustic emission data characteristics per unit time. The state evaluation value is then used to characterize the damage state analysis results, thereby characterizing the dynamic damage state of the pressure vessel using the continuously generated state evaluation values. In this way, the data processing device 3 is able to receive and analyze remotely transmitted acoustic emission data, thereby achieving continuous real-time monitoring of the pressure vessel during operation.

[0046] Furthermore, after analyzing and processing the acoustic emission data received in real time and obtaining the corresponding status evaluation value representing the analysis result, the data processing device 3 will also perform a level evaluation based on the status evaluation value generated in real time to determine whether the current damage state of the pressure vessel has reached the conditions requiring an alarm prompt, thereby prompting the operator when the pressure vessel is damaged to a certain extent.

[0047] Specifically, the data processing device 3 is also used to determine whether the real-time damage state of the current pressure vessel reaches the alarm condition based on the damage state analysis result generated in real time. If it reaches the alarm condition, an alarm instruction is generated, and the alarm device 4 responds to the instruction. The data processing device 3 is further used to compare the real-time generated damage state analysis result with the preset damage alarm threshold. When the current damage state analysis result (state evaluation value) exceeds the damage alarm threshold, it is determined that the alarm condition is currently reached. At this time, an alarm instruction is immediately generated, and the alarm device 4 responds to the instruction. Therefore, reference Figure 2 The pressure vessel monitoring system according to the embodiment of the present invention further includes an alarm device 4. The alarm device 4 is configured to respond to the alarm instruction generated by the data processing device 3, thereby notifying the pressure vessel operator of abnormal damage to the pressure vessel and indicating that the current damage level is about to reach a dangerous state.

[0048] It should be noted that the above-mentioned damage alarm threshold is the standard quantitative value of the status evaluation value corresponding to when the real-time damage status of the pressure vessel reaches the alarm condition. The present invention does not specifically limit this threshold, and those skilled in the art can set it according to factors such as evaluation accuracy.

[0049] Furthermore, in an embodiment of the present invention, the data processing device 3 is further configured to, after generating an alarm instruction, utilize a pre-set warning level damage threshold sequence to continuously diagnose the alarm level corresponding to the current damage state analysis result (state evaluation value) and generate corresponding alarm level information. In this embodiment of the present invention, the warning level damage threshold sequence includes multiple damage state evaluation thresholds, with the warning level representing the real-time damage level corresponding to the current pressure vessel after an alarm condition is met. Furthermore, each damage state evaluation threshold is greater than the damage alarm threshold, and each damage state evaluation threshold is arranged in ascending order to form a warning level damage threshold sequence. The evaluation range corresponding to each warning level is formed by adjacent damage state evaluation thresholds in the warning level damage threshold sequence. Thus, when diagnosing the alarm levels corresponding to the continuously generated state evaluation values, the data processing device 3 is configured to obtain corresponding alarm level information by determining the warning level evaluation range within which the currently generated state evaluation value falls.

[0050] Furthermore, after generating the corresponding alarm level information, the data processing device 3 is further configured to generate a corresponding early warning report based on the current alarm level information and the state evaluation value corresponding to the current alarm event. The early warning report is then sent to the corresponding terminal device of the pressure vessel operator and / or the relevant person in charge of the enterprise. Simultaneously, the currently generated alarm instruction is transmitted to the central control system, so that the central control system activates the on-site alarm bell under the control of the current alarm instruction. The early warning report includes, but is not limited to, a vessel status image containing the current pressure vessel damage status characteristics, an on-site picture containing the current pressure vessel scene, alarm level information, and predetermined solutions.

[0051] In this way, the data processing device 3 can not only realize the monitoring of the pressure vessel under real-time analysis, but also automatically send an alarm instruction to the alarm device 4 when the analysis result meets the danger alarm condition, so as to inform the operator and / or relevant person in charge of the abnormal condition of the current pressure vessel in advance. The operator and / or relevant person in charge will take corresponding measures according to the early warning report, thereby greatly reducing the harm of the surrounding environment to the operator.

[0052] In addition, when the data processing device 3 detects that the current damage status analysis result (status evaluation value) does not exceed the above-mentioned damage alarm threshold, the real-time damage status of the current pressure vessel has not reached the alarm condition. At this time, it indicates that the damage status of the current pressure vessel is still in a safe state and there is no need to alarm. The status evaluation value generated in the next unit time period is compared with the damage alarm threshold under the alarm condition, and the real-time damage status analysis result is continued to be evaluated.

[0053] Furthermore, in an embodiment of the present invention, the data processing device 3 is further configured to calculate a state evaluation value for the entire pressure vessel based on the acoustic emission data corresponding to each acoustic emission sensor, so as to use the state evaluation value to characterize the current damage state of the pressure vessel. Specifically, first, based on the acoustic emission data corresponding to each acoustic emission sensor, the evaluation features of different acoustic emission data within a unit time period are extracted and the average value of each evaluation feature is calculated. In actual application, the data processing device 3 is configured to continuously receive the acoustic emission data corresponding to each acoustic emission sensor, and then, for the acoustic emission data continuously transmitted by each acoustic emission sensor (wherein the acoustic emission data of each sensor is defined as a group of acoustic emission data), extract different evaluation features of the corresponding group of acoustic emission data within a unit time period from each group of acoustic emission data, and finally calculate the average value of each evaluation feature, thereby obtaining quantitative data of each evaluation feature within a unit time period based on the acoustic emission data continuously transmitted by each acoustic emission sensor. In an embodiment of the present invention, the evaluation features include at least: time domain feature parameters including signal energy (evaluation feature), impact intensity (evaluation feature), ring count (evaluation feature), and signal amplitude (evaluation feature), frequency domain evaluation features, and entropy value evaluation features. For example, after receiving the acoustic emission data corresponding to each acoustic emission sensor, the data processing device 3 extracts the signal energy evaluation feature, impact intensity evaluation feature, ring count evaluation feature, signal amplitude evaluation feature, frequency domain evaluation feature and entropy value feature corresponding to each group of acoustic emission data within each second based on the acoustic emission data received at each monitoring moment within each second, and calculates the average value of each time-frequency entropy feature parameter.

[0054] Next, the data processing device 3 is further configured to weight the mean values ​​of the different evaluation features based on the average values ​​of each evaluation feature within the unit time period to obtain a state evaluation value corresponding to the current unit time period. It should be noted that the embodiments of the present invention do not specifically limit the proportional distribution of weight values ​​corresponding to different evaluation features. Those skilled in the art may determine the importance of these evaluation features for damage state evaluation.

[0055] Specifically, the data processing device 3 is primarily an integrated operating system for acoustic emission data processing and early warning. It includes not only the aforementioned data analysis and monitoring and alarm functions, but also data storage. Specifically, the data storage function uses the monitoring time as the file name and the monitored object as the storage unit. The stored content includes acoustic emission data (i.e., acoustic emission monitoring data files), on-site photos, and file descriptions (including sensor layout plans, basic pressure vessel parameters, etc.).

[0056] In addition, the data analysis function also includes: by continuously acquiring (real-time acquisition) the acoustic emission data, exporting the time history diagram, scatter diagram, waveform diagram, etc. of the acoustic emission data to facilitate users to intuitively understand the real-time changes in the monitoring data.

[0057] Example 2

[0058] Based on the above-mentioned embodiment 1, in order to improve the accuracy of local data screening of the pressure vessel monitoring system described in the present invention, the embodiment of the present invention also needs to use the data processing device 3 to verify the consistency between the defect type of the acoustic emission data with defect generation or defect expansion transmitted to the background server and the damage status analysis result (status evaluation value), so as to achieve optimization correction of the container defect recognition model.

[0059] Specifically, the data processing device 3 is also used to collect acoustic emission data that does not meet the alarm conditions and the corresponding damage status analysis results and alarm level information, and use these acoustic emission data that meet the alarm conditions locally but do not meet the alarm conditions through analysis by the background server to optimize and train the container defect recognition model currently required for implementing the defect type analysis and processing process locally, thereby obtaining a corrected container defect recognition model.

[0060] In this way, the embodiment of the present invention performs secondary analysis of the alarm conditions of the acoustic emission data with alarm conditions that are screened locally through the background server, which not only completes the accurate analysis and verification of the alarm status and verifies the accuracy of the local screening, but also can correct the container defect recognition model required for local data screening.

[0061] Furthermore, in order to further improve the accuracy of the container defect recognition model and complete more accurate local data screening and processing, the data processing device 3 described in the embodiment of the present invention is also used to periodically obtain the normal acoustic signal segments of the digital quantity corresponding to the case where the defect type sent by the data transmission device 2 is no defect, and calculate the state evaluation value of these normal acoustic signal segments, and further determine the acoustic emission segments that meet the alarm conditions among these normal acoustic signal segments of the digital quantity. Then, the acoustic emission segment data that have not been screened out locally but have met the alarm conditions through analysis by the backend server, as well as the acoustic emission data that have been screened out locally but have met the alarm conditions through analysis by the backend server, are used to optimize and train the container defect recognition model currently required for implementing the defect type analysis and processing process, thereby obtaining a corrected container defect recognition model.

[0062] In this way, the present invention implements a secondary analysis of the alarm conditions of the acoustic emission data with alarm conditions that have been screened locally through the background server, which not only completes the accurate analysis and verification of the alarm status and verifies the accuracy of the local screening, but also verifies the accuracy of the local acoustic emission data that has not been screened out. Moreover, based on the above two verification results, the container defect recognition model required for local data screening is further corrected, thereby further improving the local screening accuracy of the container defect recognition model.

[0063] Furthermore, the embodiment of the present invention can also verify the effectiveness of the data transmission device 2 in judging defect formation or defect expansion. If it is effective, it is confirmed that the early warning conditions are met and the alarm device is activated to take corresponding measures. If it is not effective, the container defect recognition model used for front-end analysis is corrected, and the reliability of the container defect recognition model is continuously optimized and analyzed.

[0064] Example 3

[0065] On the other hand, based on the pressure vessel monitoring system described in the above-mentioned embodiment 1 and / or embodiment 2, an embodiment of the present invention also proposes a pressure vessel monitoring method, which uses the above-mentioned pressure vessel detection system to achieve long-term non-destructive monitoring of the pressure vessel. Figure 3 This is a step diagram of a pressure vessel monitoring method according to an embodiment of the present application.

[0066] like Figure 3 As shown, the pressure vessel monitoring method described in the embodiment of the present invention includes the following steps: step S110 collects the acoustic signal of the pressure vessel surface; step S120 performs noise filtering processing, defect type analysis and signal feature screening processing on the acoustic signal transmitted from step S110 in sequence, and transmits the processed acoustic emission data to a data processing device; step S130 the data processing device analyzes the damage status of the current pressure vessel according to the received acoustic emission data, thereby generating corresponding analysis results, and realizing online monitoring of the damage status of the pressure vessel during operation.

[0067] Furthermore, in step S120, it includes: performing noise filtering processing on the acoustic signal transmitted in step S110, and then using a preset container defect recognition model to determine whether the current pressure vessel has a crack defect and / or whether the crack defect has expanded based on the noise-filtered acoustic signal, and if so, extracting abnormal acoustic signal segments, thereby transmitting the abnormal acoustic signal segments that meet the specified time-frequency entropy conditions to the data processing device.

[0068] Furthermore, in step S130, it includes: a data processing device receives acoustic emission data for different acoustic emission sensors, and extracts features of the acoustic emission data per unit time, and calculates corresponding state evaluation values ​​according to the acoustic emission data features corresponding to the unit time, so as to use the current state evaluation value to characterize the damage state analysis result.

[0069] In addition, the pressure vessel monitoring method according to an embodiment of the present invention further includes: the data processing device 3 determines whether the current real-time damage status of the pressure vessel meets the alarm condition based on the real-time damage status analysis results. If so, an alarm instruction is generated. In this case, the alarm device 4 responds to the alarm instruction, thereby indicating that the current damage level is about to reach a dangerous state.

[0070] Embodiments of the present invention provide a pressure vessel monitoring system and method. Based on acoustic emission technology, the system and method sequentially process collected acoustic signals through a series of preliminary screening processes, including noise filtering, defect type analysis, and valid / invalid signal identification. This system and method generates acoustic emission data, which is transmitted to a data processing terminal via a wireless network. The system also determines the damage status of the pressure vessel in real time and issues an alarm when the damage status reaches a warning threshold, thereby forming a relatively comprehensive monitoring system. Based on acoustic emission technology, the present invention provides a monitoring method specifically for pressure vessel integrity research. This method can monitor defects, newly formed defects, or defect expansion in a vessel online, and provides real-time feedback on the damage status of the pressure vessel. This method avoids electromagnetic or electrical interference, effectively improves data transmission efficiency, remotely processes data, and provides real-time warnings based on the processing results. This method reduces the adverse effects of the pressure vessel itself and the surrounding environment on operators, is easy to use, and highly reliable. Furthermore, the present invention is applicable to a variety of harsh environments, improves monitoring efficiency, and has broad application prospects. It provides effective technical support for intelligent pressure vessel management and provides a foundation for the development of intelligent, networked, and real-time pressure vessel management.

[0071] The above description is merely a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by anyone skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

[0072] It should be understood that the embodiments disclosed herein are not limited to the specific structures, processing steps, or materials disclosed herein, but should extend to equivalent substitutions of these features understood by those skilled in the relevant art. It should also be understood that the terminology used herein is for the purpose of describing specific embodiments only and is not intended to be limiting.

[0073] References in this specification to "one embodiment" or "an embodiment" mean that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the present invention. Therefore, appearances of the phrases "one embodiment" or "an embodiment" in various places throughout this specification do not necessarily refer to the same embodiment.

[0074] Although the embodiments disclosed above are for facilitating understanding of the present invention, the contents described are merely embodiments adopted for facilitating understanding of the present invention and are not intended to limit the present invention. Any person skilled in the art of the present invention may make any modifications and changes in the form and details of the implementation without departing from the spirit and scope of the present invention. However, the scope of patent protection of the present invention shall still be subject to the scope defined by the appended claims.

Claims

1. A pressure vessel monitoring system, characterized in that: The pressure vessel monitoring system comprises: A sensing device for collecting acoustic signals from the surface of the pressure vessel; A data transmission device, connected to the sensing device, for sequentially performing noise filtering processing, defect type analysis, and signal feature screening processing on the acoustic signal, and transmitting the processed acoustic emission data to a data processing device; The data processing device is used to analyze the damage state of the current pressure vessel based on the acoustic emission data, thereby generating corresponding analysis results and realizing online monitoring of the damage state of the pressure vessel during operation, wherein: The data transmission device is connected to the data processing device via wireless means and is used to generate acoustic emission data according to the following process: The acoustic signal is subjected to noise filtering processing using a preset noise distribution law, wherein the noise distribution law is a feature formed by pre-collecting the distribution law of various acoustic emission parameters corresponding to the ambient sound signal including background noise in a shutdown state and equipment body noise during operation; Based on the noise-filtered acoustic signal, using a vessel defect recognition model established based on a neural network model, different stress corrosion types of the current pressure vessel, including no defect generation, crack defect generation, and crack propagation, are identified; and when the stress corrosion type of the current pressure vessel is identified as crack defect generation or defect propagation, a noise-filtered acoustic signal segment corresponding to the current crack defect generation or crack defect propagation is determined as an abnormal acoustic signal segment; and The real-time time domain parameters, real-time frequency, and real-time entropy value of the current abnormal sound signal segment are determined, and the three real-time data are compared with the corresponding threshold values ​​respectively. Based on the comparison results, it is determined whether the current abnormal sound signal segment is an invalid signal, so as to transmit the acoustic emission data formed by integrating the valid abnormal sound signal segments generated by all sensors to the data processing device. If each type of real-time data reaches or exceeds the corresponding threshold value, the current abnormal sound signal segment is a valid signal; if any type of real-time data does not reach the corresponding threshold value, the current abnormal sound signal segment is an invalid signal.

2. The pressure vessel monitoring system according to claim 1, characterized in that: The sensing device includes a plurality of acoustic emission sensors, which are evenly distributed at different positions on the outer surface of the pressure vessel.

3. The pressure vessel monitoring system according to claim 1, characterized in that: The pressure vessel monitoring system further includes: An alarm device, which is used to respond to an alarm instruction, wherein: The data processing device is further configured to determine whether the current real-time damage state of the pressure vessel has reached an alarm condition based on the damage state analysis result; if the current state evaluation value representing the damage state analysis result exceeds the damage alarm threshold, it is determined that the alarm condition has been reached, thereby generating the alarm instruction, wherein: The data processing device is further configured to receive and obtain acoustic emission data corresponding to different acoustic emission sensors, extract different evaluation features of different groups of acoustic emission data per unit time from the acoustic emission data of each acoustic emission sensor, calculate an average value of each evaluation feature, and then perform weighted processing on the average values ​​of the different evaluation features to obtain a state evaluation value corresponding to a unit time, so as to use the state evaluation value to characterize the damage state analysis result of the pressure vessel, wherein the evaluation features at least include time domain features including signal energy, impact intensity, ring count and signal amplitude, frequency features and entropy features.

4. The pressure vessel monitoring system according to claim 1 or 3, characterized in that: The data processing device is further used to collect acoustic emission data that does not meet the alarm conditions, and use the acoustic emission data that does not meet the alarm conditions to calibrate the container defect recognition model required for the defect type analysis and processing process.

5. A pressure vessel monitoring method, characterized in that: The method utilizes the pressure vessel monitoring system according to any one of claims 1 to 4 to implement long-term non-destructive monitoring of a pressure vessel, wherein the pressure vessel monitoring method comprises the following steps: Step 1: collecting acoustic signals on the surface of the pressure vessel; Step 2: performing noise filtering processing, defect type analysis and signal feature screening processing on the acoustic signal in sequence, and transmitting the processed acoustic emission data to a data processing device; Step three: the data processing device analyzes the damage state of the current pressure vessel based on the acoustic emission data, thereby generating corresponding analysis results, and realizing online monitoring of the damage state of the pressure vessel during operation.

6. The pressure vessel monitoring method according to claim 5, characterized in that: In the step three, Acoustic emission data from different acoustic emission sensors are received, and features of the acoustic emission data per unit time are extracted. A corresponding state evaluation value is calculated based on the features of the acoustic emission data corresponding to the unit time, so as to characterize the damage state analysis result using the state evaluation value.

7. The pressure vessel monitoring method according to claim 6, characterized in that: The pressure vessel monitoring method further includes: The data processing device determines whether the current real-time damage state of the pressure vessel reaches an alarm condition based on the damage state analysis result, and generates an alarm instruction if so; The alarm device responds to the alarm instruction.

Images