Intelligent hanger crack diagnosis system based on acoustic emission
Through the intelligent diagnosis system of hanger cracks based on acoustic emission, the monitoring point analysis and acoustic emission sensor array are combined with the neural network model to solve the accuracy and real-time problems of hanger crack detection, and achieve the improvement of hanger safety and reliability.
Patent Information
- Application Number
- CN202510965601.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-14
- Publication Date
- 2025-09-16
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The existing technology for hanger crack detection has the following problems: low detection efficiency, difficulty in detecting early-stage tiny cracks, inability to monitor crack development dynamics in real time, inaccurate acoustic emission signal acquisition, susceptibility to interference from environmental noise, and lack of intelligent diagnosis and early warning mechanisms.
An intelligent diagnosis system for hanger cracks based on acoustic emission is adopted, which includes a historical hanger crack collection module, a crack database, a monitoring point analysis module, an acoustic emission sensor array, a signal acquisition and processing module, a priori diagnostic knowledge base, an intelligent diagnosis module and an early warning module. The monitoring point analysis module is used to mark key stress-bearing parts, and an acoustic emission sensor array is arranged for signal collection. A crack diagnosis model is constructed by combining convolutional neural networks and recurrent neural networks to perform crack classification, severity assessment and development trend prediction.
The accuracy of acoustic emission signal acquisition is improved, intelligent diagnosis and early warning of hanger cracks are realized, the safety of hanger operation is improved, and production accidents are avoided.
Smart Images
Figure CN120651971A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of hanger online detection, and in particular to an intelligent diagnosis system for hanger cracks based on acoustic emission. Background Art
[0002] The oil machinery hanger is a key equipment used in the oil industry to support and fix casing, tubing and other pipe strings. It can effectively support the weight of the tubing, ensure its stability in the well, prevent oil and gas leakage, and ensure production safety.
[0003] Hangers are prone to cracking under long-term exposure to alternating loads, high temperatures and pressures, and complex corrosive environments. Traditional crack detection methods, such as visual inspection, magnetic particle testing, and penetrant testing, suffer from low efficiency, difficulty detecting early-stage microcracks, and inability to monitor crack development in real time. These methods are unable to meet the stringent requirements for hanger safety and reliability in modern industry.
[0004] Acoustic emission (AE) technology, as a dynamic nondestructive testing technique, can capture in real time the elastic wave signals generated by materials or structures during stress or deformation, thereby revealing their internal damage state. However, the current application of AE technology in hanger crack diagnosis faces several bottlenecks, such as inaccurate AE signal acquisition, susceptibility to environmental noise interference, inaccurate determination of crack type and severity, and the lack of effective intelligent diagnosis and early warning mechanisms. Therefore, this paper proposes an AE-based intelligent hanger crack diagnosis system to address these shortcomings. Summary of the Invention
[0005] In response to the above problems, the purpose of the present invention is to provide an intelligent diagnosis system for hanger cracks based on acoustic emission. By utilizing a monitoring point analysis module to analyze the crack elastic wave signal according to the quality inspection parameter content and the historical damage hanger, the location of the hanger prone to cracks is determined, and the key stress-bearing parts on the hanger are marked. Then, an acoustic emission sensor array is arranged for signal acquisition, which can effectively improve the accuracy of acoustic emission signal acquisition. The improved acoustic emission signal acquisition quality is conducive to the intelligent diagnosis module to perform crack classification, severity assessment and crack development trend prediction.
[0006] To achieve the above object, the present invention provides the following technical solutions:
[0007] An intelligent diagnostic system for hanger cracks based on acoustic emission, comprising a historical hanger crack collection module, a crack database, a hanger quality inspection report library, a monitoring point analysis module, an acoustic emission sensor array, a signal acquisition and processing module, a priori diagnostic knowledge base, an intelligent diagnosis module, and an early warning module;
[0008] The historical hanger crack acquisition module is used to collect crack elastic wave signals of historically damaged hangers using an acoustic emission sensor;
[0009] The crack database is used to store the crack elastic wave signals of the historical damaged hanger;
[0010] The hanger quality inspection knowledge base is used to store the quality inspection parameter content of the hanger before application;
[0011] The monitoring point analysis module is used to analyze the crack elastic wave signal based on the quality inspection parameter content and the history of damaged hangers to find the location of the hanger prone to cracks, and to mark the key stress-bearing parts on the hanger;
[0012] The acoustic emission sensor array is composed of a plurality of acoustic emission sensors, and the plurality of acoustic emission sensors are distributed and installed at locations prone to cracks and key stress-bearing locations on the hanger.
[0013] The signal acquisition and processing module is used to convert the real-time crack elastic wave signal collected by the acoustic emission sensor into an electrical signal, amplify it through a signal amplifier, and then perform signal noise reduction and feature extraction processing to obtain an acoustic emission signal;
[0014] The priori diagnosis knowledge base is used to store historical crack diagnosis experience of the hanger;
[0015] The intelligent diagnosis module builds a crack diagnosis model based on convolutional neural network (CNN) and recurrent neural network (RNN). After the crack diagnosis model is trained, the characteristic parameters of the acoustic emission signal obtained by the signal acquisition and processing module are input into the crack diagnosis model to perform crack classification, severity assessment and crack development trend prediction;
[0016] The early warning module is used to set a threshold, then compare the intelligent diagnosis with the threshold to obtain early warning data, and perform automatic early warning based on the early warning data.
[0017] A further improvement is that when the crack database stores the elastic wave signals of cracks of historically damaged hangers, the elastic wave signals of cracks of historically damaged hangers are converted into electrical signals to form crack data signals with clear corresponding relationships for storage.
[0018] A further improvement is that the quality inspection parameter content includes a hanger strength distribution diagram and a damage distribution diagram.
[0019] A further improvement is that the monitoring point analysis module analyzes the locations of the hangers that are prone to cracks based on the quality inspection parameter content and the crack elastic wave signals of historically damaged hangers, and is used to mark the key stress-bearing parts on the hangers. Specifically, the module cross-compares and analyzes the hanger strength distribution map, the damage distribution map, and the crack locations corresponding to the crack elastic wave signals of historically damaged hangers to analyze and determine the locations of multiple hangers that are prone to cracks.
[0020] A further improvement is that after the multiple acoustic emission sensors of the acoustic emission sensor array are installed, a shell is further installed at the installation position of the acoustic emission sensor to protect the acoustic emission sensor.
[0021] A further improvement is that when the signal acquisition and processing module performs signal noise reduction processing, specifically: in response to various noise interferences existing in the actual working environment of the hanger, one or a combination of wavelet transform filtering and adaptive filtering is used to reduce the noise of the collected signal to improve the signal-to-noise ratio.
[0022] A further improvement is that when the signal acquisition and processing module extracts and processes the signal features, it specifically extracts multiple characteristic parameters from the noise-reduced acoustic emission signal, including amplitude, energy, count, duration, rise time, and frequency components.
[0023] Further improvements are as follows: during the training process of the crack diagnosis model: a large number of hanger crack acoustic emission signal samples of different types and severity are collected, and after labeling and processing them, the collected data are divided into training set, validation set and test set according to a certain ratio. The training set is used to train the intelligent diagnosis model, the validation set is used to adjust the model's hyperparameters and evaluate the model's performance during the training process, and the test set is used to evaluate the model's generalization ability.
[0024] A further improvement is that when the warning module sets the threshold, the threshold setting value is comprehensively analyzed by analyzing the historical crack diagnosis experience of the hanger in the prior diagnosis knowledge base, and the threshold setting value is numerically adjusted downward as the use cycle of the hanger increases.
[0025] A further improvement is that the early warning module is connected to the alarm device of the hanger monitoring system and is used to form early warning data into an alarm control and send it to the alarm device for real-time early warning.
[0026] The beneficial effects of the present invention are as follows: the present invention utilizes a monitoring point analysis module to analyze the crack elastic wave signal based on the quality inspection parameter content and the historical damage hanger to find out the location where the hanger is prone to cracks, and to mark the key stress-bearing parts on the hanger, and then arranges an acoustic emission sensor array for signal collection, which can effectively improve the accuracy of acoustic emission signal collection. The improved quality of acoustic emission signal collection is conducive to the intelligent diagnosis module to perform crack classification, severity assessment and crack development trend prediction, which can improve the safety of the hanger and avoid production accidents. BRIEF DESCRIPTION OF THE DRAWINGS
[0027] Figure 1 Schematic diagram of the system framework of the present invention. DETAILED DESCRIPTION
[0028] In order to deepen the understanding of the present invention, the present invention will be further described in detail below with reference to the examples. The examples are only used to explain the present invention and do not constitute a limitation on the scope of protection of the present invention.
[0029] Example 1
[0030] according to Figure 1 As shown, this embodiment proposes an intelligent diagnosis system for hanger cracks based on acoustic emission, which includes a historical hanger crack collection module, a crack database, a hanger quality inspection report library, a monitoring point analysis module, an acoustic emission sensor array, a signal collection and processing module, a priori diagnostic knowledge base, an intelligent diagnosis module, and an early warning module; the historical hanger crack collection module is used to use the acoustic emission sensor to collect crack elastic wave signals of historically damaged hangers;
[0031] The crack database is used to store the crack elastic wave signals of the historical damaged hanger;
[0032] The hanger quality inspection knowledge base is used to store the quality inspection parameter content of the hanger before application;
[0033] The monitoring point analysis module is used to analyze the crack elastic wave signal based on the quality inspection parameter content and the history of damaged hangers to find the location of the hanger prone to cracks, and to mark the key stress-bearing parts on the hanger;
[0034] The acoustic emission sensor array is composed of a plurality of acoustic emission sensors, and the plurality of acoustic emission sensors are distributed and installed at locations prone to cracks and key stress-bearing locations on the hanger.
[0035] The signal acquisition and processing module is used to convert the real-time crack elastic wave signal collected by the acoustic emission sensor into an electrical signal, amplify it through a signal amplifier, and then perform signal noise reduction and feature extraction processing to obtain an acoustic emission signal;
[0036] The priori diagnosis knowledge base is used to store historical crack diagnosis experience of the hanger;
[0037] The intelligent diagnosis module builds a crack diagnosis model based on convolutional neural network (CNN) and recurrent neural network (RNN). After the crack diagnosis model is trained, the characteristic parameters of the acoustic emission signal obtained by the signal acquisition and processing module are input into the crack diagnosis model to perform crack classification, severity assessment and crack development trend prediction;
[0038] The early warning module is used to set a threshold, then compare the intelligent diagnosis with the threshold to obtain early warning data, and perform automatic early warning based on the early warning data.
[0039] The present invention utilizes a monitoring point analysis module to analyze crack elastic wave signals based on quality inspection parameter content and historically damaged hangers to determine locations where cracks are likely to occur on the hanger, and to mark key stress-bearing locations on the hanger. An acoustic emission sensor array is then arranged for signal acquisition, effectively improving the accuracy of acoustic emission signal acquisition. The improved quality of acoustic emission signal acquisition facilitates crack classification, severity assessment, and crack development trend prediction by the intelligent diagnostic module, thereby improving the safety of hanger use and avoiding production accidents.
[0040] Example 2
[0041] according to Figure 1 As shown, this embodiment proposes an intelligent hanger crack diagnosis system based on acoustic emission. The crack database stores elastic wave crack signals from historically damaged hangers, converting these signals into electrical signals to form crack data signals with clear correspondences for storage. The quality inspection parameters include a hanger strength distribution map and a damage distribution map. The monitoring point analysis module analyzes hanger crack-prone locations based on the quality inspection parameters and the elastic wave crack signals from historically damaged hangers. Furthermore, the module identifies key stress-bearing locations on the hanger by cross-comparing and analyzing the hanger strength distribution map, the damage distribution map, and the crack locations corresponding to the elastic wave crack signals from historically damaged hangers. This analysis identifies multiple hanger crack-prone locations. By pre-analyzing the hanger's quality and deploying acoustic emission sensors at locations with low strength or existing damage, the hanger's strength changes can be effectively monitored. Low strength and existing damage locations are prone to cracks, and focusing on these crack-prone locations improves crack diagnosis efficiency. By conducting a statistical analysis of the locations where the hanger is prone to cracks under historical conditions, and then placing acoustic emission sensors at these locations for monitoring, the efficiency of crack diagnosis can also be improved.
[0042] The elastic wave signals of early-stage tiny cracks are relatively weak. If monitoring is not performed at a precise location, it is easy to miss the detection problem. For crack diagnosis, the earlier the diagnosis, the more effective it is. Therefore, the present invention can effectively improve the quality of crack diagnosis through the above-mentioned settings.
[0043] After the multiple acoustic emission sensors of the acoustic emission sensor array are installed, a housing is further installed at the acoustic emission sensor installation location to protect the acoustic emission sensors. The acoustic emission sensors of the present invention are highly sensitive and broadband sensors. The housing prevents the acoustic emission sensors from being damaged in harsh operating environments, thereby affecting signal acquisition accuracy.
[0044] The signal acquisition and processing module performs signal noise reduction processing, specifically by using wavelet transform filtering, adaptive filtering, or a combination of both to reduce noise on the collected signals, addressing various noise interferences present in the actual operating environment of the hanger, thereby improving the signal-to-noise ratio. The signal acquisition and processing module also performs signal feature extraction processing, specifically by extracting various characteristic parameters from the noise-reduced acoustic emission signals, including amplitude, energy, counts, duration, rise time, and frequency components.
[0045] During the training process of the crack diagnosis model, a large number of acoustic emission signal samples of hanger cracks of varying types and severity are collected, labeled, and processed. The collected data is then divided into training, validation, and test sets according to a specific ratio. The training set is used to train the intelligent diagnosis model, the validation set is used to adjust the model's hyperparameters and evaluate the model's performance during training, and the test set is used to evaluate the model's generalization ability. When setting the threshold, the early warning module analyzes historical hanger crack diagnosis experience from a priori diagnostic knowledge base to determine a threshold value. This threshold value is adjusted downward as the hanger's service life increases. The early warning module is connected to the alarm device of the hanger monitoring system and is used to generate an alarm control signal from the early warning data and transmit it to the alarm device for real-time warning. The present invention constructs a crack diagnosis model based on a convolutional neural network (CNN) and a recurrent neural network (RNN). This model combines the advantages of CNN in feature extraction with the capabilities of RNN in processing sequential data, enabling it to better capture the temporal and spatial characteristics of acoustic emission signals. The threshold value is adjusted downward as the hanger's service life increases, and dynamic adjustment helps improve the timeliness of early warnings.
[0046] The basic principles, main features, and advantages of the present invention are shown and described above. Those skilled in the art should understand that the present invention is not limited to the foregoing embodiments. The foregoing embodiments and descriptions are merely illustrative of the principles of the present invention. Various changes and modifications may be made to the present invention without departing from the spirit and scope of the present invention. Such changes and modifications are intended to fall within the scope of the present invention. The scope of protection claimed in the present invention is defined by the appended claims and their equivalents.
Claims
1. An intelligent diagnostic system for hanger cracks based on acoustic emission, characterized by: It includes historical hanger crack collection module, crack database, hanger quality inspection report library, monitoring point analysis module, acoustic emission sensor array, signal acquisition and processing module, prior diagnosis knowledge base, intelligent diagnosis module and early warning module; The historical hanger crack acquisition module is used to collect crack elastic wave signals of historically damaged hangers using an acoustic emission sensor; The crack database is used to store the crack elastic wave signals of the historical damaged hanger; The hanger quality inspection knowledge base is used to store the quality inspection parameter content of the hanger before application; The monitoring point analysis module is used to analyze the crack elastic wave signal based on the quality inspection parameter content and the history of damaged hangers to find the location of the hanger prone to cracks, and to mark the key stress-bearing parts on the hanger; The acoustic emission sensor array is composed of a plurality of acoustic emission sensors, and the plurality of acoustic emission sensors are distributed and installed at locations prone to cracks and key stress-bearing locations on the hanger. The signal acquisition and processing module is used to convert the real-time crack elastic wave signal collected by the acoustic emission sensor into an electrical signal, amplify it through a signal amplifier, and then perform signal noise reduction and feature extraction processing to obtain an acoustic emission signal; The priori diagnosis knowledge base is used to store historical crack diagnosis experience of the hanger; The intelligent diagnosis module builds a crack diagnosis model based on convolutional neural network (CNN) and recurrent neural network (RNN). After the crack diagnosis model is trained, the characteristic parameters of the acoustic emission signal obtained by the signal acquisition and processing module are input into the crack diagnosis model to perform crack classification, severity assessment and crack development trend prediction; The early warning module is used to set a threshold, then compare the intelligent diagnosis with the threshold to obtain early warning data, and perform automatic early warning based on the early warning data.
2. The intelligent diagnostic system for hanger cracks based on acoustic emission according to claim 1, characterized in that: When the crack database stores the elastic wave signals of cracks of the historically damaged hangers, the elastic wave signals of cracks of the historically damaged hangers are converted into electrical signals to form crack data signals with clear corresponding relationships for storage.
3. The intelligent diagnostic system for hanger cracks based on acoustic emission according to claim 2, characterized in that: The quality inspection parameter content includes a hanger strength distribution diagram and a damage distribution diagram.
4. The intelligent diagnostic system for hanger cracks based on acoustic emission according to claim 3, characterized in that: The monitoring point analysis module analyzes the locations of the hangers that are prone to cracks based on the quality inspection parameters and the crack elastic wave signals of historically damaged hangers, and is used to mark the key stress-bearing parts on the hangers. Specifically, it cross-compares and analyzes the hanger strength distribution map, the damage distribution map, and the crack locations corresponding to the crack elastic wave signals of historically damaged hangers to analyze and determine the locations of multiple hangers that are prone to cracks.
5. The intelligent diagnostic system for hanger cracks based on acoustic emission according to claim 1, characterized in that: After the multiple acoustic emission sensors of the acoustic emission sensor array are installed, a shell is further installed at the installation position of the acoustic emission sensor to protect the acoustic emission sensor.
6. The intelligent diagnostic system for hanger cracks based on acoustic emission according to claim 1, characterized in that: When the signal acquisition and processing module performs signal noise reduction processing, specifically: in response to various noise interferences existing in the actual working environment of the hanger, one or a combination of wavelet transform filtering and adaptive filtering is used to reduce the noise of the collected signal to improve the signal-to-noise ratio.
7. The intelligent diagnostic system for hanger cracks based on acoustic emission according to claim 6, characterized in that: The signal acquisition and processing module extracts and processes the signal features by extracting a variety of characteristic parameters from the noise-reduced acoustic emission signal, including amplitude, energy, count, duration, rise time, and frequency components.
8. The intelligent diagnostic system for hanger cracks based on acoustic emission according to claim 1, characterized in that: During the training process of the crack diagnosis model, a large number of acoustic emission signal samples of hanger cracks of different types and severity are collected, labeled and processed, and then the collected data is divided into a training set, a validation set and a test set according to a certain ratio. The training set is used to train the intelligent diagnosis model, the validation set is used to adjust the model's hyperparameters and evaluate the model's performance during the training process, and the test set is used to evaluate the model's generalization ability.
9. The intelligent diagnostic system for hanger cracks based on acoustic emission according to claim 1, characterized in that: When setting the threshold, the warning module analyzes the historical crack diagnosis experience of the hanger in the prior diagnosis knowledge base to comprehensively analyze the threshold setting value, and the threshold setting value is numerically adjusted downward as the service life of the hanger increases.
10. The intelligent diagnostic system for hanger cracks based on acoustic emission according to claim 9, characterized in that: The early warning module is connected to the alarm device of the hanger monitoring system and is used to form early warning data into an alarm control and send it to the alarm device for real-time early warning.