Pressure vessel microcrack self-sensing and acoustic emission positioning detection system

Abstract

The invention relates to the technical field of pressure vessel detection, and discloses a pressure vessel microcrack self-sensing and acoustic emission positioning detection system, which comprises a wireless sensor network consisting of a plurality of wireless sensor nodes, and each node comprises an acoustic emission sensor, a wireless communication module, a microprocessor and a power supply; the data acquisition and preprocessing module is used for acquiring the acoustic emission signals, performing filtering, amplification and analog-to-digital conversion, and extracting characteristic parameters; and the intelligent identification and positioning module is used for classifying and identifying the acoustic emission signals by utilizing a machine learning algorithm and predicting a crack propagation path by combining finite element analysis and a digital twinborn technology. Through the combination of the wireless sensor network and the high-precision wireless synchronous clock, the problems of complex wiring and high cost of a traditional wired synchronization mode are solved. Through distributed deployment of the sensor nodes and use of the wireless communication module, the system can be quickly and flexibly applied to pressure vessel detection in various complex environments.

Description

Technical Field

[0001] The present invention relates to the technical field of pressure vessel detection, and particularly to a self-sensing and acoustic emission positioning detection system for microcracks in pressure vessels. Background Technique

[0002] The detection of microcracks in pressure vessels is an important link to ensure their safe operation. Traditional detection methods mainly rely on equipment such as ultrasonic flaw detectors, and sensors are manually operated to scan the surface of the pressure vessel. However, this method has the following problems: the scanning range of a single sensor is limited, and the sensor needs to be moved multiple times to cover the entire detection area. Repeating the same operation for a long time may cause operator fatigue and reduce work efficiency. In addition, the traditional wired synchronization method requires a large number of cables and equipment, which are complex to install and costly, and it is difficult to meet the detection requirements in complex environments.

[0003] In recent years, wireless sensor network technology has been widely used in the field of industrial detection. Wireless sensor networks have the characteristics of self-organization, dynamicity, and strong reliability, and can be flexibly deployed in various complex environments. However, existing wireless sensor networks still have deficiencies in positioning accuracy, energy management, and intelligence level. For example, the positioning accuracy is greatly affected by hardware conditions, and the ranging error will affect the accuracy of the positioning result. In addition, sensor nodes are usually powered by batteries, with limited energy, which restricts their ability for long-term online monitoring.

[0004] In the aspect of acoustic emission signal processing, the application of machine learning algorithms has gradually received attention. Existing research mainly focuses on signal denoising and identification, and methods such as the K-means clustering algorithm and wavelet analysis are used to improve the accuracy of signal processing. However, existing technologies still have deficiencies in crack propagation path prediction and intelligent detection, and it is difficult to meet the requirements of real-time monitoring and dynamic prediction.

[0005] In summary, existing technologies have obvious deficiencies in detection efficiency, intelligence level, and long-term online monitoring ability, and there is an urgent need for a new detection system that can solve these problems. Summary of the Invention

[0006] Aiming at the deficiencies of the existing technology, the present invention provides a self-sensing and acoustic emission positioning detection system for microcracks in pressure vessels, which solves the problems raised in the above background technique.

[0007] To achieve the above objectives, the present invention is realized through the following technical solutions: A self-sensing and acoustic emission positioning detection system for microcracks in pressure vessels, comprising:

[0008] A wireless sensor network, which consists of multiple wireless sensor nodes, and each node includes an acoustic emission sensor, a wireless communication module, a microprocessor, and a power supply;

[0009] A data acquisition and preprocessing module, which is used to acquire acoustic emission signals and perform filtering, amplification, and analog-to-digital conversion, and extract characteristic parameters;

[0010] An intelligent identification and positioning module, which uses machine learning algorithms to classify and identify acoustic emission signals, and combines finite element analysis and digital twin technology to predict the crack propagation path;

[0011] A data transmission and cloud platform integration module, which transmits data to the cloud platform in real time through 5G or Internet of Things technology, and supports remote monitoring and data analysis;

[0012] A user interface and report generation module, which provides an intuitive user interface, displays detection results, crack locations, and expansion predictions, and automatically generates a detection report;

[0013] Among them, the wireless sensor nodes achieve time synchronization through a high-precision wireless synchronous clock. Data is transmitted between sensor nodes through a wireless communication module. The data acquisition and preprocessing module sends the processed data to the intelligent identification and positioning module for analysis. The results of the intelligent identification and positioning module are sent to the cloud platform through the data transmission module. The user interface and report generation module obtains data from the cloud platform and displays it to the user.

[0014] Preferably, the wireless sensor nodes adopt a low-power design and are built-in with rechargeable batteries. The sensor nodes include acoustic emission sensors, wireless communication modules, microprocessors, and power management modules. The sensor nodes achieve time synchronization through a high-precision wireless synchronous clock to ensure the accurate acquisition of acoustic emission signals.

[0015] Preferably, the data acquisition and preprocessing module filters and amplifies the acquired acoustic emission signals, removes noise interference, and extracts characteristic parameters such as the amplitude, duration, energy, and frequency of the signals. The preprocessed signals are sent to the intelligent identification and positioning module through the wireless communication module.

[0016] Preferably, the intelligent identification and positioning module uses a convolutional neural network to classify acoustic emission signals. The convolutional neural network model is trained with a large amount of labeled data and can automatically learn the characteristic patterns of the signals. The input of the model is the preprocessed signal converted into a time-amplitude matrix, and the output of the model is the signal classification result.

[0017] Preferably, the crack location algorithm uses the time difference of arrival algorithm and combines sensor layout optimization to achieve precise crack location. Let the position of the sensor node be S i =(x i ,y i ,z i) The crack location is P=(x, y, z), the propagation speed of the acoustic emission signal in the medium is v, and by measuring the time difference Δt between different sensors ij =t i -t j , establish a system of equations:

[0018]

[0019] Use an optimization algorithm to solve the above system of equations to obtain the crack location P.

[0020] Preferably, the dynamic prediction model for crack propagation combines finite element analysis and machine learning algorithms, uses historical data and real-time monitoring data to predict the crack propagation path and rate. The model inputs are the initial crack location, propagation direction, material properties, and stress state, and the model outputs are the predicted path of crack propagation and the potential failure time.

[0021] Preferably, the user interface and the report generation module provide Web and mobile application interfaces, support real-time monitoring and historical data query, automatically generate inspection reports, provide maintenance suggestions, the user interface displays inspection results, crack locations, and expansion predictions, and the report generation module generates detailed inspection reports based on the analysis results.

[0022] Preferably, the data transmission and cloud platform integration module transmits data to the cloud platform in real time through 5G or Internet of Things technology. The cloud platform provides data storage, analysis, and visualization functions, integrates machine learning models and finite element analysis tools to achieve crack identification and expansion prediction.

[0023] Preferably, the power management module of the sensor node supports solar charging or external power supply to ensure the long-term stable operation of the system. The power management module extends the battery life of the sensor node by optimizing the hardware design and software algorithms and supports long-term online monitoring.

[0024] The present invention provides a pressure vessel micro-crack self-sensing and acoustic emission location detection system. It has the following beneficial effects:

[0025] 1. By combining a wireless sensor network and a high-precision wireless synchronous clock, the present invention solves the problems of complex wiring and high cost of traditional wired synchronization methods. The distributed deployment of sensor nodes and the use of wireless communication modules enable the system to be quickly and flexibly applied to the detection of pressure vessels in various complex environments. At the same time, the high-precision wireless synchronous clock ensures the accurate acquisition of acoustic emission signals, improving the accuracy of detection. In addition, the intelligent identification and positioning module uses a convolutional neural network to classify and identify acoustic emission signals, and combines finite element analysis and digital twin technology to predict the crack propagation path, further enhancing the intelligence level and accuracy of detection. This efficient and accurate detection method greatly reduces manual intervention, improves detection efficiency, and provides a reliable guarantee for the safe operation of pressure vessels.

[0026] 2. The present invention can monitor the microcrack state of a pressure vessel in real time, and through a dynamic prediction model of crack propagation, combined with finite element analysis and machine learning algorithms, use historical data and real-time monitoring data to predict the crack propagation path and rate. This real-time monitoring and prediction ability enables the system to timely discover potential safety hazards and provide a scientific basis for equipment maintenance and safety management. By providing the predicted path of crack propagation and the potential failure time, the system helps operators take preventive measures in advance and effectively prevent the sudden failure of pressure vessels.

[0027] 3. The present invention uses a wireless sensor network, reducing the large amount of cables and equipment required by traditional wired synchronization methods, and significantly reducing the system installation and maintenance costs. At the same time, this system is applicable to the detection of pressure vessels in various complex environments, especially in long-distance and high-complexity pipelines and containers, and has broad application value. This low-cost and high-applicability design enables the system to be widely applied to different industrial scenarios, providing an economical and efficient solution for the safety detection and maintenance of pressure vessels. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 is the system module interaction flowchart of the present invention;

[0029] Figure 2 is the acoustic emission signal processing flowchart of the present invention;

[0030] Figure 3 is the crack location algorithm flowchart of the present invention;

[0031] Figure 4 is the data transmission and cloud platform integration flowchart of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0032] Next, in conjunction with the accompanying drawings of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0033] Please refer to the attached Figure 1 - attached Figure 4 , the embodiment of the present invention provides a self-sensing and acoustic emission location detection system for microcracks in pressure vessels, including:

[0034] A wireless sensor network, composed of multiple wireless sensor nodes, each node includes an acoustic emission sensor, a wireless communication module, a microprocessor, and a power supply;

[0035] The wireless sensor node adopts a low-power design and is built-in with a rechargeable battery. The sensor node includes an acoustic emission sensor, a wireless communication module, a microprocessor, and a power management module. The sensor node realizes time synchronization through a high-precision wireless synchronous clock to ensure the accurate acquisition of acoustic emission signals. The low-power design is to extend the battery life of the sensor node so that it can support long-term online monitoring. The built-in rechargeable battery and power management module further optimize the energy usage. The high-precision wireless synchronous clock ensures that all nodes work under the same time reference, thereby improving the acquisition accuracy of acoustic emission signals and avoiding errors caused by time asynchronization.

[0036] Specifically, the wireless sensor network is a distributed network that can be flexibly deployed on the surface of the pressure vessel. Each node detects the acoustic emission signal generated by the microcrack through the acoustic emission sensor. The wireless communication module is used for data transmission between nodes. The microprocessor is responsible for signal processing, and the power supply provides energy for the node. This design enables the system to quickly adapt to complex environments and reduce wiring costs.

[0037] A data acquisition and preprocessing module, used to acquire acoustic emission signals and perform filtering, amplification, and analog-to-digital conversion, and extract characteristic parameters;

[0038] The data acquisition and preprocessing module filters and amplifies the acquired acoustic emission signals, removes noise interference, and extracts characteristic parameters such as the amplitude, duration, energy, and frequency of the signals. The preprocessed signals are sent to the intelligent identification and location module through the wireless communication module. The data acquisition module is responsible for transmitting the acoustic emission signals from the sensors to the system. The preprocessing module removes noise interference through filtering, amplifies the signals to improve the signal-to-noise ratio, and converts the analog signals into digital signals through analog-to-digital conversion. Feature extraction is to extract key parameters (such as amplitude, duration, energy, and frequency) from the processed signals to provide basic data for subsequent intelligent identification. Filtering and amplification are to improve the quality of the signals and ensure the accuracy of subsequent analysis. Characteristic parameters (such as amplitude, duration, energy, and frequency) are the key indicators for crack identification and location. The preprocessed signals are transmitted to the intelligent identification module through the wireless communication module for further analysis.

[0039] The intelligent identification and location module uses machine learning algorithms to classify and identify acoustic emission signals, and combines finite element analysis and digital twin technology to predict the crack propagation path; the machine learning algorithm (convolutional neural network) is used to automatically identify the crack characteristics in the acoustic emission signals. Finite element analysis is a numerical simulation method used to predict the crack propagation path in pressure vessels. Digital twin technology constructs a virtual model of the pressure vessel to reflect the crack propagation situation in real time, providing a scientific basis for equipment maintenance. This combination enables the system to achieve intelligent crack detection and prediction.

[0040] The crack location algorithm uses the time difference of arrival algorithm and combines sensor layout optimization to achieve precise crack location. Let the position of the sensor node be S i =(x i ,y i ,z i ), the crack position is P=(x,y,z), the propagation speed of the acoustic emission signal in the medium is v, and by measuring the time difference Δt ij =t i -t j , a system of equations is established:

[0041]

[0042] Use an optimization algorithm to solve the above system of equations to obtain the crack position P. The dynamic prediction model of crack propagation combines finite element analysis and machine learning algorithms, uses historical data and real-time monitoring data to predict the crack propagation path and rate. The model inputs are the initial crack position, propagation direction, material properties, and stress state, and the model outputs are the predicted path of crack propagation and the potential failure time.

[0043] Specifically, the dynamic prediction model for crack propagation combines finite element analysis (FEA) and machine learning algorithms. Using historical data and real-time monitoring data, it predicts the crack propagation path and rate. The model inputs include parameters such as the initial crack position, propagation direction, material properties, and stress state. The model output is the predicted crack propagation path and the potential failure time, providing a scientific basis for equipment maintenance and safety management.

[0044] Finite element analysis is used to simulate the crack propagation path under different stress conditions. The machine learning algorithm is trained through historical data to learn the laws of crack propagation, thereby achieving the dynamic prediction of crack propagation.

[0045] The intelligent identification and localization module classifies acoustic emission signals using a convolutional neural network. The convolutional neural network model is trained through a large amount of labeled data and can automatically learn the characteristic patterns of the signals. The model input is the preprocessed signal converted into a time-amplitude matrix, and the model output is the signal classification result.

[0046] The data transmission and cloud platform integration module transmits data to the cloud platform in real time through 5G or Internet of Things technology, supporting remote monitoring and data analysis. The data transmission and cloud platform integration module transmits data to the cloud platform in real time through 5G or Internet of Things technology. The cloud platform provides data storage, analysis, and visualization functions, integrating machine learning models and finite element analysis tools to achieve crack identification and propagation prediction.

[0047] Specifically, a convolutional neural network is a deep learning model suitable for processing image and signal data. It automatically extracts the characteristic patterns of the data through convolutional layers, reduces the data dimension through pooling layers, and performs classification through fully connected layers. The convolutional neural network model in the intelligent identification and localization module is trained through a large amount of labeled data. This data includes different types of acoustic emission signals and their corresponding crack characteristics, enabling the model to learn the association between signals and cracks. The preprocessed acoustic emission signals are converted into a time-amplitude matrix as the input of the CNN model. This matrix form can retain the time series information and amplitude changes of the signals, helping the model accurately identify signal characteristics. The model output is the signal classification result.

[0048] The real-time data transmission module uses 5G or Internet of Things technology to transmit the data collected by sensors to the cloud platform in real time. 5G technology provides high-speed and low-latency network connections, while Internet of Things technology supports the connection and data transmission of large-scale devices.

[0049] The cloud platform functions include the following:

[0050] Data storage: The cloud platform provides a large amount of storage space for saving historical data and real-time monitoring data.

[0051] Data analysis: The cloud platform integrates data analysis tools, enabling real-time processing and analysis of data to extract valuable information.

[0052] Data visualization: Through visualization tools, users can intuitively view information such as detection results, crack locations, and expansion predictions.

[0053] Integrated tools: The cloud platform integrates machine learning models and finite element analysis tools. The machine learning models are used to further analyze and predict the development trend of cracks, while the finite element analysis tools are used to simulate the expansion path of cracks under different stress conditions, providing a scientific basis for equipment maintenance.

[0054] User interface and report generation module, providing an intuitive user interface to display detection results, crack locations, and expansion predictions, and automatically generating detection reports;

[0055] The user interface and report generation module provides Web and mobile application interfaces, supports real-time monitoring and historical data querying, automatically generates detection reports, provides maintenance suggestions, the user interface displays detection results, crack locations, and expansion predictions, and the report generation module generates detailed detection reports based on the analysis results.

[0056] Specifically, the user interface includes the following functions:

[0057] Intuitive display: The user interface is designed to be simple and intuitive, enabling operators to easily view key information such as detection results, crack locations, and expansion predictions.

[0058] Real-time monitoring: Supports real-time monitoring functionality, allowing operators to view the current status of the pressure vessel at any time and promptly detect potential problems.

[0059] Historical data query: Provides historical data query functionality, allowing users to view past detection records for trend analysis and comparison.

[0060] Report generation includes the following functions:

[0061] Automatic report generation: Automatically generates a detailed detection report based on the analysis results. The report content includes detection time, crack location, expansion prediction, maintenance suggestions, etc.

[0062] Maintenance suggestions: Based on the detection results and analysis, provides specific maintenance suggestions to help operators take appropriate measures to prevent equipment failure.

[0063] Report format: The report can be generated in multiple formats (such as PDF, Excel, etc.) for easy user archiving, sharing, and further analysis.

[0064] The Web interface is accessed through a browser and is suitable for use in offices or fixed workstations, providing comprehensive functions and detailed data analysis. The mobile application interface supports smartphones and tablets, enabling operators to view test results and reports at any time on-site or while on the move, improving work efficiency. By providing detailed test reports and maintenance suggestions, it helps managers make scientific decisions, optimize equipment maintenance plans, and reduce downtime and maintenance costs.

[0065] Wireless sensor nodes achieve time synchronization through a high-precision wireless synchronous clock. Data is transmitted between sensor nodes through a wireless communication module. The data acquisition and preprocessing module sends the processed data to the intelligent identification and positioning module for analysis. The results of the intelligent identification and positioning module are sent to the cloud platform through the data transmission module. The user interface and report generation module obtains data from the cloud platform and presents it to the user.

[0066] Specifically, the data acquisition and preprocessing module is responsible for the preliminary processing of the acquired acoustic emission signals, including operations such as filtering, amplification, and analog-to-digital conversion, to remove noise interference and extract the key characteristic parameters of the signals. The processed data is then sent to the intelligent identification and positioning module through the wireless communication module. This module uses machine learning algorithms such as convolutional neural networks to classify and identify the signals, and at the same time combines finite element analysis and digital twin technology to predict the crack propagation path. The analysis results of the intelligent identification and positioning module are sent to the cloud platform through the data transmission module. The data transmission module uses 5G or Internet of Things technology to ensure that data can be transmitted in real time and efficiently. The cloud platform not only provides data storage and analysis functions, but also integrates machine learning models and finite element analysis tools, further enhancing the intelligence level and prediction ability of the system. Finally, the user interface and report generation module obtains data from the cloud platform and presents it to the user in an intuitive way. Users can monitor test results in real time, view crack locations and propagation predictions, and obtain detailed test reports and maintenance suggestions automatically generated by the system through the Web interface or mobile application interface. This modular design and efficient data flow ensure that the entire system can efficiently and accurately complete the detection and positioning tasks of micro-cracks in pressure vessels.

[0067] The power management module of the sensor node supports solar charging or external power supply to ensure the long-term stable operation of the system. The power management module extends the battery life of the sensor node by optimizing the hardware design and software algorithms, supporting long-term online monitoring.

[0068] Specifically, in terms of hardware design, the power management module maximizes the utilization of solar energy resources by adopting efficient solar panels and energy storage devices. At the same time, the module also has intelligent power management functions, which can dynamically adjust the power supply strategy according to the energy requirements of the sensor nodes to ensure the efficient use of energy. In terms of software algorithms, the power management module optimizes the algorithms to extend the battery life of the sensor nodes. For example, it can dynamically adjust the power consumption according to the working state of the sensor nodes (such as data acquisition, transmission, etc.) to reduce unnecessary energy consumption. In addition, the module also has an energy monitoring function, which can monitor the battery power and solar charging efficiency in real time to ensure that the sensor nodes can switch to an external power supply or enter a low-power mode in time when the energy is insufficient, thereby extending the online monitoring time.

[0069] This design not only improves the reliability and stability of the system, but also enhances its adaptability in complex environments, ensuring the continuity and accuracy of the detection of microcracks in pressure vessels.

[0070] To better demonstrate the performance and advantages of the pressure vessel microcrack self-sensing and acoustic emission localization detection system of the present invention, we conducted tests on multiple embodiments. These embodiments cover applications in laboratory environments, industrial sites, and complex environments, aiming to verify the detection efficiency, localization accuracy, intelligence level, and long-term monitoring capabilities of the system in different scenarios. By comparing with traditional methods, we can clearly see the significant advantages of the system of the present invention.

[0071] Example 2: Performance test in a laboratory environment

[0072] Experimental conditions

[0073] Pressure vessel model: A laboratory-level pressure vessel model with dimensions of 1m × 1m × 1m and made of common industrial steel.

[0074] Crack simulation: Artificial microcracks were made on the pressure vessel model, with crack lengths of 1 - 5 mm and depths of 0.5 - 2 mm.

[0075] Sensor deployment: 10 wireless sensor nodes were deployed, evenly distributed on the surface of the pressure vessel model.

[0076] Data acquisition: Acoustic emission signals were collected at a sampling rate of 1 MHz.

[0077] Experimental results

[0078] Detection efficiency: The system can complete the detection of the entire pressure vessel model within 5 minutes, while traditional methods require more than 30 minutes.

[0079] Localization accuracy: The crack localization accuracy reaches ±2 mm, while the localization accuracy of traditional methods is ±10 mm.

[0080] Intelligent level: The system can automatically identify cracks and predict the propagation path, while traditional methods require manual analysis.

[0081] Long-term monitoring ability: The system supports long-term online monitoring, and the battery life of sensor nodes reaches 30 days. Traditional methods require frequent battery replacement.

[0082] The test results are shown in Table 1.

[0083]

[0084]

[0085] Table 1

[0086] Example 3: Application in industrial site environment

[0087] Experimental conditions

[0088] Pressure vessel: An actual pressure vessel in a chemical plant, with dimensions of 5m × 3m × 3m and made of high-strength alloy steel.

[0089] Crack simulation: Micro-cracks are simulated on the pressure vessel, with crack lengths of 2 - 8mm and depths of 1 - 3mm.

[0090] Sensor deployment: 20 wireless sensor nodes are deployed at key parts of the pressure vessel.

[0091] Data acquisition: Acoustic emission signals are collected at a sampling rate of 2MHz.

[0092] Experimental results

[0093] Detection efficiency: The system can complete the detection of the entire pressure vessel within 15 minutes, while traditional methods require more than 2 hours.

[0094] Location accuracy: The crack location accuracy reaches ±3mm, and the location accuracy of traditional methods is ±15mm.

[0095] Intelligent level: The system can monitor the crack status in real time and predict the propagation path, while traditional methods require regular manual inspections.

[0096] Long-term monitoring ability: The system supports long-term online monitoring, and the battery life of sensor nodes reaches 60 days. Traditional methods require frequent maintenance.

[0097] The test results are shown in Table 2

[0098] Detection index The system of the present invention Traditional method Detection time (minutes) 15 >120 Positioning accuracy (mm) ±3 ±15 Intelligent level Real-time monitoring Regular inspection Endurance (days) 60 14

[0099] Table 2.

[0100] Example 4: Performance Verification in Complex Environments

[0101] Experimental Conditions

[0102] Pressure Vessel: A certain section of an oil pipeline, with a length of 100 m, a diameter of 0.5 m, and the material being corrosion-resistant alloy.

[0103] Crack Simulation: Simulate micro-cracks on the pipeline, with the crack length being 1 - 6 mm and the depth being 0.5 - 2.5 mm.

[0104] Sensor Deployment: Deploy 30 wireless sensor nodes, evenly distributed along the pipeline.

[0105] Data Acquisition: Acquire acoustic emission signals, with a sampling rate of 1.5 MHz.

[0106] Experimental Results

[0107] Detection Efficiency: The system can complete the detection of the entire pipeline section within 30 minutes, while the traditional method takes more than 4 hours.

[0108] Location Accuracy: The crack location accuracy reaches ±4 mm, and the location accuracy of the traditional method is ±20 mm.

[0109] Intelligent Level: The system can automatically identify cracks and predict the propagation path, while the traditional method requires manual analysis.

[0110] Long-term Monitoring Ability: The system supports long-term online monitoring, and the battery life of the sensor nodes reaches 90 days, while the traditional method requires frequent battery replacement.

[0111] The test results are shown in Table III.

[0112] Detection index The system of the present invention Traditional method Detection time (minutes) 30 >240 Positioning accuracy (mm) ±4 ±20 Intelligent level Automatic identification Manual analysis Endurance (days) 90 21

[0113] Table III.

[0114] Example 5: Multi-modal Data Fusion Performance Test

[0115] Experimental Conditions

[0116] Pressure Vessel Model: A laboratory-level pressure vessel model, with dimensions of 1 m × 1 m × 1 m and the material being common industrial steel.

[0117] Crack Simulation: Manually create micro-cracks on the pressure vessel model, with the crack length being 1 - 5 mm and the depth being 0.5 - 2 mm.

[0118] Sensor Deployment: Deploy 10 wireless sensor nodes, evenly distributed on the surface of the pressure vessel model.

[0119] Multi-modal Data Acquisition: Simultaneously acquire data such as acoustic emission signals, temperature, and pressure.

[0120] Experimental results

[0121] Detection efficiency: The system can complete the detection of the entire pressure vessel model within 5 minutes, while the traditional method takes more than 30 minutes.

[0122] Positioning accuracy: The crack positioning accuracy reaches ±2 mm, and the positioning accuracy of the traditional method is ±10 mm.

[0123] Intelligent level: The system can automatically identify cracks and predict the propagation path, while the traditional method requires manual analysis.

[0124] Data fusion effect: Through multi-modal data fusion, the crack recognition accuracy rate is increased to over 95%, and the accuracy rate of the traditional method is about 80%.

[0125] The test results are shown in Table IV.

[0126] Detection index The system of the present invention Traditional method Detection time (minutes) 5 >30 Positioning accuracy (mm) ±2 ±10 Intelligent level Automatic identification Manual analysis Recognition accuracy rate (%) >95 ～80

[0127] Table IV.

[0128] As can be seen from the above embodiments, the self-sensing and acoustic emission positioning detection system for micro-cracks in pressure vessels of the present invention is superior to the traditional method in terms of detection efficiency, positioning accuracy, intelligent level, and long-term monitoring ability, and has significant innovation and application value.

[0129] Although the embodiments of the present invention have been shown and described, it will be understood by those of ordinary skill in the art that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention, and the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A self-sensing and acoustic emission location detection system for microcracks in a pressure vessel, characterized in that Including: A wireless sensor network, composed of multiple wireless sensor nodes, each node including an acoustic emission sensor, a wireless communication module, a microprocessor, and a power supply; A data acquisition and preprocessing module, used to acquire acoustic emission signals and perform filtering, amplification, and analog-to-digital conversion, and extract characteristic parameters; An intelligent identification and positioning module, using machine learning algorithms to classify and identify acoustic emission signals, and combining finite element analysis and digital twin technology to predict crack propagation paths; A data transmission and cloud platform integration module, which transmits data to the cloud platform in real time through 5G or Internet of Things technology, supporting remote monitoring and data analysis; A user interface and report generation module, providing an intuitive user interface, displaying detection results, crack locations, and propagation predictions, and automatically generating detection reports; Among them, the wireless sensor nodes achieve time synchronization through a high-precision wireless synchronous clock, the sensor nodes transmit data through the wireless communication module, the data acquisition and preprocessing module sends the processed data to the intelligent identification and positioning module for analysis, the results of the intelligent identification and positioning module are sent to the cloud platform through the data transmission module, and the user interface and report generation module obtains data from the cloud platform and displays it to the user.

2. The self-sensing and acoustic emission location detection system for micro-cracks of a pressure vessel according to claim 1, wherein The wireless sensor nodes adopt a low-power design and are built with rechargeable batteries. Among them, the sensor nodes include an acoustic emission sensor, a wireless communication module, a microprocessor, and a power management module. The sensor nodes achieve time synchronization through a high-precision wireless synchronous clock to ensure the accurate acquisition of acoustic emission signals.

3. A self-sensing and acoustic emission location detection system for microcracks in pressure vessels according to claim 1, wherein, The data acquisition and preprocessing module filters and amplifies the acquired acoustic emission signals, removes noise interference, and extracts characteristic parameters such as the amplitude, duration, energy, and frequency of the signals. The preprocessed signals are sent to the intelligent identification and positioning module through the wireless communication module.

4. A self-sensing and acoustic emission location detection system for microcracks in a pressure vessel according to claim 1, characterized in that, The intelligent identification and positioning module classifies acoustic emission signals using a convolutional neural network. The convolutional neural network model is trained with a large amount of labeled data and can automatically learn the characteristic patterns of the signals. The input of the model is the preprocessed signal converted into a time-amplitude matrix, and the output of the model is the signal classification result.

5. A self-sensing and acoustic emission location detection system for micro-cracks in a pressure vessel according to claim 1, characterized in that, The crack location algorithm uses the time difference of arrival algorithm and combines it with optimized sensor layout to achieve precise crack location. Let the position of the sensor node be S i =(x i , y i , z i ), the crack position is P = (x, y, z), the propagation speed of the acoustic emission signal in the medium is v, and by measuring the time difference Δt ij = t i - t j , a system of equations is established: Use an optimization algorithm to solve the above equations to obtain the crack location P.

6. The self-sensing and acoustic emission location detection system for micro-cracks of a pressure vessel according to claim 1, characterized in that, The dynamic prediction model of crack propagation combines finite element analysis and machine learning algorithms, uses historical data and real-time monitoring data to predict the propagation path and rate of cracks. The input of the model is the initial crack location, propagation direction, material properties, and stress state, and the output of the model is the predicted path of crack propagation and the potential failure time.

7. A self-sensing and acoustic emission location detection system for microcracks in a pressure vessel according to claim 1, characterized in that The user interface and report generation module provides Web and mobile application interfaces, supports real-time monitoring and historical data query, automatically generates detection reports, provides maintenance suggestions, the user interface displays detection results, crack locations, and propagation predictions, and the report generation module generates a detailed detection report based on the analysis results.

8. A self-sensing and acoustic emission location detection system for microcracks in a pressure vessel according to claim 1, characterized in that, The data transmission and cloud platform integration module transmits data to the cloud platform in real time through 5G or Internet of Things technology. The cloud platform provides data storage, analysis, and visualization functions, integrates machine learning models and finite element analysis tools to achieve crack identification and propagation prediction.

9. The self-sensing and acoustic emission location detection system for micro-cracks of a pressure vessel according to claim 1, characterized in that, The power management module of the sensor node supports solar charging or external power supply to ensure the long-term stable operation of the system. By optimizing the hardware design and software algorithm, the power management module extends the battery life of the sensor node and supports long-term online monitoring.

Images