Pressure swing adsorption device monitoring method, system, equipment and medium
Through digital twin technology and neural network model, a digital twin model of voltage-switch adsorption device is built, and the device status is monitored and evaluated in real time, which solves the problems of slow response speed and low prediction accuracy in traditional monitoring methods, and improves fault detection and maintenance efficiency. Enhanced safety and reliability of the equipment.
Patent Information
- Application Number
- CN202411108283.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-13
- Publication Date
- 2024-11-26
AI Technical Summary
The monitoring methods of traditional pressure-switch adsorption devices rely on manual inspection and regular maintenance, with slow response speed and low prediction accuracy, resulting in low equipment fault detection and maintenance efficiency, posing safety hazards and economic losses.
Using digital twin technology to build a digital twin model of the voltage-transforming adsorption device, combined with neural network model to perform defect prediction and evaluation, obtain three-dimensional spatial data sets through various non-destructive detection methods, monitor the operating status of the device in real time, and evaluate three-dimensionally in the digital twin model result.
Accurate monitoring and real-time evaluation of the pressure-switch adsorption device is realized, fault prediction accuracy and response speed are improved, maintenance costs are reduced, and equipment safety and reliability are enhanced.
Smart Images

Figure CN119026049A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of coal chemical industry, and in particular to a monitoring method, system, equipment and medium for a pressure swing adsorption device. Background Art
[0002] Pressure swing adsorption device is an important equipment widely used in chemical, petroleum, natural gas and other fields. The stability of its adsorption performance and health have an important impact on the stability and safety of the production process. At present, the main failure modes of domestic pressure swing adsorption devices are corrosion under the insulation layer and fatigue cracks caused by mechanical fatigue. Since the pressure swing adsorption device is located upstream of the production process, if a certain device fails and causes gas leakage, it is easy to cause a special equipment safety accident when encountering an open flame. The consequences and losses are immeasurable. At the same time, it will cause insufficient supply of upstream raw gas, which will lead to a reduction in production or shutdown of the entire production chain. The indirect economic losses are also more than 30 million, and the direct economic losses range from 4 to 10 million (mainly in the replacement of catalysts and equipment maintenance). It is not conducive to social security and the safety of people’s lives and property, and it has increased the direct costs of enterprises invisibly. The monitoring methods of traditional pressure swing adsorption devices often rely on manual inspections and regular maintenance, and there are problems such as slow response speed and low prediction accuracy.
[0003] Digital Twins (DT) was first proposed by Professor Michael-Grieves in the Product Life Cycle Management (PLM) course at the University of Michigan in 2002. In 2016, ANSYS and General Electric entered into a technical cooperation to expand and integrate the engineering simulation and embedded software development platforms of both parties, providing power for a wide range of industry digital twin solutions. The development of digital twin technology in China started late. The background and connotation of digital twin technology were analyzed, the concept of digital twin technology and the architecture of digital twin technology products were proposed, the application of digital twin technology in product design, production and maintenance were studied, and the development trend of digital twin technology in products in the future was analyzed. Zhao Haoran et al. proposed a three-dimensional real-time monitoring method for production workshops based on digital twin technology to realize visual monitoring of industrial robots. However, due to the complex production mechanism and equipment mechanism of the petrochemical industry, the application of digital twin technology in the petrochemical industry is still relatively small. Most of the research is still in the theoretical stage and lacks practical application, which limits the development of digital twin technology in the petrochemical industry.
[0004] In summary, there is an urgent need to provide an application solution for digital twins in coal chemical production processes to improve the above problems. Summary of the invention
[0005] The purpose of the present invention is to provide a pressure swing adsorption device monitoring method, system, equipment and medium to improve the traditional defect detection scheme that relies too much on manual inspection and regular maintenance, and has problems such as slow response speed and low prediction accuracy.
[0006] In a first aspect, an embodiment of the present invention provides a defect detection method for a pressure swing adsorption device, comprising:
[0007] Based on the three-dimensional model of the pressure swing adsorption device, the mechanism model of the pressure swing adsorption device and the operating parameters of the pressure swing adsorption device, a digital twin model of the pressure swing adsorption device is constructed;
[0008] Based on the formation mechanism, conditions and detection results of various indicators of local defects of the pressure swing adsorption device under the action of positive and negative pressure cycles, combined with the digital twin model, a defect prediction and evaluation system of the pressure swing adsorption device is constructed;
[0009] Based on the defect prediction and evaluation system of the pressure swing adsorption device, a neural network model is trained to obtain a defect prediction and evaluation model of the pressure swing adsorption device;
[0010] Based on the defect prediction and evaluation model of the pressure swing adsorption device, the operating status of the pressure swing adsorption device is evaluated in real time, and the evaluation results are displayed in three dimensions in the digital twin model.
[0011] In a possible implementation, the operating state of the pressure swing adsorption device is evaluated in real time, including:
[0012] Using a PLC robot sensing and control system, the defect prediction and evaluation system communicates with the pressure swing adsorption device through a communication protocol to obtain real-time operating data and real-time nondestructive testing results of the pressure swing adsorption device;
[0013] The real-time operation data and the real-time nondestructive testing results of the pressure swing adsorption device are input into the defect prediction and evaluation model to perform a real-time evaluation on the operation status of the pressure swing adsorption device.
[0014] In another possible implementation, the sources of the test results of the various indicators include:
[0015] By using non-destructive testing methods such as wall thickness measurement, magnetic particle testing, and ultrasonic testing, a data set of crack defects, stress defects, and bulge defects in three dimensions of the pressure swing adsorption device during service is obtained;
[0016] Based on the data set, the wall thickness reduction, stress concentration, crack distribution, and location distribution of local temperature rise defects during the service process are determined, and a service equipment failure distribution database including the detection results of various indicators is established.
[0017] In other possible implementations, the digital twin model of the pressure swing adsorption device is constructed based on the three-dimensional model of the pressure swing adsorption device, the mechanism model of the pressure swing adsorption device and the operating parameters of the pressure swing adsorption device, including:
[0018] The calculation process and logic of the mechanism model are embedded into the corresponding components or objects of the 3D model, and the mechanism model is called for real-time calculation and simulation during the operation of the 3D model;
[0019] The calculation results of the mechanism model are mapped to the corresponding positions of the three-dimensional model and compared with the operating status data of the pressure swing adsorption device during operation, so as to couple the multi-dimensional data and laws of multi-physical fields, realize the real-time symbiosis of the physical entity and virtual entity of the pressure swing adsorption device system, and obtain the digital twin model.
[0020] In a possible implementation, based on the defect prediction and evaluation system of the pressure swing adsorption device, a neural network model training is performed to obtain a defect prediction and evaluation model of the pressure swing adsorption device, including:
[0021] Cleaning and preprocessing the operating parameters and test results of various indicators of the pressure swing adsorption device, removing noise and outliers, marking defective data, and obtaining a training data set;
[0022] The training data set is used to train the neural network, and the network weights and biases are continuously adjusted through the back propagation algorithm to minimize the prediction error. Regularization is used to prevent the model from overfitting, so as to obtain the defect prediction and evaluation model of the pressure swing adsorption device.
[0023] In a possible implementation, the real-time operation data and the real-time nondestructive testing results of the pressure swing adsorption device are input into the defect prediction and evaluation model to perform a real-time evaluation on the operation status of the pressure swing adsorption device, including:
[0024] Collecting real-time operating parameters and real-time nondestructive testing results of the pressure swing adsorption device;
[0025] Calculating time domain features and frequency domain features corresponding to the real-time operating parameters and the real-time nondestructive testing results to form a plurality of initial sample vectors;
[0026] Using principal component analysis to reduce the dimension of the initial sample vector to obtain main characteristic components, and obtaining target sample vectors corresponding to the main characteristic components to form a defect analysis sample library;
[0027] The target sample vector in the defect analysis sample library is input into the defect prediction and evaluation model; the defect prediction and evaluation model is used to perform real-time evaluation to obtain a defect evaluation result.
[0028] In a possible implementation, the initial sample vector is reduced in dimension by using the principal component analysis method to obtain the main characteristic components, and the target sample vector corresponding to the main characteristic components is obtained to form a defect analysis sample library, including:
[0029] splicing the initial sample vectors into a sample feature matrix, and standardizing the sample feature matrix;
[0030] The principal component analysis method is used to analyze the standardized sample feature matrix to obtain eigenvalues, and the main characteristic components are selected from multiple eigenvalues. The corresponding eigenvectors are matched as target sample vectors to form a defect analysis sample library.
[0031] In a second aspect, an embodiment of the present invention further provides a defect detection device for a pressure swing adsorption device, comprising: a physical layer, a data layer, a model layer, and an application layer, wherein:
[0032] The physical layer includes the pressure swing adsorption device, the data acquisition terminal, the multi-data integrated sensor system, and the communication interface;
[0033] The data layer is used to collect the operating data of the pressure swing adsorption device and the test results of various indicators;
[0034] The model layer is used to construct a digital twin model of the pressure swing adsorption device based on the three-dimensional model of the pressure swing adsorption device, the mechanism model of the pressure swing adsorption device and the operating parameters of the pressure swing adsorption device; based on the formation mechanism, conditions and detection results of various indicators of local defects of the pressure swing adsorption device under the action of positive and negative pressure cycles, in combination with the digital twin model, a defect prediction and evaluation system of the pressure swing adsorption device is constructed; based on the defect prediction and evaluation system of the pressure swing adsorption device, a neural network model is trained to obtain a defect prediction and evaluation model of the pressure swing adsorption device; based on the defect prediction and evaluation model of the pressure swing adsorption device, the operating status of the pressure swing adsorption device is evaluated in real time;
[0035] The application layer is used to provide a user interface to display the evaluation results in three dimensions in the digital twin model.
[0036] In a third aspect, an embodiment of the present invention further provides an electronic device, comprising a memory and a processor, wherein the memory stores a program that can be run on the processor, and when the program is executed by the processor, the electronic device implements a method in any possible embodiment of the first aspect.
[0037] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium includes a program, and when the program runs on an electronic device, the electronic device executes a method in any possible implementation of the first aspect.
[0038] In a fifth aspect, an embodiment of the present invention further provides a computer program product, which, when executed on an electronic device, enables the electronic device to execute a method of any possible implementation of the first aspect.
[0039] The present invention provides a method and system for monitoring a pressure swing adsorption device. The beneficial effects of the technical solution are mainly reflected in the following aspects:
[0040] 1. Accurate monitoring and real-time evaluation: By integrating various non-destructive testing methods such as wall thickness measurement, magnetic particle testing, and ultrasonic testing, a detailed three-dimensional spatial data set of the pressure swing adsorption device during service is obtained, making the monitoring of crack defects, stress defects, bulge defects and other problems more accurate. Combined with the digital twin model, real-time evaluation of the device's operating status is achieved, and potential safety hazards can be discovered in a timely manner.
[0041] 2. Efficient data processing and modeling: The present invention uses a neural network model to perform defect prediction and evaluation. Through data cleaning, preprocessing, labeling and other steps, a high-quality training data set is constructed to improve the accuracy and generalization ability of the model. At the same time, the principal component analysis method is used to reduce the dimension of the initial sample vector and extract the main characteristic components, which reduces the amount of calculation and improves the evaluation efficiency.
[0042] 3. Multi-dimensional coupling and real-time symbiosis: The calculation process and logic of the mechanism model are embedded into the corresponding components or objects of the three-dimensional model. Through real-time calculation and simulation, the coupling of multi-dimensional data and laws of multi-physical fields is realized. This real-time symbiosis enables the digital twin model to more accurately reflect the actual operating status of the pressure swing adsorption device.
[0043] 4. Broad application prospects: The present invention is not only applicable to the monitoring and evaluation of pressure swing adsorption devices, but can also be extended to the monitoring and evaluation of other similar equipment. By building a digital twin model, comprehensive monitoring and predictive maintenance of the equipment can be achieved, the safety and reliability of the equipment can be improved, and the maintenance cost can be reduced.
[0044] 5. Improve decision-making efficiency: By evaluating the operating status of the pressure swing adsorption device in real time and displaying the evaluation results in three dimensions, decision-makers can intuitively understand the operating status of the equipment and make correct decisions quickly. This helps improve production efficiency and safety and reduce losses caused by equipment failure.
[0045] In summary, the pressure swing adsorption device monitoring method and system provided by the present invention have the advantages of accurate monitoring, efficient processing, multi-dimensional coupling, comprehensive architecture and broad application prospects, which are of great significance for improving the safety and reliability of the equipment. BRIEF DESCRIPTION OF THE DRAWINGS
[0046] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0047] Figure 1 A schematic diagram of the architecture of a pressure swing adsorption device monitoring system provided by an embodiment of the present invention;
[0048] Figure 2 A schematic flow chart of a method for monitoring a pressure swing adsorption device provided by an embodiment of the present invention;
[0049] Figure 3 A schematic flow chart of a method for generating a defect prediction, evaluation and monitoring model provided by an embodiment of the present invention;
[0050] Figure 4 A schematic diagram of a defect detection device for a pressure swing adsorption device provided by an embodiment of the present invention;
[0051] Figure 5 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION
[0052] In the description of the embodiments of the present invention, the terms used in the following embodiments are only for the purpose of describing specific embodiments, and are not intended to be used as limitations of the present invention. As used in the specification of the present invention and the appended claims, the singular expressions "a", "said", "above", "the" and "this" are intended to also include expressions such as "one or more", unless there is a clear contrary indication in the context. It should also be understood that in the following embodiments of the present invention, "at least one", "one or more" refer to one or more (including two). The term "and / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist; for example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone, where A and B can be singular or plural. The character " / " generally indicates that the objects associated before and after are in an "or" relationship.
[0053] References to "one embodiment" or "some embodiments" etc. described in this specification mean that one or more embodiments of the present invention include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the statements "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways. The term "connection" includes direct connection and indirect connection, unless otherwise specified. "First" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated.
[0054] In the embodiments of the present invention, the words "exemplarily" or "for example" are used to indicate examples, illustrations or explanations. Any embodiment or implementation scheme described as "exemplarily" or "for example" in the embodiments of the present invention should not be interpreted as being more preferred or more advantageous than other embodiments or implementation schemes. Specifically, the use of words such as "exemplarily" or "for example" is intended to present related concepts in a specific way.
[0055] The present invention proposes a pressure swing adsorption device monitoring method and system, which aims to achieve real-time monitoring, fault warning and intelligent optimization of the operating status of the pressure swing adsorption device by building a high-fidelity digital twin model. The method integrates advanced technologies such as the Internet of Things, big data, and artificial intelligence. The system has the advantages of high prediction accuracy, fast response speed, and simple operation, and provides new ideas and technical means for the intelligent management of pressure swing adsorption devices.
[0056] Figure 1It is a schematic diagram of the structure of a pressure swing adsorption device monitoring system provided by an embodiment of the present invention, which mainly includes four parts: physical layer, data layer, model layer, and application layer. Among them: the physical layer is the actual operating physical entity part, specifically including the pressure swing adsorption device, data acquisition terminal, multi-data integrated sensor system, communication interface, etc. The pressure swing adsorption device includes components such as the main structure, adsorbent, control system, import and export pipelines, and safety protection measures. The data acquisition terminal includes a sensor network, a video monitoring network, an infrared monitoring network, etc. The communication interface includes Ethernet RS232, etc., and remote data acquisition is realized through embedded devices such as MCU / Zynq / FPGA and wireless communication modules. The data layer is responsible for collecting the operating data of the pressure swing adsorption device, including parameters such as pressure, flow, temperature, and efficiency; the model layer constructs a digital twin model, a three-dimensional model, an evaluation algorithm, etc. of the pressure swing adsorption device to realize real-time processing and analysis of data. The application layer provides a user interface to display monitoring results and realize fault warning and intelligent optimization.
[0057] This embodiment is based on the working principle and physical characteristics of the pressure swing adsorption device, and uses multi-physics field coupling simulation technology to build a digital twin model that includes processes such as fluid dynamics, heat transfer, and chemical reactions. For example, the digital twin model of the pressure swing adsorption device can be established using solid works software and 3Dmax software, and the digital twin environment can be built through the Unity 3D platform. By establishing a process parameter collaborative control model, Vuforia can be used to achieve observation of the entire operation process of the pressure swing adsorption device.
[0058] During the operation of the pressure swing adsorption device, the communication between the pressure swing adsorption device and the defect prediction and evaluation system can be achieved by applying the sensing and control system of the programmable logic controller (PLC) and relying on the development of the Tongxin protocol. This process involves effectively connecting the physical equipment with the digital model to ensure seamless data exchange and command transmission between the two. In this process, the data layer can monitor and respond to changes in the external environment in real time through its highly integrated sensors and actuators. At the same time, the model layer is responsible for processing and analyzing the collected data to generate corresponding control strategies. The development of the Tongxin protocol has become a key bridge connecting the two layers. It not only ensures the stability and reliability of data transmission, but also improves the overall response speed and flexibility of the system. The realization of this communication method has greatly improved the intelligence level of the robot, enabling it to complete various complex tasks more efficiently and accurately.
[0059] The data collection end monitors and collects multi-source heterogeneous information in real time, and uses adaptive calibration and cleaning methods. It uses a deep learning model based on data mining and feature fusion to establish a mapping relationship between "information features and defect prediction and evaluation" and use defect prediction and evaluation models to predict and evaluate device status defects. The application layer provides a user interface to display monitoring results, achieve fault warning and intelligent optimization. The evaluation algorithm can be used for state evaluation model training to achieve equipment life prediction, equipment fault diagnosis and system performance analysis, as well as real-time evaluation of operating efficiency and response time. The state evaluation analysis results are then mapped to a three-dimensional model for visual display of the results. Driven by historical data and real-time data, the model parameters are continuously updated to improve the accuracy and real-time performance of the model.
[0060] Based on the above monitoring system, this embodiment provides a pressure swing adsorption device monitoring method based on digital twin, which mainly includes:
[0061] (1) Real-time data collection and preprocessing: Using the Internet of Things technology, the operation data and non-destructive testing data of the pressure swing adsorption device are collected in real time, and preprocessing operations such as cleaning and denoising are performed to ensure data quality.
[0062] (2) Condition monitoring and evaluation: The preprocessed data is input into the digital twin model, a defect prediction and evaluation system of the pressure swing adsorption device is constructed, a neural network model is trained, and a defect prediction and evaluation model of the pressure swing adsorption device is obtained. The defect prediction and evaluation model is used to perform real-time simulation and analysis to evaluate the operating status of the pressure swing adsorption device. The abnormal status is identified by comparing the actual operating data with the model prediction data.
[0063] (3) Fault warning and diagnosis: Based on the model prediction results, combined with the expert knowledge base and machine learning algorithm, early warning and intelligent diagnosis of faults are achieved. Through visualization technology, the fault location, cause and possible impact are displayed.
[0064] (4) Intelligent optimization and decision support: Based on the monitoring results and fault warning information, targeted optimization suggestions are put forward, such as adjusting operating parameters, replacing adsorption materials, etc. At the same time, decision support is provided to managers to improve production efficiency and safety.
[0065] Specifically, if Figure 2 As shown, the pressure swing adsorption device monitoring method comprises the following steps:
[0066] S201, constructing a digital twin model of the pressure swing adsorption device based on the three-dimensional model of the pressure swing adsorption device, the mechanism model of the pressure swing adsorption device, and the operating parameters of the pressure swing adsorption device;
[0067] S202, based on the formation mechanism and conditions of local defects of the pressure swing adsorption device under the action of positive and negative pressure cycles and the nondestructive testing data of the pressure swing adsorption device, combined with the digital twin model, construct a defect prediction and evaluation system for the pressure swing adsorption device;
[0068] S203, performing neural network model training based on the defect prediction and evaluation system of the pressure swing adsorption device to obtain a defect prediction and evaluation model of the pressure swing adsorption device;
[0069] S204: Based on the defect prediction and evaluation model of the pressure swing adsorption device, the operating status of the pressure swing adsorption device is evaluated in real time, and the evaluation result is displayed in three dimensions in the digital twin model.
[0070] To expand, in S201, first, it is necessary to obtain a three-dimensional model of the pressure swing adsorption device, which usually includes detailed information such as geometric structure and material properties. Subsequently, it is necessary to have a deep understanding of the mechanism model of the pressure swing adsorption device, which involves scientific principles such as its working principle, gas flow characteristics, adsorption and desorption process, etc. At the same time, combined with the actual operating parameters of the pressure swing adsorption device, such as pressure, flow, adsorption time, regeneration conditions, gas composition, etc., a highly accurate digital twin model can be constructed. This model will be able to simulate the operation of the pressure swing adsorption device under various working conditions, providing a solid foundation for subsequent defect prediction and evaluation. In S202, based on the formation mechanism and conditions of local defects of the pressure swing adsorption device under the action of positive and negative pressure cycles, it is necessary to deeply analyze how these defects are generated under specific working conditions. At the same time, combined with the test results of various indicators, such as pressure fluctuations, temperature anomalies, etc., the existence and severity of defects can be further confirmed. Then, using the digital twin model constructed in step S201, the impact of defects on the operating performance of the pressure swing adsorption device can be simulated, and a defect prediction and evaluation system can be constructed accordingly. The system will be able to monitor the operating status of the pressure swing adsorption unit in real time, predict possible defects, and assess their potential risks.
[0071] On the basis of the defect prediction and evaluation system constructed in step S202, the neural network model can be used for further training and optimization. By inputting a large amount of historical data and simulation data, the neural network model can learn and identify various abnormal patterns and defect characteristics that occur during the operation of the pressure swing adsorption device. After repeated training and adjustment, a defect prediction and evaluation model with high accuracy and stability can be obtained. This model will be able to more accurately predict the defects of the pressure swing adsorption device and provide strong support for its maintenance and management. Finally, after obtaining the defect prediction and evaluation model of the pressure swing adsorption device, it can be combined with the digital twin model to achieve real-time three-dimensional display of the evaluation results. This means that the operating status and potential defects of the pressure swing adsorption device can be intuitively seen in the digital twin model. Through three-dimensional display technology, the location, size, nature and other information of the defects can be more clearly understood, providing a more accurate decision-making basis for subsequent maintenance and management. At the same time, this real-time display method also helps to improve the monitoring capability of the operating status of the pressure swing adsorption device and ensure that it operates in a safe and efficient state.
[0072] In a possible embodiment, constructing the defect prediction and evaluation system of the pressure swing adsorption device requires predetermining the interaction mechanism between defects and processes. Considering that the pressure swing adsorption device is a technology that uses pressure changes to adsorb and separate gases, the process of the pressure swing adsorption device refers to the material transfer and reaction process inside the device, including adsorption, desorption, regeneration and other processes. Research has found that the interaction mechanism between defects and processes mainly includes the following aspects:
[0073] ①Interaction between the service conditions and defects of the pressure swing adsorption device
[0074] This embodiment can comprehensively obtain defects such as cracks, stresses, bulges, etc. during the service of the device through non-destructive testing methods such as wall thickness measurement, magnetic particle testing, and ultrasonic testing, obtain comprehensive data such as the formation type, quantity, and distribution location of various defects in the operating equipment, and establish a database of the operating parameters of the service equipment (such as pressure, flow, adsorption time, regeneration conditions, gas composition, etc.) and its failure behavior.
[0075] Wall thickness detection is a method of measuring the wall thickness of the equipment to determine whether the equipment has defects such as thinning and cracks. Magnetic particle testing is a method that uses magnetic fields and magnetic particles to detect surface and near-surface cracks of equipment. Ultrasonic testing is a method that uses the propagation characteristics of ultrasonic waves in materials to detect internal defects of materials. These three non-destructive testing methods have their own characteristics and complement each other, and can comprehensively cover the types of defects that may exist in the equipment. Through these non-destructive testing methods, a three-dimensional spatial data set of equipment defects can be obtained, including information such as the location, size, shape and depth of the defects. These data are of great significance for understanding the operating status of the equipment, evaluating the reliability of the equipment, and formulating equipment maintenance and repair plans. In general, the use of non-destructive testing methods such as wall thickness detection, magnetic particle testing and ultrasonic testing can comprehensively and accurately collect three-dimensional spatial data sets of defects such as cracks, stress and bulges that occur in the equipment during operation, providing a scientific and reliable basis for the maintenance and repair of the equipment.
[0076] ② In-depth correlation mechanism between the operating parameters of the pressure swing adsorption device and its failure behavior
[0077] In this case, a strategy of meticulous sampling of the damaged area of the equipment was adopted. Through a series of advanced microstructure characterization and analysis techniques, such as optical microscopy (OM), scanning electron microscopy (SEM) and X-ray diffraction analysis (XRD), the complex process of the germination, growth and evolution of the main defects of the pressure swing adsorption device during service was comprehensively analyzed. This process not only reveals the close relationship between service conditions and the formation of macroscopic defects of the equipment, but also provides an effective method for accurately evaluating the defect level of the equipment in operation. Specifically, through a detailed analysis of the damaged parts of the equipment, key microstructure characteristic parameters including grain size, phase type, elemental composition and strain degree were obtained. The acquisition of these parameters provides valuable data support for a deep understanding of the nucleation, growth, concentrated distribution location and the entire evolution process of defects in the adsorption device during service. Furthermore, combining the results of non-destructive testing and the microstructure characteristics of the materials used in the device, a comprehensive correlation mechanism covering "production process-defect characteristics-equipment status" was creatively constructed. The establishment of this mechanism not only provides a scientific basis for predicting and preventing equipment failure, but also lays a solid foundation for achieving safe and long-term operation of the equipment and optimizing the design of the equipment structure. In conclusion, the research results are of great significance for improving the overall performance of the pressure swing adsorption device and extending its service life.
[0078] Based on the interaction mechanism between defects and processes mentioned above, Figure 3 The exemplary embodiment shows a flow chart of a method for generating a defect prediction and evaluation model. Figure 3 As shown, the process includes:
[0079] Step 301, constructing a digital twin model of the pressure swing adsorption device.
[0080] This step is to build a digital twin model of the pressure swing adsorption device, which is a complex but crucial process. This involves data integration and technical applications in the following aspects. First, build a three-dimensional model: 1. Geometric structure modeling: Using CAD (computer-aided design) software, accurately build a three-dimensional model of its geometric structure according to the actual size and shape of the pressure swing adsorption device, including all key components such as adsorption towers, pipes, valves, sensors, etc. 2. Material property definition: Define the material properties of each component in the model, such as density, strength, thermal conductivity, elastic modulus, etc. These properties will affect the simulation of the physical behavior of the model. Secondly, integrate the mechanism model: 1. Working principle simulation: Based on the working principle of the pressure swing adsorption device, scientific principles such as gas flow, adsorption and desorption processes are integrated into the model. This usually requires the use of CFD (computational fluid dynamics) software to simulate the flow and distribution of gas in the device. 2. Thermodynamic and kinetic models: Build a thermodynamic model to simulate the behavior of the device under temperature changes, and a kinetic model to simulate the speed and efficiency of the adsorption and desorption processes. Finally, integrate the operating parameters: 1. Real-time data access: Through the Industrial Internet of Things (IOT) technology, the actual operating parameters of the pressure swing adsorption device (such as pressure, flow, temperature, etc.) are connected to the digital twin model in real time. 2. Historical data integration: At the same time, historical operating data is also integrated into the model for long-term trend analysis and fault prediction.
[0081] S302, generating a defect prediction and evaluation model.
[0082] In this step, first analyze the fatigue damage and crack propagation that may occur in the material under the action of positive and negative pressure cycles. Identify possible stress concentration areas in the device, which are more prone to defects. Use non-destructive testing technology (such as wall thickness testing, magnetic particle testing, ultrasonic testing, etc.) to obtain defect information in the device, and combine non-destructive testing data with the digital twin model to achieve accurate positioning and quantification of defects. Use machine learning algorithms (such as support vector machines, random forests, neural networks, etc.) to train defect data, build a defect prediction and evaluation model, and verify the accuracy and reliability of the model by comparing the prediction results with the actual test results.
[0083] In this embodiment, the neural network model training process includes: the first step is to prepare the data: 1. Data cleaning: clean and preprocess the raw data to remove noise and outliers. 2. Data labeling: label the defect data so that the neural network can learn how to identify different types of defects. The second step is to design the model: 1. Network architecture: select a suitable neural network architecture (such as convolutional neural network CNN, recurrent neural network RNN, etc.) to adapt to the defect prediction task. 2. Hyperparameter adjustment: adjust the model's hyperparameters (such as learning rate, batch size, number of iterations, etc.) through methods such as cross-validation to optimize model performance. The third step is to train the model: the training process is to train the neural network using the prepared data set, continuously adjust the network weights and biases through the back-propagation algorithm to minimize the prediction error, and use regularization, dropout and other techniques to prevent the model from overfitting.
[0084] It should be understood that after the above-mentioned defect prediction and evaluation model is generated, the actual operation data of the pressure swing adsorption device is connected to the digital twin model and the defect prediction and evaluation model in real time. The defect prediction and evaluation model is used to process and analyze the real-time data to evaluate the operating status and potential defect risks of the device. Moreover, the operating status and defect location of the device can be intuitively displayed in the digital twin model through color, transparency, shape, etc., and an interactive interface can also be provided for users to view and analyze the evaluation results.
[0085] In an embodiment of the present invention, a defect prediction and evaluation model based on a pressure swing adsorption device is constructed, and the model can evaluate the operating state of the pressure swing adsorption device in real time. Specifically, the process includes the following steps: First, it is necessary to collect the real-time operating parameters and real-time non-destructive testing results of the pressure swing adsorption device. These parameters and results are the basis for evaluating the operating state of the pressure swing adsorption device. Next, it is necessary to calculate the time domain characteristics and frequency domain characteristics corresponding to these real-time operating parameters and real-time non-destructive testing results. These characteristics will help to better understand the operating state of the pressure swing adsorption device. Through these characteristics, multiple initial sample vectors can be constructed. Then, the principal component analysis method is used to reduce the dimension of these initial sample vectors to obtain the main characteristic components. These main characteristic components can help capture the key information of the operating state of the pressure swing adsorption device. The target sample vectors corresponding to these main characteristic components are combined to form a defect analysis sample library. Next, the target sample vectors in the defect analysis sample library are input into the defect prediction and evaluation model. This model will use these sample vectors for training, so as to learn how to evaluate the operating state of the pressure swing adsorption device. Finally, the defect prediction and evaluation model is used to perform real-time evaluation to obtain the defect evaluation results of the pressure swing adsorption device. This result can help to timely discover possible problems with the pressure swing adsorption device, so that appropriate measures can be taken to repair or maintain it to ensure the normal operation of the pressure swing adsorption device.
[0086] Specifically, the process of using principal component analysis to reduce the dimension of the initial sample vector to obtain the main characteristic components is as follows: First, the initial sample vector is spliced into a sample feature matrix, and the matrix is standardized to make it more processable. Then, the standardized sample feature matrix is analyzed using principal component analysis to obtain eigenvalues. These eigenvalues represent the main change direction of the sample feature matrix. From these eigenvalues, the main characteristic components are selected and the corresponding eigenvectors are matched as target sample vectors. These target sample vectors will constitute the defect analysis sample library. In this way, the operating status of the pressure swing adsorption device can be effectively evaluated in real time to ensure its normal operation.
[0087] In some embodiments of the present invention, the present invention also discloses a defect detection device for a pressure swing adsorption device, such as Figure 4 As shown, the device is used to implement the methods described in the above method embodiments, which includes: a model building unit 401, a model training unit 402 and a monitoring unit 403, wherein:
[0088] The model building unit 401 is used to build a digital twin model of the pressure swing adsorption device based on the three-dimensional model of the pressure swing adsorption device, the mechanism model of the pressure swing adsorption device and the operating parameters of the pressure swing adsorption device; based on the formation mechanism, conditions and detection results of various indicators of local defects of the pressure swing adsorption device under the action of positive and negative pressure cycles, in combination with the digital twin model, build a defect prediction and evaluation system for the pressure swing adsorption device;
[0089] A model training unit 402 is used to perform neural network model training based on the defect prediction and evaluation system of the pressure swing adsorption device to obtain a defect prediction and evaluation model of the pressure swing adsorption device;
[0090] The monitoring unit 403 is used to evaluate the operating status of the pressure swing adsorption device in real time based on the defect prediction and evaluation model of the pressure swing adsorption device, and display the evaluation result in three dimensions in the digital twin model.
[0091] All relevant contents of each step involved in the above method embodiment can be referred to the functional description of the corresponding functional module, and will not be repeated here.
[0092] The above device may be executed by a chip or a chip module. The various modules / units included in the various devices and products described in the above embodiments may be software modules / units or hardware modules / units, or may be partially software modules / units and partially hardware modules / units.
[0093] Next, an electronic device provided by an embodiment of the present application is introduced. Figure 5 , Figure 5 A schematic diagram of a structure of an electronic device provided in an embodiment of the present application, wherein the electronic device 500 implements Figure 1 Functions of the corresponding method embodiments. Specifically, the electronic device 500 includes: a receiver 501, a transmitter 502, a processor 503 and a memory 504 (wherein the number of processors 503 in the electronic device 500 can be one or more), wherein the processor 503 may include an application processor 5031 and a communication processor 5032. In some embodiments of the present application, the receiver 501, the transmitter 502, the processor 503 and the memory 504 may be connected via a bus or other means.
[0094] The memory 504 may include a read-only memory and a random access memory, and provides instructions and data to the processor 503. A portion of the memory 504 may also include a non-volatile random access memory (NVRAM). The memory 504 stores processor and operation instructions, executable modules or data structures, or subsets thereof, or extended sets thereof, wherein the operation instructions may include various operation instructions for implementing various operations.
[0095] The processor 503 controls the operation of the electronic device. In a specific application, the various components of the electronic device are coupled together through a bus system, wherein the bus system includes not only a data bus but also a power bus, a control bus, and a status signal bus, etc. However, for the sake of clarity, various buses are referred to as bus systems in the figure.
[0096] The method disclosed in the above embodiment of the present application can be applied to the processor 503, or implemented by the processor 503. The processor 503 can be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by an integrated logic circuit of hardware in the processor 503 or an instruction in the form of software. The above processor 503 can be a general-purpose processor, a digital signal processor (digital signal processing, DSP), a microprocessor or a microcontroller, and a vision processor (vision processing unit, VPU), a tensor processor (tensorprocessing unit, TPU) and other processors suitable for AI computing, and can further include an application specific integrated circuit (application specific integrated circuit, ASIC), a field programmable gate array (field-programmable gate array, FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The processor 503 can implement or execute the disclosed methods, steps and logic block diagrams in the embodiments of the present application. The general processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in the embodiment of the present application can be directly embodied as being executed by a hardware decoding processor, or being executed by a combination of hardware and software modules in the decoding processor. The software module can be located in a storage medium mature in the art such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory 504, and the processor 503 reads the information in the memory 504, and completes the steps in the above method embodiment in combination with its hardware.
[0097] The receiver 501 can be used to receive input digital or character information and generate signal input related to the relevant settings and function control of the electronic device. The transmitter 502 can be used to output digital or character information through the first interface; the transmitter 502 can also be used to send instructions to the disk group through the first interface to modify the data in the disk group; the transmitter 502 can also include a display device such as a display screen.
[0098] The embodiment of the present application also provides a computer program product, which, when executed on a computer, enables the computer to execute the steps executed by the aforementioned device, or enables the computer to execute the steps executed by the aforementioned device.
[0099] A computer-readable storage medium is also provided in an embodiment of the present application. The computer-readable storage medium stores a program for performing signal processing. When the program is run on a computer, the computer executes the steps executed by the aforementioned device, or the computer executes the steps executed by the aforementioned device.
[0100] The execution device, training device or electronic device provided in the embodiments of the present application may specifically be a chip, and the chip includes: a processing unit and a communication unit, wherein the processing unit may be, for example, a processor, and the communication unit may be, for example, an input / output interface, a pin or a circuit, etc. The processing unit may execute the computer execution instructions stored in the storage unit so that the chip in the execution device executes the data processing method described in the above embodiment, or so that the chip in the training device executes the data processing method described in the above embodiment. Optionally, the storage unit is a storage unit in the chip, such as a register, a cache, etc. The storage unit may also be a storage unit located outside the chip in the wireless access device, such as a read-only memory (ROM) or other types of static storage devices that can store static information and instructions, a random access memory (RAM), etc.
[0101] It should also be noted that the device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed over multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, in the drawings of the device embodiments provided by the present application, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines.
[0102] Through the description of the above implementation mode, the technicians in the field can clearly understand that the present application can be implemented by means of software plus necessary general hardware, and of course, it can also be implemented by special hardware including special integrated circuits, special CPUs, special memories, special components, etc. In general, all functions completed by computer programs can be easily implemented by corresponding hardware, and the specific hardware structure used to implement the same function can also be various, such as analog circuits, digital circuits or special circuits. However, for the present application, software program implementation is a better implementation mode in more cases. Based on such an understanding, the technical solution of the present application is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a readable storage medium, such as a computer floppy disk, a U disk, a mobile hard disk, a ROM, a RAM, a disk or an optical disk, etc., including a number of instructions to enable a computer device (which can be a personal computer, a training device, or a network device, etc.) to execute the methods described in each embodiment of the present application.
[0103] In the above embodiments, all or part of the embodiments may be implemented by software, hardware, firmware or any combination thereof. When implemented by software, all or part of the embodiments may be implemented in the form of a computer program product.
[0104] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from a website site, a computer, a training device, or a data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, training device, or data center. The computer-readable storage medium may be any available medium that a computer can store or a data storage device such as a training device, a data center, etc. that includes one or more available media integrations. The available medium may be a magnetic medium, (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive (SSD)), etc.
Claims
1. A method for monitoring a pressure swing adsorption device, characterized in that: include: Based on the three-dimensional model of the pressure swing adsorption device, the mechanism model of the pressure swing adsorption device and the operating parameters of the pressure swing adsorption device, a digital twin model of the pressure swing adsorption device is constructed; Based on the formation mechanism, conditions and detection results of various indicators of local defects of the pressure swing adsorption device under the action of positive and negative pressure cycles, combined with the digital twin model, a defect prediction and evaluation system of the pressure swing adsorption device is constructed; Based on the defect prediction and evaluation system of the pressure swing adsorption device, a neural network model is trained to obtain a defect prediction and evaluation model of the pressure swing adsorption device; Based on the defect prediction and evaluation model of the pressure swing adsorption device, the operating status of the pressure swing adsorption device is evaluated in real time, and the evaluation results are displayed in three dimensions in the digital twin model.
2. The method according to claim 1, characterized in that: The operating status of the pressure swing adsorption device is evaluated in real time, including: Using a PLC robot sensing and control system, the defect prediction and evaluation system communicates with the pressure swing adsorption device through a communication protocol to obtain real-time operating data and real-time nondestructive testing results of the pressure swing adsorption device; The real-time operation data and the real-time nondestructive testing results of the pressure swing adsorption device are input into the defect prediction and evaluation model to perform a real-time evaluation on the operation status of the pressure swing adsorption device.
3. The method according to claim 1, characterized in that The sources of the test results of the various indicators include: By means of nondestructive testing methods, a data set of crack defects, stress defects, and bulge defects in three dimensions is obtained during the service of the pressure swing adsorption device; Based on the data set, the wall thickness reduction, stress concentration, crack distribution, and location distribution of local temperature rise defects during the service process are determined, and a service equipment failure distribution database including the test results of various indicators is established.
4. The method according to any one of claims 1 to 3, characterized in that: The digital twin model of the pressure swing adsorption device is constructed based on the three-dimensional model of the pressure swing adsorption device, the mechanism model of the pressure swing adsorption device and the operating parameters of the pressure swing adsorption device, including: The calculation process and logic of the mechanism model are embedded into the corresponding components or objects of the 3D model, and the mechanism model is called for real-time calculation and simulation during the operation of the 3D model; The calculation results of the mechanism model are mapped to the corresponding positions of the three-dimensional model and compared with the operating status data of the pressure swing adsorption device during operation, so as to couple the multi-dimensional data and laws of multi-physical fields, realize the real-time symbiosis of the physical entity and virtual entity of the pressure swing adsorption device system, and obtain the digital twin model.
5. The method according to any one of claims 1 to 3, characterized in that: Based on the defect prediction and evaluation system of the pressure swing adsorption device, a neural network model training is performed to obtain a defect prediction and evaluation model of the pressure swing adsorption device, including: Cleaning and preprocessing the operating parameters and test results of various indicators of the pressure swing adsorption device, removing noise and outliers, marking defective data, and obtaining a training data set; The training data set is used to train the neural network, and the network weights and biases are continuously adjusted through the back propagation algorithm to minimize the prediction error. Regularization is used to prevent the model from overfitting, so as to obtain the defect prediction and evaluation model of the pressure swing adsorption device.
6. The method according to claim 2, characterized in that Inputting the real-time operation data and the real-time nondestructive testing results of the pressure swing adsorption device into the defect prediction and evaluation model to perform real-time evaluation on the operation status of the pressure swing adsorption device, including: Calculating time domain features and frequency domain features corresponding to the real-time operating parameters and the real-time nondestructive testing results to form a plurality of initial sample vectors; Using principal component analysis to reduce the dimension of the initial sample vector to obtain main characteristic components, and obtaining target sample vectors corresponding to the main characteristic components to form a defect analysis sample library; The target sample vector in the defect analysis sample library is input into the defect prediction and evaluation model; the defect prediction and evaluation model is used to perform real-time evaluation to obtain a defect evaluation result.
7. The method according to claim 6, characterized in that The initial sample vector is reduced in dimension by using the principal component analysis method to obtain the main characteristic components, and the target sample vector corresponding to the main characteristic components is obtained to form a defect analysis sample library, including: splicing the initial sample vectors into a sample feature matrix, and standardizing the sample feature matrix; The principal component analysis method is used to analyze the standardized sample feature matrix to obtain eigenvalues, and the main characteristic components are selected from multiple eigenvalues. The corresponding eigenvectors are matched as target sample vectors to form a defect analysis sample library.
8. A pressure swing adsorption device monitoring system, characterized in that: Including physical layer, data layer, model layer, application layer: The physical layer includes the pressure swing adsorption device, the data acquisition terminal, the multi-data integrated sensor system, and the communication interface; The data layer is used to collect the operating data of the pressure swing adsorption device and the test results of various indicators; A model layer is used to construct a digital twin model of the pressure swing adsorption device based on the three-dimensional model of the pressure swing adsorption device, the mechanism model of the pressure swing adsorption device and the operating parameters of the pressure swing adsorption device; based on the formation mechanism, conditions and detection results of various indicators of local defects of the pressure swing adsorption device under the action of positive and negative pressure cycles, in combination with the digital twin model, a defect prediction and evaluation system of the pressure swing adsorption device is constructed; based on the defect prediction and evaluation system of the pressure swing adsorption device, a neural network model is trained to obtain a defect prediction and evaluation model of the pressure swing adsorption device; based on the defect prediction and evaluation model of the pressure swing adsorption device, the operating status of the pressure swing adsorption device is evaluated in real time; The application layer is used to provide a user interface to display the evaluation results in three dimensions in the digital twin model.
9. An electronic device, characterized in that: include: At least one memory, at least one processor, the at least one processor is used to execute instructions stored in the at least one memory, so that the electronic device performs the method as described in any one of claims 1 to 7.
10. A computer-readable storage medium, wherein a program is stored in the computer-readable storage medium, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.
Citation Information
Patent Citations
Compressed air energy storage system state evaluation method based on digital twinborn model
CN117993281A
Transformer vibration monitoring and fault diagnosis method
CN118070180A
Cited By
Damage assessment method and device for pressure swing adsorption device and storage medium
CN119803988A
Method and device for assessing damage to pressure swing adsorption apparatus and storage medium
CN119803988B
Pressure release two-way rapid cutting-off method and device for gas supply system of pressure swing adsorption device
CN120742991A
Pressure relief bidirectional quick cut-off method and device for gas supply system of pressure swing adsorption device
CN120742991B
