Pressure vessel safety evaluation early warning method based on digital twinning

By analyzing the wall thickness difference distribution and stress analysis of the pressure vessel, the resonance buckling instability probability is calculated and a safety evaluation logic model is constructed, which solves the problems of inaccurate analysis and large errors in traditional methods, and improves the safety management level and early warning efficiency.

CN119047376AActive Publication Date: 2024-11-29菏泽市产品检验检测研究院

Patent Information

Application Number
CN202411488511.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-10-24
Publication Date
2024-11-29
Estimated Expiration
2044-10-24

AI Technical Summary

Technical Problem

The traditional pressure vessel safety evaluation and early warning method based on digital twins has problems such as inaccurate analysis of the probability state of instability and large errors in safety evaluation and early warning.

Method used

By obtaining pressure vessel design data, calculating the wall thickness difference distribution, dividing structural similarity areas, analyzing the difference in stress distribution of connecting structures, calculating the probability of resonance buckling instability, and building a safety evaluation logic model, and finally uploading the model to the data cloud platform for Early warning.

Benefits of technology

Improve the accuracy of the probability state analysis of the pressure vessel instability and the accuracy of safety evaluation and early warning, reduce errors, realize the automated and intelligent safety management of the pressure vessel, and ensure its reliability and stability under complex working conditions. sex.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119047376A_ABST
    Figure CN119047376A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of safety evaluation early warning, in particular to a pressure vessel safety evaluation early warning method based on digital twinning. The method comprises the following steps: carrying out wall thickness difference distribution calculation on pressure vessel design data, and carrying out structural similarity region division to obtain structural form wall thickness region division data; carrying out resonance buckling instability probability calculation based on the structural form wall thickness region division data to obtain resonance buckling instability probability data; performing safety evaluation logic model construction on the resonance buckling instability probability data to obtain a pressure vessel safety evaluation logic model; and sending the pressure vessel safety evaluation logic model to a data cloud platform to execute the pressure vessel safety evaluation early warning method. According to the invention, the safety evaluation early warning technology is optimized, so that the safety evaluation early warning technology is more perfect.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of safety evaluation and early warning technology, and in particular to a pressure vessel safety evaluation and early warning method based on digital twins. Background Art

[0002] Pressure vessels are widely used in the fields of petroleum, chemical industry, energy, etc. They are subjected to high temperature, high pressure and corrosive environment for a long time, which is very easy to produce fatigue, cracks or structural failure, leading to equipment damage or safety accidents. Digital twin technology constructs a virtual model of the pressure vessel to simulate the working state of its physical entity under different working conditions, and combines the real-time data collected by the sensor to conduct real-time dynamic monitoring of key parameters such as temperature, pressure, stress, corrosion, and cracks of the vessel. The virtual model can not only reflect the current operating state of the vessel, but also predict the future working conditions, and simulate the damage evolution process of the vessel based on a variety of physical and chemical models. By combining the historical data, design parameters and real-time operation data of the pressure vessel, the digital twin model can be continuously updated and optimized to improve the simulation accuracy and realize the accurate assessment of the health status of the pressure vessel. However, a traditional pressure vessel safety evaluation and early warning method based on digital twins has the problems of inaccurate analysis of the probability of instability of the pressure vessel and large error in the safety evaluation and early warning of the pressure vessel. Summary of the invention

[0003] Based on this, it is necessary to provide a pressure vessel safety evaluation and early warning method based on digital twins to solve at least one of the above technical problems.

[0004] To achieve the above purpose, a pressure vessel safety evaluation and early warning method based on digital twin is provided, and the method comprises the following steps: Step S1: obtaining pressure vessel design data; performing wall thickness difference distribution calculation on the pressure vessel design data to obtain structural morphology wall thickness difference distribution data; performing structural similarity region division according to the structural morphology wall thickness difference distribution data to obtain structural morphology wall thickness region division data; Step S2: performing a stress distribution difference analysis of the connection structure based on the structural morphology wall thickness area division data to obtain stress distribution difference data of the connection structure; performing a resonance buckling instability probability calculation based on the stress distribution difference data of the connection structure to obtain resonance buckling instability probability data; Step S3: constructing a safety evaluation logic model for the resonant buckling instability probability data to obtain a safety evaluation logic model for the pressure vessel; Step S4: Send the pressure vessel safety assessment logic model to the data cloud platform to execute the pressure vessel safety assessment early warning method.

[0005] The present invention obtains the design data of the pressure vessel and performs detailed calculations on the wall thickness difference, which can reveal the distribution characteristics of the wall thickness under different structural forms. This process can not only clarify the stress state of each area, but also help designers to deeply understand the influence of wall thickness on the overall performance of the structure. By dividing the area of ​​wall thickness difference, designers can more accurately identify potential weak links and provide a basis for subsequent improvement measures. In addition, the stress distribution difference analysis based on these wall thickness area divisions can reveal the stress characteristics of the connection structure under different working conditions and evaluate its stability. By analyzing the stress distribution difference of the connection structure, important data support can be provided for the probability calculation of resonant buckling instability. Understanding the stress state of different areas is crucial to predicting its instability risk. By comprehensively considering these data, designers can identify the key stress points of the structure and evaluate its safety under actual working conditions. This data-based analysis method enables designers to optimize the structure more specifically, thereby improving the safety and durability of the pressure vessel, reducing the risk of failure, and extending the service life. Therefore, the entire process not only improves the scientificity and rationality of the design, but also provides an important theoretical basis for subsequent engineering applications. Through the implementation of step S3 and step S4, the safety evaluation capability and early warning efficiency of pressure vessels can be significantly improved. First, in step S3, a safety evaluation logic model based on resonant buckling instability probability data is constructed to provide a scientific theoretical basis and decision support for the safety evaluation of pressure vessels. This logic model can not only comprehensively consider multiple influencing factors, but also comprehensively analyze the stability of pressure vessels by simulating the stress state under different working conditions. This modeled safety evaluation method can effectively identify potential safety hazards and help designers to formulate corresponding preventive measures in a timely manner, thereby improving the safety of pressure vessels in practical applications. Next, step S4 realizes the automation and intelligence of pressure vessel safety evaluation by uploading the safety evaluation logic model to the data cloud platform. The powerful computing power and data processing capabilities of the cloud platform can monitor and analyze the operating status of the pressure vessel in real time. Once a potential safety risk is detected, the cloud platform can issue an early warning in time to help relevant personnel take measures quickly to prevent accidents. This cloud computing-based safety evaluation and early warning method greatly improves the safety management level of pressure vessels and ensures their reliability and stability under complex working conditions. In addition, this method also provides real-time data support for subsequent safety management decisions, which helps to optimize maintenance strategies and improve resource allocation efficiency, ultimately achieving safe, economical and sustainable engineering management goals.Therefore, the present invention optimizes a traditional pressure vessel safety evaluation and early warning method based on digital twins, solves the problems of inaccurate analysis of the pressure vessel instability probability state and large error in the pressure vessel safety evaluation and early warning method based on digital twins, improves the accuracy of the pressure vessel instability probability state analysis, and reduces the problem of error in the pressure vessel safety evaluation and early warning.

[0006] Preferably, step S1 comprises the following steps: Step S11: obtaining pressure vessel design data; Step S12: performing structural morphology analysis on the pressure vessel design data to obtain structural morphology data of the pressure vessel; Step S13: performing wall thickness difference distribution calculation on the pressure vessel structure morphology data to obtain structure morphology wall thickness difference distribution data; Step S14: dividing the pressure vessel structure data into structural similarity regions according to the structure-form wall thickness difference distribution data to obtain structure-form wall thickness region division data.

[0007] The present invention can systematically improve the design accuracy and safety of the pressure vessel through the preferred implementation of step S1. First, in step S11, obtaining the design data of the pressure vessel lays the foundation for subsequent analysis. These data cover all aspects of the design and provide important information for in-depth understanding of structural performance. In step S12, the structural morphology analysis of the pressure vessel can help designers identify the geometric characteristics and force characteristics of different components, thereby obtaining the structural morphology data of the pressure vessel. This analysis process is crucial for a comprehensive understanding of the response of the structure under various loads. Subsequently, in step S13, by calculating the wall thickness difference distribution of the structural morphology data, the wall thickness changes in each area can be revealed. This differentiated wall thickness distribution data not only helps to identify weak links, but also provides a specific basis for design optimization. In addition, the structural similarity area division based on the wall thickness difference data in step S14 further subdivides the structural morphology data to form different wall thickness area division data. The result of this area division provides designers with a more refined perspective, enabling them to make design improvements and safety assessments more targeted. Overall, the implementation of this series of steps not only improves the design quality of pressure vessels, but also provides scientific data support for subsequent safety analysis and optimization, ensuring the reliability and safety of pressure vessels during use.

[0008] Preferably, step S2 comprises the following steps: Step S21: constructing a virtualized pressure vessel regional structure model based on the structural morphology and wall thickness regional division data based on the digital twin to obtain a virtualized pressure vessel regional structure model; Step S22: performing regional structural fluid load simulation on the virtualized pressure vessel regional structural model to obtain regional structural fluid load data; Step S23: Based on the structural form wall thickness difference distribution data and the structural form wall thickness regional division data, the regional structural fluid load data is subjected to a connection structure force distribution difference analysis to obtain the connection structure force distribution difference data; Step S24: Calculate the resonant buckling instability probability of the regional structure fluid load data according to the force distribution difference data of the connection structure to obtain the resonant buckling instability probability data.

[0009] The present invention constructs a virtualized pressure vessel regional structure model based on digital twin technology, laying a solid foundation for subsequent simulation and analysis. The virtualized model can accurately reflect the actual structural characteristics of the pressure vessel, and provides a necessary virtual environment for in-depth fluid load analysis, helping designers to intuitively understand the performance of the structure under different working conditions. Next, in step S22, the regional structural fluid load simulation is performed on the virtualized model to obtain regional structural fluid load data. These data not only reveal the way the fluid acts inside the structure, but also help analyze the force characteristics of each region, ensuring that designers can effectively evaluate the dynamic load of the fluid. In step S23, based on the structural morphology wall thickness difference distribution data and the regional division data, the force distribution difference analysis of the connection structure is performed, and the force conditions of each connection part are further clarified, helping to identify potential weak connections and stress concentration areas. Finally, in step S24, the probability of resonant buckling instability of the regional structural fluid load data is calculated according to the force distribution difference data of the connection structure, and this process is extremely critical. It provides designers with the probability of resonant buckling instability of the relevant structure under actual operating conditions, helping to identify risks in advance and take appropriate preventive measures. Overall, the implementation of these steps not only improves the safety performance assessment capabilities of pressure vessels, but also provides data support for optimizing designs and improving solutions, ensuring the stability and safety of pressure vessels under complex working conditions.

[0010] Preferably, step S23 includes the following steps: Step S231: performing geometric characteristic analysis of connection structures between different regions based on the structural form wall thickness difference distribution data and the structural form wall thickness region division data to obtain geometric characteristic data of the connection structure; Step S232: Calculating adjacent connection angles between different regions of the connection structure geometric feature data to obtain structure adjacent connection angle data; Step S233: calculating the connection surface slopes between different regions of the connection structure geometric feature data to obtain structure connection surface slope data; Step S234: performing fluid force dispersion analysis between different regions on the regional structural fluid load data according to the structural adjacent connection angle data and the structural connection surface slope data, to obtain regional structural force dispersion data; Step S235: performing connection structure force distribution difference analysis on the regional structure fluid load data according to the regional structure force dispersion data to obtain connection structure force distribution difference data.

[0011] First, in step S231, the present invention performs a connection structure geometric feature analysis based on the structural morphology wall thickness difference distribution data and the regional division data, and the obtained connection structure geometric feature data provides a basis for subsequent force analysis. This analysis can deeply understand the geometric relationship and characteristics of the connection parts in different regions, and help identify potential problems in the design. Then, in step S232, the adjacent connection angle calculation performed on the connection structure geometric feature data can reveal the geometric shape of the connection part and its influence on the fluid load. This analysis is crucial to ensure the stability of the structure, because too large an angle leads to stress concentration, thereby affecting the overall performance of the structure. In step S233, the calculation of the slope of the connection surface further provides detailed information on the relative position between the connection surfaces, helping to analyze the stress state of each connection surface under the action of the fluid. In step S234, according to the structural adjacent connection angle data and the connection surface slope data, the regional structure fluid load data is subjected to fluid force dispersion analysis, which can effectively identify the force dispersion of each region. This process ensures that designers can clearly understand the distribution and action mode of the fluid inside the structure, thereby optimizing the design to reduce risks. Finally, step S235 performs a differential analysis of the stress distribution of the connection structure based on the regional structural stress dispersion data, further enhancing the understanding of the stress conditions of each connection point. The implementation of this series of steps makes the design process of the pressure vessel more scientific and refined, which helps to improve its safety and reliability in practical applications and ensure stability under various working conditions.

[0012] Preferably, step S234 includes the following steps: Decomposing the connection surface normal component according to the structural adjacent connection angle data and the structural connection surface slope data to obtain the connection surface normal component data; Conducting interface force dispersion analysis on the connection surface normal component data to obtain interface force dispersion data; According to the connection surface normal component data and the interface force dispersion data, the fluid shear force orientation difference between different regions is calculated for the regional structural fluid load data to obtain the fluid shear force orientation difference data; Based on the fluid shear force azimuth difference data and the connection surface normal component data, the fluid force dispersion between different regions is analyzed to obtain the regional structural force dispersion data.

[0013] The present invention can significantly improve the design safety of pressure vessels and the prediction accuracy of fluid loads by implementing the detailed steps of fluid force dispersion analysis. In the analysis process, the connection surface normal component is first decomposed through the structural adjacent connection angle data and the connection surface slope data to obtain the connection surface normal component data. This process lays the foundation for understanding the direction of the force acting on the connection surface by the fluid, ensuring the accuracy of subsequent analysis. The clarification of the connection surface normal component can help designers identify which connection areas are facing greater fluid pressure, so as to make design adjustments in time. Next, the interface force dispersion analysis is performed based on the connection surface normal component data to obtain the interface force dispersion data. This analysis reveals the difference in stress state between different connection surfaces, helps identify stress concentration areas caused by improper design or structural shape, and provides an important basis for optimizing design. Then, according to the connection surface normal component data and the interface force dispersion data, the fluid shear force azimuth difference of the regional structural fluid load data is calculated. This calculation can clarify the force direction and intensity changes of the fluid on the connection surface, thereby revealing the potential impact of fluid shear force between different regions. Finally, the regional structural force dispersion analysis based on the fluid shear force azimuth difference data and the connection surface normal component data can comprehensively evaluate the force dispersion of the fluid in different regions. Through the implementation of this series of steps, not only the understanding of the fluid force distribution of the pressure vessel under working conditions is improved, but also detailed data support is provided for subsequent safety assessment and optimization design. This refined analysis method ensures the safety and stability of the pressure vessel in actual use and promotes the scientific and rational design process.

[0014] Preferably, step S24 comprises the following steps: Step S241: performing fluid resonance frequency simulation on the regional structure fluid load data according to the force distribution difference data of the connection structure to obtain regional fluid resonance frequency data; Step S242: Calculating the load section moment of inertia between different regional structures on the regional structural fluid load data based on the regional fluid resonance frequency data to obtain structural load section moment of inertia data; Step S243: performing inertial distortion integration on the structural load section inertia moment data to obtain section inertial distortion integration data; Step S244: performing structural tolerance limit simulation on the regional structural fluid load data according to the cross-section inertial distortion integral data to obtain regional structural tolerance limit data; Step S245: Calculate the resonant buckling instability probability of the regional structure fluid load data according to the cross-section inertial distortion integral data and the regional structure tolerance limit data to obtain the resonant buckling instability probability data.

[0015] The present invention first, in step S241, performs fluid resonance frequency simulation on the regional structural fluid load data based on the force distribution difference data of the connection structure, and the regional fluid resonance frequency data obtained provides key parameters for subsequent analysis. This step can reveal the vibration characteristics of the fluid in different regions, help identify potential resonance risk areas, and ensure that designers can optimize in a targeted manner. Subsequently, in step S242, based on the regional fluid resonance frequency data, the load section inertia moment between the structures in different regions is calculated to obtain the structural load section inertia moment data. This calculation can not only help understand the stress conditions of each region under different loads, but also provide an important basis for evaluating the stability of the structure. In the following step S243, by performing inertial distortion integration on the structural load section inertia moment data, the section inertia distortion integral data is obtained, and the structural deformation characteristics under the action of the fluid are further analyzed, which lays a foundation for evaluating the adaptability of the structure. In step S244, the structural tolerance limit simulation is performed on the regional structural fluid load data based on the section inertia distortion integral data, and the regional structural tolerance limit data can be obtained. This step is very important because it helps designers to clarify the bearing capacity of the structure under extreme working conditions and provides a scientific basis for safety assessment. Finally, step S245 calculates the probability of resonant buckling instability based on the cross-sectional inertial distortion integral data and the regional structural tolerance limit data. The resonant buckling instability probability data obtained provides a key decision-making basis for designers to identify and prevent potential instability risks. The implementation of this series of steps not only improves the safety assessment capability of pressure vessels under complex working conditions, but also provides practical data support for the optimal design of the structure, ensuring its long-term stability and safety in practical applications.

[0016] Preferably, step S243 includes the following steps: Discretize the moment of inertia of different sections of the structural load section moment of inertia data to obtain the discretized data of the section moment of inertia; According to the discretized data of section inertia moment, nonlinear regression analysis of the structure load section inertia moment data of different sections is performed to obtain the nonlinear regression data of the moment of inertia; Performing piecewise interpolation processing on the nonlinear regression data of the moment of inertia to obtain the nonlinear interpolation data of the moment of inertia; The inertia distortion integral is performed on the nonlinear interpolation data of the moment of inertia using the Gauss-Legendre integration method to obtain the section inertia distortion integral data.

[0017] The present invention can greatly improve the design accuracy and safety of pressure vessels by implementing the detailed steps of inertial distortion integration of structural load section inertia data. First, in this process, the section inertia discretization data is obtained by discretizing the inertia of different sections of the structural load section inertia data. The implementation of this step enables designers to convert the overall complex section characteristics into discrete data that are easier to analyze, which is convenient for subsequent nonlinear regression analysis. Next, the nonlinear regression analysis based on the section inertia discretization data can reveal the complex relationship between the section inertia moment and other variables, thereby obtaining the inertia moment nonlinear regression data. This analysis not only helps to understand the response characteristics of each section under the action of fluid load, but also provides the necessary mathematical model for subsequent interpolation processing, ensuring the accuracy and effectiveness of the calculation. In the subsequent segmented interpolation processing, the nonlinear interpolation data of the inertia moment is obtained by processing the nonlinear regression data of the inertia moment, which further improves the smoothness and continuity of the section inertia moment data, which is convenient for use in subsequent calculations. Finally, by using the Gauss-Legendre integral method to perform inertial distortion integration on the nonlinear interpolation data of the moment of inertia, the cross-sectional inertial distortion integral data was obtained. This efficient numerical integration method can effectively calculate the degree of distortion of complex structures under the action of fluids, providing an important basis for evaluating the stability and reliability of the structure. The implementation of this series of steps not only deepens the understanding of the structural stress characteristics, but also provides strong support for the safe design and operation of pressure vessels, ensuring their safety and stability under various working conditions.

[0018] Preferably, step S244 includes the following steps: Extract the extreme value of section distortion resistance according to the section inertial distortion integral data to obtain the extreme value data of section distortion resistance; Based on the extreme value data of cross-section distortion resistance, the structural distortion evolution response analysis is performed on the regional structural fluid load data to obtain the structural distortion evolution response data; Perform structural fracture failure simulation on the structural distortion evolution response data to obtain structural fracture failure data; According to the structural fracture failure data and the structural distortion evolution response data, the structural tolerance limit of the regional structural fluid load data is simulated to obtain the regional structural tolerance limit data.

[0019] The present invention can effectively improve the safety assessment capability of pressure vessels under complex working conditions by implementing the steps of simulating the structural tolerance limit of regional structural fluid load data. First, the section distortion resistance extreme value extraction is performed based on the section inertial distortion integral data to obtain the section distortion resistance extreme value data. The acquisition of this data lays the foundation for subsequent structural analysis, which can help designers to clarify the resistance characteristics of the section under different fluid loads, thereby identifying the existing weak links and potential failure modes. Next, the structural distortion evolution response analysis based on the section distortion resistance extreme value data can deeply explore the influence of fluid load on the structural distortion characteristics, and the obtained structural distortion evolution response data provides detailed information for evaluating the performance of the structure under dynamic load. This analysis can reveal the behavioral changes of the structure under stress state, helping designers to better understand the influence of fluid on structural stability and overall safety. In addition, further structural fracture failure simulation of the structural distortion evolution response data can identify the fracture point and failure mode in advance, providing a practical basis for the optimization design of the structure. Finally, the regional structural tolerance limit simulation based on the structural fracture failure data and the structural distortion evolution response data can obtain the regional structural tolerance limit data. This simulation result provides a scientific basis for designers to ensure that the structure can maintain sufficient safety margin under extreme conditions to prevent structural failure caused by fluid load. Through the implementation of this series of steps, not only the understanding of the stress and failure of pressure vessels under complex conditions is strengthened, but also solid data support is provided for subsequent design optimization and safety verification, ensuring the long-term safety and stable operation of pressure vessels.

[0020] Preferably, step S3 comprises the following steps: Step S31: normalizing the resonant buckling instability probability data to obtain resonant buckling instability probability normalized data; Step S32: constructing a safety evaluation logic model for the normalized data of the resonant buckling instability probability based on the random forest algorithm to obtain a safety evaluation logic model for the pressure vessel.

[0021] The present invention can effectively improve the safety assessment capability of pressure vessels and ensure its reliability in practical applications through the implementation of step S3. First, in step S31, the resonant buckling instability probability data is normalized to obtain resonant buckling instability probability normalized data. The main purpose of normalization is to convert data of different dimensions and ranges into a unified standard, thereby eliminating the relative differences between data. This step can ensure the consistency of data in subsequent analysis and facilitate comparison and analysis. The normalized data is easier to process mathematically, while enhancing the stability and reliability of the model, providing a clearer basis for subsequent safety evaluation. Next, in step S32, based on the normalized data, a safety evaluation logic model of the pressure vessel is constructed using a random forest algorithm. Random forest is an integrated learning method that has the advantages of processing high-dimensional data and complex relationships, and can effectively capture the nonlinear relationship between the probability of resonant buckling instability and other related factors. This step not only provides an effective tool for safety evaluation, but also can improve the prediction accuracy and robustness of the model through the combination of multiple decision trees. The safety evaluation logic model constructed can help engineers quickly evaluate the safety of pressure vessels under various working conditions during the design phase, identify potential failure risks in advance, and provide a scientific basis for optimizing design and safety verification. Through the implementation of this series of steps, the safety assessment capability of pressure vessels is ultimately improved, providing solid data support and decision-making basis for subsequent safety design, ensuring its reliability and stability under extreme working conditions.

[0022] Preferably, step S32 includes the following steps: Step S321: performing a logic error test on the normalized data of the resonant buckling instability probability to obtain the resonant buckling instability logic error data; Step S322: performing error balance iteration on the resonant buckling instability probability normalized data according to the resonant buckling instability logic error data to obtain resonant buckling error balance data; Step S323: performing time-effect characteristic analysis on the resonant buckling error balance data to obtain error time-effect characteristic data; Step S324: constructing a safety evaluation logic model for the resonant buckling error balance data and the error time-effect characteristic data based on the random forest algorithm to obtain a safety evaluation logic model for the pressure vessel.

[0023] The present invention can significantly enhance the safety assessment capability of pressure vessels under dynamic conditions and ensure its reliability and stability in actual use through the system implementation of step S32. First, in step S321, a logical error test is performed on the normalized data of the probability of resonant buckling instability to obtain the logical error data of resonant buckling instability. This step is crucial because it can help identify potential error sources and inconsistencies in the model and provide data support for subsequent error correction. The results of the logical error test will provide key clues for understanding the influencing factors of the probability of resonant buckling instability and ensure that subsequent analysis is based on more reliable data. Next, in step S322, based on the logical error data of resonant buckling instability, the normalized data of the probability of resonant buckling instability is subjected to error balance iteration to obtain the resonant buckling error balance data. Through the process of error balance iteration, the original data can be effectively corrected and optimized, the influence caused by errors can be reduced, and the accuracy and reliability of the data can be improved. The implementation of this step provides more accurate data for the subsequent analysis of the model, so that the actual situation can be better reflected when constructing a safety evaluation model. In step S323, the time-effect characteristic analysis of the resonant buckling error balance data is performed to obtain the error time-effect characteristic data. This analysis process can help engineers understand the impact of time factors on the probability of resonant buckling instability, identify changes in the performance of pressure vessels after long-term use or under specific conditions, and thus provide a basis for maintenance and monitoring. These data will enhance the adaptability and accuracy of the model to actual working conditions. Finally, in step S324, based on the random forest algorithm, a safety evaluation logic model is constructed for the resonant buckling error balance data and the error time-effect characteristic data to obtain a pressure vessel safety evaluation logic model. By taking advantage of the random forest, important features can be extracted from complex data sets to generate an efficient logic model. This model can not only quickly evaluate the safety of pressure vessels, but also provide real-time feedback for optimal design, and improve the safety and reliability of pressure vessels under various working conditions. Through the implementation of this series of steps, the pressure vessel safety evaluation logic model finally constructed will provide solid data support and decision-making basis for engineering design, operation and maintenance.

[0024] The beneficial effect of the present invention is that the design data of the pressure vessel is obtained and the wall thickness difference is calculated in detail, which can reveal the distribution characteristics of the wall thickness under different structural forms. This process can not only clarify the stress state of each area, but also help designers to deeply understand the influence of wall thickness on the overall performance of the structure. By dividing the area of ​​wall thickness difference, designers can more accurately identify potential weak links and provide a basis for subsequent improvement measures. In addition, the stress distribution difference analysis based on these wall thickness area divisions can reveal the stress characteristics of the connection structure under different working conditions and evaluate its stability. By analyzing the stress distribution difference of the connection structure, important data support can be provided for the probability calculation of resonant buckling instability. Understanding the stress state of different areas is crucial to predicting its instability risk. By comprehensively considering these data, designers can identify the key stress points of the structure and evaluate its safety under actual working conditions. This data-based analysis method enables designers to optimize the structure more specifically, thereby improving the safety and durability of the pressure vessel, reducing the risk of failure, and extending the service life. Therefore, the entire process not only improves the scientificity and rationality of the design, but also provides an important theoretical basis for subsequent engineering applications. Through the implementation of step S3 and step S4, the safety evaluation capability and early warning efficiency of pressure vessels can be significantly improved. First, in step S3, a safety evaluation logic model based on resonant buckling instability probability data is constructed to provide a scientific theoretical basis and decision support for the safety evaluation of pressure vessels. This logic model can not only comprehensively consider multiple influencing factors, but also comprehensively analyze the stability of pressure vessels by simulating the stress state under different working conditions. This modeled safety evaluation method can effectively identify potential safety hazards and help designers to formulate corresponding preventive measures in a timely manner, thereby improving the safety of pressure vessels in practical applications. Next, step S4 realizes the automation and intelligence of pressure vessel safety evaluation by uploading the safety evaluation logic model to the data cloud platform. The powerful computing power and data processing capabilities of the cloud platform can monitor and analyze the operating status of the pressure vessel in real time. Once a potential safety risk is detected, the cloud platform can issue an early warning in time to help relevant personnel take measures quickly to prevent accidents. This cloud computing-based safety evaluation and early warning method greatly improves the safety management level of pressure vessels and ensures their reliability and stability under complex working conditions. In addition, this method also provides real-time data support for subsequent safety management decisions, which helps to optimize maintenance strategies and improve resource allocation efficiency, ultimately achieving safe, economical and sustainable engineering management goals.Therefore, the present invention optimizes a traditional pressure vessel safety evaluation and early warning method based on digital twins, solves the problems of inaccurate analysis of the pressure vessel instability probability state and large error in the pressure vessel safety evaluation and early warning method based on digital twins, improves the accuracy of the pressure vessel instability probability state analysis, and reduces the problem of error in the pressure vessel safety evaluation and early warning. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 It is a schematic diagram of the steps of a pressure vessel safety evaluation and early warning method based on digital twins; Figure 2 for Figure 1 Detailed implementation steps of step S2 in FIG.

[0026] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0027] The technical method of the present invention is described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by technicians in this field without creative work are within the scope of protection of the present invention.

[0028] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0029] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0030] To achieve this, please refer to Figure 1 to Figure 2, a pressure vessel safety evaluation and early warning method based on digital twins, the method comprising the following steps: Step S1: obtaining pressure vessel design data; performing wall thickness difference distribution calculation on the pressure vessel design data to obtain structural morphology wall thickness difference distribution data; performing structural similarity region division according to the structural morphology wall thickness difference distribution data to obtain structural morphology wall thickness region division data; Step S2: performing a stress distribution difference analysis of the connection structure based on the structural morphology wall thickness area division data to obtain stress distribution difference data of the connection structure; performing a resonance buckling instability probability calculation based on the stress distribution difference data of the connection structure to obtain resonance buckling instability probability data; Step S3: constructing a safety evaluation logic model for the resonant buckling instability probability data to obtain a safety evaluation logic model for the pressure vessel; Step S4: Send the pressure vessel safety assessment logic model to the data cloud platform to execute the pressure vessel safety assessment early warning method.

[0031] In the embodiment of the present invention, reference Figure 1 The above is a schematic diagram of the steps of a pressure vessel safety evaluation and early warning method based on digital twins of the present invention. In this example, the pressure vessel safety evaluation and early warning method based on digital twins includes the following steps: Step S1: obtaining pressure vessel design data; performing wall thickness difference distribution calculation on the pressure vessel design data to obtain structural morphology wall thickness difference distribution data; performing structural similarity region division according to the structural morphology wall thickness difference distribution data to obtain structural morphology wall thickness region division data; In an embodiment of the present invention, the acquisition of pressure vessel design data includes extracting key information such as structural dimensions and material properties from the manufacturer, design drawings and engineering standards and specifications of the pressure vessel. These design data are standardized to ensure that data from different sources have consistent formats and units. In the calculation of wall thickness difference distribution, based on finite element analysis (FEA) technology, stress analysis theory is applied to calculate different wall thickness areas of the pressure vessel. By refining the discrete distribution of wall thickness data, the difference between the wall thickness fluctuation range of each area and the standard wall thickness is extracted. In order to ensure the calculation accuracy, the deviation between the local wall thickness and the overall wall thickness distribution is calculated using the numerical integration method, and "structural morphology wall thickness difference distribution data" is generated. The data will be stored in a matrix form, and the degree of difference of each wall thickness area will be distinguished to support subsequent structural morphology analysis.

[0032] Step S2: performing a stress distribution difference analysis of the connection structure based on the structural morphology wall thickness area division data to obtain stress distribution difference data of the connection structure; performing a resonance buckling instability probability calculation based on the stress distribution difference data of the connection structure to obtain resonance buckling instability probability data; In the embodiment of the present invention, the "structural morphology wall thickness area division data" generated in step S1 is used, combined with the stress distribution theory, and the connection structure (such as welds, flange connections, etc.) of the pressure vessel is subjected to stress analysis by the finite element method. The stress field distribution model of the connection structure under different load conditions is established by using elastic-plastic mechanics and structural analysis methods. The "connection structure force distribution difference data" is generated by numerically iteratively calculating the stress distribution in the wall thickness difference area. In this process, the stress concentration coefficient of different connection parts under various stresses (such as internal pressure, external pressure and torsion, etc.) needs to be calculated, and then the distribution difference of the force in each area is evaluated. The final data represents the force changes in different areas in the form of a three-dimensional stress field diagram. The probability of resonant buckling instability is calculated for the "connection structure force distribution difference data" through the buckling theory in structural mechanics. The buckling instability calculation needs to consider the nonlinear characteristics of the material and the stress hardening effect of the pressure vessel, and the critical stress value of different connection parts is analyzed by the energy method. By establishing a random vibration analysis model, the probability of structural buckling instability under the action of external random vibration is calculated. The Monte Carlo method is used to perform multiple iterative calculations on the probability of resonant buckling instability, and finally the "resonant buckling instability probability data" is obtained, which represents the instability risks of different parts in the form of a multi-dimensional matrix.

[0033] Step S3: constructing a safety evaluation logic model for the resonant buckling instability probability data to obtain a safety evaluation logic model for the pressure vessel; In an embodiment of the present invention, after obtaining the "resonant buckling instability probability data", a pressure vessel safety evaluation logic model is constructed. The construction of the logic model is based on fuzzy logic theory and the analytic hierarchy process (AHP). Through quantitative and qualitative analysis of each key risk factor, factors such as structural buckling, force difference, wall thickness distribution, and environmental load of the pressure vessel are incorporated into the model. Using the safety margin assessment method, different parameters are weighted, and the comprehensive safety factor of each structural part is obtained through logical operations. The model output is the safety evaluation level of each structure, and the logical relationship is expressed in the form of a decision tree. The final generated pressure vessel safety evaluation logic model can automatically adapt to different load conditions and environmental conditions.

[0034] Step S4: Send the pressure vessel safety assessment logic model to the data cloud platform to execute the pressure vessel safety assessment early warning method.

[0035] In an embodiment of the present invention, the operating status of the pressure vessel is monitored in real time by an Internet of Things (IoT) device, and the real-time monitoring data is combined with the safety evaluation logic model constructed in step S3. In the data cloud platform, the operating safety of the current pressure vessel is evaluated by real-time analysis of the received data. The data cloud platform will regularly analyze the monitoring data and trigger the warning signal according to the risk warning rules in the safety evaluation logic model. When the safety factor of a certain area is lower than the set threshold, the platform will automatically send a safety warning message to remind the relevant maintenance personnel to intervene. At the same time, the data cloud platform will generate a detailed safety assessment report, including historical operating data and risk trend analysis, to ensure continuous monitoring and warning of the operating status of the pressure vessel.

[0036] Preferably, step S1 comprises the following steps: Step S11: obtaining pressure vessel design data; Step S12: performing structural morphology analysis on the pressure vessel design data to obtain structural morphology data of the pressure vessel; Step S13: performing wall thickness difference distribution calculation on the pressure vessel structure morphology data to obtain structure morphology wall thickness difference distribution data; Step S14: dividing the pressure vessel structure data into structural similarity regions according to the structure-form wall thickness difference distribution data to obtain structure-form wall thickness region division data.

[0037] In an embodiment of the present invention, the pressure vessel design data is obtained by extracting drawings and specification documents from the pressure vessel manufacturer, including key structural information and parameters, such as diameter, length, material properties, weld position, design pressure and design temperature. The data includes two-dimensional CAD drawings, three-dimensional model data, and technical requirements in engineering standards and specifications. Through standardized technical processes, the design data is converted into a format that can be further processed to ensure that each value and parameter is consistent in the subsequent calculations. These data will be used for calculation and analysis in subsequent steps. The pressure vessel structural morphology analysis is carried out using the finite element analysis (FEA) method. First, the design data of the pressure vessel is meshed to establish an accurate three-dimensional model. The meshing adopts adaptive mesh generation technology to divide the key parts of the pressure vessel, such as connection points, welds and local weak areas, into denser meshes to improve the calculation accuracy. The key structural areas of the pressure vessel are identified by static and dynamic analysis of the structural model. Then, the morphological analysis technology is applied to extract the geometric morphological features of the pressure vessel, including the surface shape, corner radius and spatial position distribution of the weld. After multiple iterations and corrections, the "pressure vessel structural morphology data" is generated. The data is stored in a multidimensional space and contains detailed information on each local structure. When processing the "pressure vessel structural morphology data", the wall thickness of each structural area is first measured and compared with the standard wall thickness in the design specification. The wall thickness difference distribution calculation uses the deviation analysis method in statistics to calculate the difference between the wall thickness values ​​of different areas and the standard value. Specifically, the area of ​​each wall thickness area is calculated using the numerical integration method to ensure the accuracy of the data. At the same time, in order to capture subtle changes in wall thickness, multidimensional difference calculation is used, especially in the areas around local weak points and stress concentration points, to ensure that the wall thickness changes in these areas are accurately recorded. The generated "structural morphology wall thickness difference distribution data" is stored in the form of two-dimensional and three-dimensional matrices, and the spatial distribution of wall thickness differences is displayed in the form of color gradient maps for subsequent analysis. Based on the "structural morphology wall thickness difference distribution data", a clustering algorithm is used to divide the structure of the pressure vessel by similarity. This step uses the K-means clustering algorithm to classify areas with similar wall thickness differences by calculating the similarity between the wall thickness difference values ​​of each area and its neighboring areas. First, the wall thickness difference data of each area is standardized, and then the similarity between the areas is calculated by Euclidean distance, and divided into multiple structural areas according to the similarity. The divided areas are marked and stored as "structural morphology wall thickness area division data". The output data of this step is used to generate a heat map through a spatial data visualization tool to show the division results of each similar area, showing the wall thickness change trend of the pressure vessel and the distribution of the structural difference area.

[0038] Preferably, step S2 comprises the following steps: Step S21: constructing a virtualized pressure vessel regional structure model based on the structural morphology and wall thickness regional division data based on the digital twin to obtain a virtualized pressure vessel regional structure model; Step S22: performing regional structural fluid load simulation on the virtualized pressure vessel regional structural model to obtain regional structural fluid load data; Step S23: Based on the structural form wall thickness difference distribution data and the structural form wall thickness regional division data, the regional structural fluid load data is subjected to a connection structure force distribution difference analysis to obtain the connection structure force distribution difference data; Step S24: Calculate the resonant buckling instability probability of the regional structure fluid load data according to the force distribution difference data of the connection structure to obtain the resonant buckling instability probability data.

[0039] As an example of the present invention, refer to Figure 2 As shown, in this example, step S2 includes: Step S21: constructing a virtualized pressure vessel regional structure model based on the structural morphology and wall thickness regional division data based on the digital twin to obtain a virtualized pressure vessel regional structure model; In an embodiment of the present invention, a virtualized regional structural model of a pressure vessel is constructed based on the "structural morphology and wall thickness regional division data". Digital twin technology is used to generate a high-precision virtual model of a pressure vessel to ensure that the model can truly reproduce the physical properties and dynamic behavior of the pressure vessel. First, the "structural morphology and wall thickness regional division data" is imported into a three-dimensional modeling environment, and the corresponding three-dimensional regional structural model is generated through multi-point mesh refinement technology. The model construction process uses the principles of geometric topology to ensure that the spatial characteristics of each region are accurately expressed, and the differences in wall thickness distribution in each region are clearly reflected. The model also integrates information such as the material properties, temperature distribution, and pressure gradient of the container to provide a comprehensive virtual representation.

[0040] Step S22: performing regional structural fluid load simulation on the virtualized pressure vessel regional structural model to obtain regional structural fluid load data; In an embodiment of the present invention, after the construction of the regional structural model of the virtualized pressure vessel is completed, a fluid load simulation is performed on the model. This process uses computational fluid dynamics (CFD) methods to simulate the flow behavior of the fluid inside the pressure vessel. First, set the physical property parameters of the fluid (such as density, viscosity, etc.) and the flow boundary conditions in the pressure vessel, including inlet pressure, outlet velocity, and fluid temperature changes. Then, the Navier-Stokes equations are used to numerically solve the fluid behavior to simulate the pressure distribution, velocity field, and shear stress of the fluid in each region. The simulation uses the finite volume method for spatial discretization to ensure that the dynamic changes of the fluid in the complex structure area are accurately captured. The final output "regional structural fluid load data" contains the fluid pressure and its distribution in each region.

[0041] Step S23: Based on the structural form wall thickness difference distribution data and the structural form wall thickness regional division data, the regional structural fluid load data is subjected to a connection structure force distribution difference analysis to obtain the connection structure force distribution difference data; In an embodiment of the present invention, in this step, the "structural morphology wall thickness difference distribution data" and the "structural morphology wall thickness area division data" are combined with the "regional structure fluid load data" to perform a difference analysis of the force distribution of the connection structure. First, through the superposition principle of the mechanical field, the fluid load on the regional structure is associated with the wall thickness difference of the structure, and the mechanical stress changes at each structural connection point are analyzed. Stress analysis technology (such as the von Mises stress criterion) is used to calculate the stress distribution of different wall thickness areas when subjected to stress, with special attention paid to the stress concentration area at the connection. In order to ensure the accuracy of the analysis results, the difference finite element method is used to further refine the force changes in these areas to generate "connection structure force distribution difference data", which reflects the force differences at each connection point and identifies the structural areas that produce fatigue damage or local stress overload.

[0042] Step S24: Calculate the resonant buckling instability probability of the regional structure fluid load data according to the force distribution difference data of the connection structure to obtain the resonant buckling instability probability data.

[0043] In an embodiment of the present invention, the calculation of the probability of resonant buckling instability is based on the "connection structure force distribution difference data" and the "regional structure fluid load data". First, through natural frequency analysis, the vibration mode of the pressure vessel under stress is identified, and the natural frequency of the structure under different stress states is calculated in combination with the elastic modulus of the material and the structural geometric characteristics. Subsequently, the vibration response of the pressure vessel under fluid load is compared with its natural frequency to determine the area where resonance occurs. Using Euler's buckling theory, the critical buckling load of these areas under the resonant state is further calculated to evaluate the risk of instability. Finally, using the extreme value distribution theory in probability theory, combined with the stress distribution and critical buckling load of each structural area, the probability of resonant buckling instability in each area is calculated to generate "resonant buckling instability probability data", which provides a basis for the safety evaluation of pressure vessels.

[0044] Preferably, step S23 includes the following steps: Step S231: performing geometric characteristic analysis of connection structures between different regions based on the structural form wall thickness difference distribution data and the structural form wall thickness region division data to obtain geometric characteristic data of the connection structure; Step S232: Calculating adjacent connection angles between different regions of the connection structure geometric feature data to obtain structure adjacent connection angle data; Step S233: calculating the connection surface slopes between different regions of the connection structure geometric feature data to obtain structure connection surface slope data; Step S234: performing fluid force dispersion analysis between different regions on the regional structural fluid load data according to the structural adjacent connection angle data and the structural connection surface slope data, to obtain regional structural force dispersion data; Step S235: performing connection structure force distribution difference analysis on the regional structure fluid load data according to the regional structure force dispersion data to obtain connection structure force distribution difference data.

[0045] In the embodiment of the present invention, by comparing the "structure morphology wall thickness difference distribution data" and the "structure morphology wall thickness area division data", the geometric characteristics of the connection structure in different areas are analyzed. First, according to the wall thickness change trend of each area, combined with the geometric characteristic analysis method, a three-dimensional geometric model of each area is established. By extracting the geometric information of the connection part region by region, special attention is paid to the connection boundary of the adjacent areas and the local geometric morphology of the connection part, such as arc connection, straight connection and transition section. The curvature analysis method of differential geometry is used to quantify the geometric characteristics at the connection point, and finally generate "connection structure geometric characteristic data", which contains the geometric difference information between different areas and provides a basis for subsequent mechanical analysis. Using the "connection structure geometric characteristic data" generated in the previous step, the connection structure angle of the adjacent areas is calculated by the geometric topology analysis method. The specific method is to select the connection surface of each pair of adjacent areas, and based on the difference in the normal vectors between these surfaces, the vector dot product formula is used to calculate the connection angle between the two areas. At the same time, considering the influence of the angle change on the stress concentration area, the multi-point fitting algorithm is used to accurately calculate the average angle and local angle difference at each connection point. The calculation process includes point-by-point sampling of the geometric data of each region to ensure the high precision and accuracy of the angle calculation results, and finally obtains the "structural adjacent connection angle data", which will be used for subsequent mechanical load analysis. Based on the "connection structure geometric feature data", the slope of the connection surface between each region is calculated by gradient analysis. First, the connection surface of each region is locally planarized to extract the spatial coordinate information of each point on the surface. Then, the local slope of each point on the connection surface is calculated using derivative and difference methods. To ensure the accuracy of the slope calculation, multiple points on each connection surface are sampled multiple times, and the weighted average method is used to integrate the local slope changes of each point. This calculation can not only quantify the degree of inclination of the connection surface of each region, but also capture the bending or mutation of the local area to generate "structural connection surface slope data". This data will help analyze the stress of the connection structure. Combining "structural adjacent connection angle data" and "structural connection surface slope data", the force dispersion analysis of the fluid load distribution in the regional structure is carried out. The analytical method is based on the principle of pressure dispersion in fluid mechanics. By analyzing the dispersion effect of the angle and slope on the fluid load at the connection, the stress conditions of each area are determined. The specific operation includes using the Laplace equation to reconstruct the fluid load in each area, inputting the angle and slope data as boundary conditions, and gradually calculating the pressure gradient changes and the dispersion path of the fluid load in each area. The final output "regional structural force dispersion data" reflects in detail the force dispersion law of the fluid in different connection areas, laying the foundation for further force analysis. Through the "regional structural force dispersion data", the "regional structural fluid load data" is deeply analyzed for the difference in force distribution of the connection structure. The finite element method in stress analysis is used to combine the fluid load distribution in each area with the force dispersion data to analyze the force differences of the fluid load in the connection structure.First, the stress concentration area at each connection point is analyzed in detail, and the stress distribution of the connection structure is calculated using the von Mises stress criterion. Secondly, based on the fluid load distribution and structural geometric characteristics, the strain gradient change in the stress concentration area is calculated. Finally, the generated "connection structure stress distribution difference data" will reflect the stress differences of each connection part under different load conditions, which will help evaluate the stability and safety of the structure.

[0046] Preferably, step S234 includes the following steps: Decomposing the connection surface normal component according to the structural adjacent connection angle data and the structural connection surface slope data to obtain the connection surface normal component data; Conducting interface force dispersion analysis on the connection surface normal component data to obtain interface force dispersion data; According to the connection surface normal component data and the interface force dispersion data, the fluid shear force orientation difference between different regions is calculated for the regional structural fluid load data to obtain the fluid shear force orientation difference data; Based on the fluid shear force azimuth difference data and the connection surface normal component data, the fluid force dispersion between different regions is analyzed to obtain the regional structural force dispersion data.

[0047] In an embodiment of the present invention, based on the "structural adjacent connection angle data" and the "structural connection surface slope data", the vector decomposition method of analytical geometry is used to calculate the normal component of each connection surface. The specific operation includes first extracting the normal vector of the connection surface and decomposing it into components parallel to each main axis. In the decomposition process, the normal component of each connection surface in three-dimensional space is calculated by the angle cosine method. A large number of connection surfaces are processed one by one by matrix operations to ensure the comprehensiveness and accuracy of the calculation, especially in the surface connection area, more sampling points are added to improve the accuracy. The "connection surface normal component data" finally generated is used to describe the normal distribution characteristics of each connection surface between different regions, ensuring that in the subsequent mechanical analysis, the direction and magnitude of the force can accurately reflect the actual situation of the regional structure. Based on the "connection surface normal component data", the mechanical dispersion method in the finite element analysis (FEM) is used to analyze the stress condition of the interface force. First, for each connection surface, different external force loads are applied according to the data of its normal component. Through the step-by-step loading method, the effect of fluid pressure on each connection surface is simulated, focusing on the distribution characteristics of the normal force. In this process, the stress-strain relationship law is used, combined with the elastic modulus and structural morphology of the material, to analyze the distribution path and stress concentration area of ​​the normal force on the connection surface. After multiple rounds of calculations, the "interface force dispersion data" is generated, which accurately describes the diffusion and dispersion of the fluid force on each connection surface, and can further reveal the force law of the connection surface. The shear force under the action of the fluid load is calculated through the "connection surface normal component data" and "interface force dispersion data". The specific implementation method is to first calculate the direction and magnitude of the shear force based on the normal component of the connection surface, and evaluate the shear force distribution on each connection surface through the boundary layer theory of fluid mechanics. For the connection surfaces between different regions, the angle differential method is used to calculate the azimuth difference of the shear force, that is, the change in the direction of the shear force on adjacent connection surfaces. This method can effectively capture the shear force imbalance caused by the change in the inclination angle and slope of the connection surface between regions. The final generated "fluid shear force azimuth difference data" will be used to evaluate the force changes of the fluid in the pressure vessel structure, especially the stress concentration phenomenon in the high shear force area. Combined with the "fluid shear force azimuth difference data" and the "connection surface normal component data", the fluid force dispersion in the regional structure is finally analyzed. Using the principle of force decomposition and synthesis, the fluid forces in different regions are vector superimposed and dispersed. By constructing a force network model, the fluid force is gradually dispersed to each connection surface, and the distribution of the fluid force in the regional structure is evaluated based on the difference in the direction of the shear force. Especially for complex geometric structures and adjacent connection areas, a multi-dimensional stress path analysis method is used to ensure the uniformity and continuity of the fluid force. The output of this step is the "regional structure force dispersion data", which describes the dispersion path of the fluid force in different structural areas, providing key data basis for structural safety assessment and early warning.

[0048] Preferably, step S24 comprises the following steps: Step S241: performing fluid resonance frequency simulation on the regional structure fluid load data according to the force distribution difference data of the connection structure to obtain regional fluid resonance frequency data; Step S242: Calculating the load section moment of inertia between different regional structures on the regional structural fluid load data based on the regional fluid resonance frequency data to obtain structural load section moment of inertia data; Step S243: performing inertial distortion integration on the structural load section inertia moment data to obtain section inertial distortion integration data; Step S244: performing structural tolerance limit simulation on the regional structural fluid load data according to the cross-section inertial distortion integral data to obtain regional structural tolerance limit data; Step S245: Calculate the resonant buckling instability probability of the regional structure fluid load data according to the cross-section inertial distortion integral data and the regional structure tolerance limit data to obtain the resonant buckling instability probability data.

[0049] In an embodiment of the present invention, the fluid resonance frequency is simulated based on the "connection structure force distribution difference data". The specific operation includes first combining the fluid load data with the structural force distribution, and determining the potential resonance area by analyzing the non-uniformity of the force distribution. The natural frequency calculation method in fluid dynamics is used to combine the morphological characteristics of the regional structure with the fluid load to establish a mathematical model of the resonance frequency. The Fourier transform is applied to perform frequency domain analysis on the fluid force fluctuation, identify the peak value of the resonance frequency, and refine the resonance frequency of each region through local resonance analysis, and finally generate "regional fluid resonance frequency data". This data will be used to further evaluate the dynamic response and stability of the structure. Based on the "regional fluid resonance frequency data", the section moment of inertia of the structural load in different regions is calculated. As an important parameter for measuring the torsion and bending resistance of the structure, the moment of inertia is first calculated in detail by the mechanical integration method to the stress field distribution caused by the fluid load on each section. The cross-sectional shape integral formula is used, combined with the structural geometric characteristics of different regions, to gradually calculate the moment of inertia of each section under the action of the fluid load. Considering the influence of resonant frequency on the moment of inertia, the dynamic moment equation is used to correct the load moment of inertia, and finally the "structure load section moment of inertia data" is obtained, which lays the foundation for the subsequent inertial distortion analysis. Based on the "structure load section moment of inertia data", the distortion integral method in mechanics is used to analyze the inertial distortion effect of different regions. In the specific operation, the structural mechanics integral method is used to first integrate the section moment of inertia of each region point by point, and calculate the stress distribution of each section under the distortion state. This process uses the Lagrange integral method to calculate the change path of the moment of inertia in sections to ensure that the inertial distortion effect of each region can be accurately captured and described. By introducing the energy conservation equation, the total effect of inertial distortion is analyzed, and finally the "section inertial distortion integral data" is obtained. This data provides a key mechanical basis for the structural tolerance assessment under fluid load. This step uses the "section inertial distortion integral data" to simulate the tolerance limit of the regional structure. In the specific implementation process, the limit analysis method is used to evaluate the tolerance of the structure in each region. By comparing the stress under the fluid load with the physical limit value of the structure, the critical stress points of different regions are calculated. The yield criterion in material mechanics (such as the Von Mises criterion) is used to match the distortion effect and load limit of each region, and the tolerance limit of each region is confirmed one by one. After multiple rounds of iterations and data correction, the "regional structural tolerance limit data" is finally generated, which defines the tolerance limit state of different regions under the action of complex fluids. Combined with the "section inertia distortion integral data" and the "regional structural tolerance limit data", the probability of resonant buckling instability under fluid load is calculated. Using the buckling instability theory, the stability of the stress state of each region is first analyzed, and the buckling instability risk caused by fluid resonance is calculated by introducing the probability distribution model of the random resonance effect.The process uses the Poisson distribution function to quantify the probability of instability, and further refines the instability risk of different areas through the critical load equation in the structural buckling theory. The final "resonance buckling instability probability data" is used to evaluate the buckling instability risk of pressure vessels under dynamic loads, ensuring the accuracy and comprehensiveness of structural safety evaluation.

[0050] Preferably, step S243 includes the following steps: Discretize the moment of inertia of different sections of the structural load section moment of inertia data to obtain the discretized data of the section moment of inertia; According to the discretized data of section inertia moment, nonlinear regression analysis of the structure load section inertia moment data of different sections is performed to obtain the nonlinear regression data of the moment of inertia; Performing piecewise interpolation processing on the nonlinear regression data of the moment of inertia to obtain the nonlinear interpolation data of the moment of inertia; The inertia distortion integral is performed on the nonlinear interpolation data of the moment of inertia using the Gauss-Legendre integration method to obtain the section inertia distortion integral data.

[0051] In the embodiment of the present invention, the moment of inertia data of the structural load section is first analyzed in sections, and the corresponding moment of inertia data is discretized based on the geometric and physical characteristics of each section. In order to ensure the accuracy of the discretization result, multiple groups of discrete points are selected to characterize the moment of inertia distribution of the section. Each discrete point is refined by the cross-sectional area, position and the influence of the fluid load on the structure. By adopting a discretization method, such as the trapezoidal method or the equidistant method, the discrete value of the moment of inertia of each section is determined one by one to ensure that all discrete points can fully describe the change of the moment of inertia of the section. The final "section moment of inertia discretization data" will provide data support for subsequent nonlinear regression analysis. Based on the discretized moment of inertia data, the nonlinear regression analysis method is used to fit the moment of inertia data of each section. First, a suitable nonlinear function form (such as a polynomial function or an exponential function) is selected to express the trend of the moment of inertia changing with the position of the section. Then, the least squares method is used to fit the moment of inertia discretization data of each section point, and the function parameters are adjusted iteratively step by step to achieve the best fitting effect. Through such nonlinear regression analysis, the changing law of the moment of inertia between different sections can be accurately captured, and finally the "nonlinear regression data of the moment of inertia" can be obtained, which reflects the nonlinear mechanical characteristics of each section under the action of fluid load. In order to further refine the change of the moment of inertia, the "nonlinear regression data of the moment of inertia" is interpolated using the segmented interpolation method. In the specific operation, the nonlinear regression curve of each section is first divided into several small sections to ensure that the interpolation function can accurately fit the actual data in each small section. The cubic spline interpolation method or the Lagrange interpolation method is used to perform local interpolation on each small section, so that the interpolated curve can smoothly connect each regression segment and accurately reflect the nonlinear change of the moment of inertia of each section. Through the precise processing of interpolation, the "nonlinear interpolation data of the moment of inertia" is generated, which can provide more accurate input for the next step of integral calculation. Finally, based on the "nonlinear interpolation data of the moment of inertia", the inertial distortion integral of the moment of inertia of different sections is performed by the Gauss-Legendre integral method. The Gauss-Legendre integral method is an efficient numerical integration method suitable for handling the integral problem of nonlinear functions. This method approximates the integral value of the moment of inertia by selecting multiple integral points (weight coefficients). First, the moment of inertia interpolation data is divided into several integral intervals, and then each interval is integrated using the Gauss-Legendre integral, and the integral results of each interval are accumulated to obtain the final inertial distortion effect value. Through precise integration processing, the "section inertial distortion integral data" is generated, which can be used to evaluate the distortion response of the structure under fluid load and its tolerance limit.

[0052] Preferably, step S244 includes the following steps: Extract the extreme value of section distortion resistance according to the section inertial distortion integral data to obtain the extreme value data of section distortion resistance; Based on the extreme value data of cross-section distortion resistance, the structural distortion evolution response analysis is performed on the regional structural fluid load data to obtain the structural distortion evolution response data; Perform structural fracture failure simulation on the structural distortion evolution response data to obtain structural fracture failure data; According to the structural fracture failure data and the structural distortion evolution response data, the structural tolerance limit of the regional structural fluid load data is simulated to obtain the regional structural tolerance limit data.

[0053] In the embodiment of the present invention, the extreme value extraction of the distortion resistance of each section under different fluid loads is performed based on the "section inertia distortion integral data". First, the extreme value extraction algorithm is used to identify the extreme value points in the section distortion integral data by using the judgment conditions based on the first-order derivative and the second-order derivative. The local maximum and minimum values ​​are determined by solving the derivative of the curve. This process involves using the linear interpolation method to improve the positioning accuracy of the extreme value, ensuring that the extracted distortion resistance extreme value can truly reflect the bearing capacity of the section. After rigorous analysis, the obtained "section distortion resistance extreme value data" can effectively describe the distortion resistance of the structure under the external fluid load. The "section distortion resistance extreme value data" is combined with the "regional structure fluid load data" to perform structural distortion evolution response analysis. Using the finite element analysis method, the regional structure is first divided into multiple small units, and the fluid load is weighted by applying the distortion resistance extreme value by establishing a mechanical model of each unit to simulate the distortion response under different conditions. In the simulation process, the dynamic analysis technology is used to gradually calculate the deformation and stress distribution of the structure under the action of the fluid, taking into account the change of the fluid load over time. Finally, the generated "structural distortion evolution response data" can fully reflect the dynamic response characteristics of the regional structure under the influence of fluid load. The "structural distortion evolution response data" is used to simulate the structural fracture failure. First, based on the fracture mechanics theory, the stress intensity factor of the structure is calculated to evaluate the possibility of structural fracture under the existing fluid load. The interaction model of stress field and strain field is used to convert the distortion evolution response data into stress distribution, and the stress concentration area is determined by the threshold method. Then, the fracture process is simulated by the stepwise increment method, and the failure behavior of the structure under different load conditions is observed to obtain the "structural fracture failure data". This data not only reflects the critical point of fracture, but also provides important information about the failure mode. The tolerance limit of the regional structure is simulated by combining the "structural fracture failure data" and "structural distortion evolution response data". By establishing a comprehensive mechanical model, combining the maximum bearing capacity and failure mode of the structure under fluid load, the tolerance limit of the structure is evaluated by the limit analysis method and the limit equilibrium method. First, by calculating the ultimate bearing capacity of the structure under different fluid pressures and combining the fracture failure mode, the tolerance limit is gradually derived. In this process, the stress and deformation state of the structure are visualized by graphical means, making the tolerance limit results more intuitive, and finally obtaining the "regional structural tolerance limit data". This data provides a scientific basis for the safety assessment of pressure vessels and helps to improve the effectiveness of safety warnings.

[0054] Preferably, step S3 comprises the following steps: Step S31: normalizing the resonant buckling instability probability data to obtain resonant buckling instability probability normalized data; Step S32: constructing a safety evaluation logic model for the normalized data of the resonant buckling instability probability based on the random forest algorithm to obtain a safety evaluation logic model for the pressure vessel.

[0055] In the embodiment of the present invention, firstly, "resonant buckling instability probability data" are collected, which represent the probability of buckling instability of pressure vessels under different working conditions. In order to improve the consistency and comparability of the data, the data is normalized. The specific method is to use the minimum-maximum normalization algorithm to scale the data to a range of 0 to 1. This process can ensure that data from different sources have equal weights, avoid the evaluation bias caused by the difference in the range of the original data, and the final "resonant buckling instability probability normalized data" lays the foundation for subsequent safety evaluation. Construct a "pressure vessel safety evaluation logic model" to more accurately evaluate the safety of pressure vessels. First, use the "resonant buckling instability probability normalized data" as input features to establish a random forest model. Random forest is an integrated learning algorithm that performs classification or regression by constructing multiple decision trees. The normalized data is divided into a training set and a test set, usually 70% of the data is used for training and 30% for testing. In the training stage, several decision trees are constructed. Each tree uses randomly selected features for node splitting to ensure the diversity and anti-overfitting ability of the model. The output of each tree is voted and its evaluation result on the safety of pressure vessels is comprehensively judged. The hyperparameters in the random forest, such as the number of trees, maximum depth and minimum number of sample splits, are adjusted through cross-validation to ensure the generalization ability of the model.

[0056] Preferably, step S32 includes the following steps: Step S321: performing a logic error test on the normalized data of the resonant buckling instability probability to obtain the resonant buckling instability logic error data; Step S322: performing error balance iteration on the resonant buckling instability probability normalized data according to the resonant buckling instability logic error data to obtain resonant buckling error balance data; Step S323: performing time-effect characteristic analysis on the resonant buckling error balance data to obtain error time-effect characteristic data; Step S324: constructing a safety evaluation logic model for the resonant buckling error balance data and the error time-effect characteristic data based on the random forest algorithm to obtain a safety evaluation logic model for the pressure vessel.

[0057] In the embodiment of the present invention, in this step, a logical error test is performed on the "normalized data of the probability of resonant buckling instability" to evaluate the reliability and validity of the data under different working conditions. The test is performed by comparing the actual buckling instability events with the predicted probability values. First, the number of buckling instability events that actually occurred is counted, and the relevant working condition data is recorded. Then, the normalized data is fitted using logistic regression analysis to obtain the predicted value of the model. By comparing the predicted value with the actual occurrence value, the logical error is calculated, and the error values ​​of all data are counted, and finally the "logical error data of resonant buckling instability" is obtained, which provides a basis for subsequent error balance and model optimization. First, the logical error is analyzed to determine its influence degree and law. The weighted average method is used to adjust the normalized data in combination with the size and direction of the error, and the difference between the updated data and the original data is evaluated in each iteration until the difference is less than the preset threshold, and the "resonant buckling error balance data" is obtained. This data provides corrected basic data for the subsequent analysis of the time-effect influence characteristics. The influence of the time-effect factor on the probability of instability in the "resonant buckling error balance data" is analyzed. First, time factors, such as operating time and service life, are defined and correlated with error balance data. A linear regression model is constructed by time series analysis to explore the relationship between time factors and errors. The significance of time factors is evaluated by variance analysis. The resulting "error time-dependent characteristic data" contains the influence of time factors on the probability of resonant buckling instability, providing the necessary features for model construction. The focus is on building a "pressure vessel safety evaluation logic model". Based on the "resonant buckling error balance data" and "error time-dependent characteristic data", the model is constructed using the random forest algorithm. First, the data set is divided into a training set and a test set with a ratio of 7:3 to ensure the generalization ability of the model. In the training stage, the random forest algorithm is trained by constructing multiple decision trees, and each tree uses random feature selection to enhance the robustness of the model. Subsequently, cross-validation is performed to optimize model parameters, such as the number of trees and the maximum depth of each tree. After the model training is completed, the accuracy and recall of the model are evaluated using the test set, and the performance of the model is evaluated by confusion matrix analysis. The final "pressure vessel safety evaluation logic model" will provide a scientific basis for real-time monitoring and early warning, and help ensure the safe operation of pressure vessels.

[0058] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0059] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A pressure vessel safety evaluation and early warning method based on digital twins, It is characterized in that The following steps are involved: Step S1: obtaining pressure vessel design data; performing wall thickness difference distribution calculation on the pressure vessel design data to obtain structural morphology wall thickness difference distribution data; According to the structural morphology and wall thickness difference distribution data, the structural similarity region division is performed to obtain the structural morphology and wall thickness region division data; Step S2: performing a stress distribution difference analysis of the connection structure based on the structural morphology wall thickness area division data to obtain stress distribution difference data of the connection structure; performing a resonance buckling instability probability calculation based on the stress distribution difference data of the connection structure to obtain resonance buckling instability probability data; Step S3: constructing a safety evaluation logic model for the resonant buckling instability probability data to obtain a safety evaluation logic model for the pressure vessel; Step S4: Send the pressure vessel safety assessment logic model to the data cloud platform to execute the pressure vessel safety assessment early warning method.

2. According to claim 1, the pressure vessel safety evaluation and early warning method based on digital twins, It is characterized in that Step S1 includes the following steps: Step S11: obtaining pressure vessel design data; Step S12: performing structural morphology analysis on the pressure vessel design data to obtain structural morphology data of the pressure vessel; Step S13: performing wall thickness difference distribution calculation on the pressure vessel structure morphology data to obtain structure morphology wall thickness difference distribution data; Step S14: dividing the pressure vessel structure data into structural similarity regions according to the structure-form wall thickness difference distribution data to obtain structure-form wall thickness region division data.

3. According to claim 1, the pressure vessel safety evaluation and early warning method based on digital twins, It is characterized in that Step S2 includes the following steps: Step S21: constructing a virtualized pressure vessel regional structure model based on the structural morphology and wall thickness regional division data based on the digital twin to obtain a virtualized pressure vessel regional structure model; Step S22: performing regional structural fluid load simulation on the virtualized pressure vessel regional structural model to obtain regional structural fluid load data; Step S23: Based on the structural form wall thickness difference distribution data and the structural form wall thickness regional division data, the regional structural fluid load data is subjected to a connection structure force distribution difference analysis to obtain the connection structure force distribution difference data; Step S24: Calculate the resonant buckling instability probability of the regional structure fluid load data according to the force distribution difference data of the connection structure to obtain the resonant buckling instability probability data.

4. According to claim 3, the pressure vessel safety evaluation and early warning method based on digital twins, It is characterized in that Step S23 includes the following steps: Step S231: performing geometric characteristic analysis of connection structures between different regions based on the structural form wall thickness difference distribution data and the structural form wall thickness region division data to obtain geometric characteristic data of the connection structure; Step S232: Calculating adjacent connection angles between different regions of the connection structure geometric feature data to obtain structure adjacent connection angle data; Step S233: calculating the connection surface slopes between different regions of the connection structure geometric feature data to obtain structure connection surface slope data; Step S234: performing fluid force dispersion analysis between different regions on the regional structural fluid load data according to the structural adjacent connection angle data and the structural connection surface slope data, to obtain regional structural force dispersion data; Step S235: performing connection structure force distribution difference analysis on the regional structure fluid load data according to the regional structure force dispersion data to obtain connection structure force distribution difference data.

5. According to claim 4, the pressure vessel safety evaluation and early warning method based on digital twins, It is characterized in that Step S234 includes the following steps: Decomposing the connection surface normal component according to the structural adjacent connection angle data and the structural connection surface slope data to obtain the connection surface normal component data; Conducting interface force dispersion analysis on the connection surface normal component data to obtain interface force dispersion data; According to the connection surface normal component data and the interface force dispersion data, the fluid shear force orientation difference between different regions is calculated for the regional structural fluid load data to obtain the fluid shear force orientation difference data; Based on the fluid shear force azimuth difference data and the connection surface normal component data, the fluid force dispersion between different regions is analyzed to obtain the regional structural force dispersion data.

6. According to claim 3, the pressure vessel safety evaluation and early warning method based on digital twins, It is characterized in that Step S24 includes the following steps: Step S241: performing fluid resonance frequency simulation on the regional structure fluid load data according to the force distribution difference data of the connection structure to obtain regional fluid resonance frequency data; Step S242: Calculating the load section moment of inertia between different regional structures on the regional structural fluid load data based on the regional fluid resonance frequency data to obtain structural load section moment of inertia data; Step S243: performing inertial distortion integration on the structural load section inertia moment data to obtain section inertial distortion integration data; Step S244: performing structural tolerance limit simulation on the regional structural fluid load data according to the cross-section inertial distortion integral data to obtain regional structural tolerance limit data; Step S245: Calculate the resonant buckling instability probability of the regional structure fluid load data according to the cross-section inertial distortion integral data and the regional structure tolerance limit data to obtain the resonant buckling instability probability data.

7. According to claim 6, the pressure vessel safety evaluation and early warning method based on digital twins, It is characterized in that Step S243 includes the following steps: Discretize the moment of inertia of different sections of the structural load section moment of inertia data to obtain the discretized data of the section moment of inertia; According to the discretized data of section inertia moment, nonlinear regression analysis of the structure load section inertia moment data of different sections is performed to obtain the nonlinear regression data of the moment of inertia; Performing piecewise interpolation processing on the nonlinear regression data of the moment of inertia to obtain the nonlinear interpolation data of the moment of inertia; The inertia distortion integral is performed on the nonlinear interpolation data of the moment of inertia using the Gauss-Legendre integration method to obtain the section inertia distortion integral data.

8. According to claim 6, the pressure vessel safety evaluation and early warning method based on digital twins, It is characterized in that Step S244 includes the following steps: Extract the extreme value of section distortion resistance according to the section inertial distortion integral data to obtain the extreme value data of section distortion resistance; Based on the extreme value data of cross-section distortion resistance, the structural distortion evolution response analysis is performed on the regional structural fluid load data to obtain the structural distortion evolution response data; Perform structural fracture failure simulation on the structural distortion evolution response data to obtain structural fracture failure data; According to the structural fracture failure data and the structural distortion evolution response data, the structural tolerance limit of the regional structural fluid load data is simulated to obtain the regional structural tolerance limit data.

9. The pressure vessel safety evaluation and early warning method based on digital twin according to claim 1, It is characterized in that Step S3 includes the following steps: Step S31: normalizing the resonant buckling instability probability data to obtain resonant buckling instability probability normalized data; Step S32: constructing a safety evaluation logic model for the normalized data of the resonant buckling instability probability based on the random forest algorithm to obtain a safety evaluation logic model for the pressure vessel.

10. The pressure vessel safety evaluation and early warning method based on digital twin according to claim 9, It is characterized in that Step S32 includes the following steps: Step S321: performing a logic error test on the normalized data of the resonant buckling instability probability to obtain the resonant buckling instability logic error data; Step S322: performing error balance iteration on the resonant buckling instability probability normalized data according to the resonant buckling instability logic error data to obtain resonant buckling error balance data; Step S323: performing time-effect characteristic analysis on the resonant buckling error balance data to obtain error time-effect characteristic data; Step S324: constructing a safety evaluation logic model for the resonant buckling error balance data and the error time-effect characteristic data based on the random forest algorithm to obtain a safety evaluation logic model for the pressure vessel.

Citation Information

Patent Citations

  • Method for assessing reliability of tall tower equipment in typhoon weather

    CN112784386A

  • Pressure vessel safety evaluation and risk early warning method based on digital twinning

    CN114065589A

  • Dam safety early warning and alarm eliminating method and system based on digital twinning

    CN114707227A

  • Component-based reduced order modeling method and system for industrial scale structural digital twinning

    CN114830152A

  • Construction and safety evaluation method of safety evaluation model for tunnel underneath pass construction

    CN115879199A

Cited By

  • Virtual reality inspection emergency method and system for safe operation of pressure vessel

    CN120745343A

  • Virtual reality inspection emergency method and system for safe operation of pressure vessel

    CN120745343B

  • Pressure vessel operation state real-time monitoring system based on edge calculation

    CN121959463A