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

Abstract

The application discloses a virtual reality inspection emergency method and system for safe operation of a pressure container, and particularly relates to the technical field of pressure containers; fatigue data in multiple time periods is collected and integrated into a VR system, a fault test environment is created, equipment material fatigue data is loaded, failure response of the equipment in a fatigue state is simulated, abnormal fluctuations of equipment data and emergency responses of operating personnel are recorded in real time, the accuracy of VR simulation is evaluated, and the simulation is divided into high-accuracy and low-accuracy simulation levels; for low-accuracy simulation, the abnormality degree is further analyzed, and the failure time and failure conditions of the equipment are dynamically adjusted, so that the simulation accuracy and the emergency response capability of the operating personnel are improved, the problem that existing VR simulation systems cannot fully consider extreme failure modes of the equipment is effectively solved, and it is ensured that the operating personnel can make more accurate and timely decisions when facing actual emergencies.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of pressure vessels, specifically to a virtual reality inspection emergency method and system for safe operation of pressure vessels. BACKGROUND

[0002] The virtual reality (VR) inspection emergency method for safe operation of pressure vessels is a method that uses virtual reality technology to simulate the actual operating environment of pressure vessels in order to improve the efficiency and safety of inspection and emergency handling. This method is mainly used for safety operation training of pressure vessels. By simulating the actual scene through VR, the operator can perform pressure vessel inspection and emergency drills in a virtual environment. In this virtual environment, users can experience a realistic operation feeling, thereby familiarizing themselves with the working principle, operation process and possible emergency situations of pressure vessels, laying a foundation for dealing with operational risks in the future.

[0003] Specifically, this method can simulate the performance of various pressure vessels under different working conditions, such as sudden problems under conditions of overpressure, leakage or structural damage. Through the VR system, the operator can learn how to identify and handle these problems, including detecting faults, taking emergency measures and protecting themselves. Compared with traditional paper training or pure theoretical knowledge, the VR inspection emergency method can enable operators to intuitively understand and handle dangerous situations in a safe environment, greatly improving their practical experience and emergency response capabilities and reducing the risk of making mistakes in real operating environments.

[0004] The prior art has the following disadvantages:

[0005] VR simulation of pressure vessel emergency situations is usually based on a normal state device model, without considering the possible physical failure modes of the device under extreme conditions. When the device is in an overpressure or material fatigue state, critical component breakage, control system failure, and emergency valve failure may occur. If these extreme physical failures are not accurately simulated in the VR system, the operator may not be able to realize the severity of the device failure in a real emergency and may still attempt to operate according to the steps in the VR, resulting in a missed opportunity to escape or respond to the emergency. At the same time, if the VR system cannot simulate the physical failure mode of the device, the operator may become accustomed to ideal state emergency measures such as opening the emergency valve or starting the control system during virtual training. The actual device may fail under conditions of overpressure or material fatigue, causing the emergency valve to jam or the control system to lose power. The operator may attempt to repeatedly operate the ineffective emergency valve or system in a real emergency due to reliance on the virtual training steps, wasting valuable escape time and even exacerbating the spread of danger. Misleading decisions can prevent operators from evacuating or properly handling risks in a timely manner, which may lead to secondary explosions or large-scale leaks of the device, posing a serious threat to life safety. SUMMARY

[0006] The application aims to provide a virtual reality inspection emergency method and system for safe operation of pressure vessels, to solve the problems in the background art.

[0007] To achieve the above-mentioned purpose, the application provides the following technical solution: a virtual reality inspection emergency method for safe operation of pressure vessels, comprising the following steps:

[0008] S1: According to the physical properties of the equipment material and the use conditions of the pressure vessel, the equipment material fatigue data in several time periods is obtained respectively, and the obtained equipment material fatigue data is integrated into the VR system;

[0009] S2: A fault test environment is created in the VR system, the equipment material fatigue data in the VR system is dynamically loaded to simulate the real failure response, and the abnormal fluctuation of the equipment material fatigue data and the emergency response ability of the operator during the simulation process are recorded;

[0010] S3: The abnormal fluctuation of the equipment material fatigue data and the emergency response ability of the operator during the simulation process are comprehensively analyzed to evaluate the accuracy of the VR system simulation equipment under the physical failure mode;

[0011] S4: According to the evaluation result, the accuracy of the VR system simulation equipment under the physical failure mode is divided into different levels, which is divided into high accuracy simulation and low accuracy simulation, and corresponding processing is carried out;

[0012] S5: For low accuracy simulation, the abnormal degree of the accuracy of the VR system simulation equipment under the physical failure mode is predicted and analyzed, if the abnormal degree is high, the failure time and failure condition of the equipment are dynamically adjusted to improve the emergency simulation precision.

[0013] Preferably, in S2, the fatigue crack propagation rate abnormality index is generated after analyzing the abnormal increase of the fatigue crack propagation rate, and the fatigue crack propagation rate abnormality index is obtained by:

[0014] Collect and organize crack propagation rate data X = [x1, x2,..., xn], calculate quartiles according to the data set: Q1: Sort the data and find the 25% data point, Q3: Sort the data and find the 75% data point, Q2: Sort the data and find the middle data point, i.e. 50% position, quartile range IQR: According to the rules of box plot, calculate the lower and upper boundaries of outliers: Lower Bound: Upper Bound: According to the calculated upper and lower bounds, identify outliers: any crack propagation rate xi is considered as an outlier if or The degree of deviation between each outlier and Q1 or Q3 is calculated, and then the fatigue crack propagation rate anomaly index is calculated, expressed as: ; wherein: The i-th outlier, The total number of outliers, n is the total number of all data points, The degree of deviation of each data point from the median, HK is the fatigue crack propagation rate anomaly index.

[0015] Preferably, in S2, the error operation frequency change drift index is generated after analyzing the change of the error operation frequency of the operator in the face of abnormal situations, and the method for obtaining the error operation frequency change drift index is:

[0016] Collect the error operation frequency data of the operator and arrange them in chronological order, and set the error operation frequency time series data as , where t represents the time point, and the order of the ARIMA model is determined, where: p is the autoregressive order, indicating the number of past values included in the model, d is the difference number, indicating the number of differences required for data stabilization, q is the moving average order, indicating the number of past values of error terms, and the ARIMA model is fitted, expressed as: ; wherein, α is the constant term, is the parameter of the autoregressive term, is the parameter of the moving average term, is the white noise error term, and the estimated value of the parameter after fitting the model is ; the fitted value of the model is calculated using the fitted ARIMA model , expressed as: ; the residual is the difference between the actual data and the fitted data, indicating the prediction error of the model: ; the operation frequency change drift index is calculated, and the calculation expression is: ; in the formula, GH is the error operation frequency change drift index, is the j-th value in the residual sequence, is the average value of the residual.

[0017] Preferably, in S3, the fatigue crack propagation rate anomaly index and the error operation frequency change drift index are converted into a comprehensive feature vector, the comprehensive feature vector is taken as the input of a machine learning model, the machine learning model takes the accuracy value label of the VR system simulation device in the physical failure mode as the prediction target, the training target is to minimize the sum of the prediction errors of the accuracy value labels of all VR system simulation devices in the physical failure mode, the machine learning model is trained until the sum of the prediction errors converges, and the model training is stopped, and the accuracy value of the VR system simulation device in the physical failure mode is determined according to the model output result, wherein the machine learning model is a polynomial regression model.

[0018] Preferably, in S4, according to the evaluation result, the accuracy of the VR system simulation device in the physical failure mode is divided into different levels, specifically:

[0019] The accuracy value of the VR system simulation device in the physical failure mode is compared with the accuracy value reference threshold set in the historical data, if the accuracy value of the VR system simulation device in the physical failure mode is greater than or equal to the accuracy value reference threshold set in the historical data, it means that the accuracy of the VR system simulation device in the physical failure mode is high, at this time, a high-accuracy simulation signal is generated, and it is divided into high-accuracy simulation; if the accuracy value of the VR system simulation device in the physical failure mode is less than the accuracy value reference threshold set in the historical data, it means that the accuracy of the VR system simulation device in the physical failure mode is low, at this time, a low-accuracy simulation signal is generated, and it is divided into low-accuracy simulation.

[0020] Preferably, in S5, for low-accuracy simulation, the abnormal degree of the accuracy of the VR system simulation device in the physical failure mode is predicted and analyzed, if the abnormal degree is high, the failure time and failure condition of the device are dynamically adjusted to improve the emergency simulation accuracy, specifically:

[0021] The accuracy abnormal degree is calculated by comparing the difference between the actual failure response of the VR simulation device and the theoretical failure response ; Set is the response value of the actual device in the failure mode, is the expected device failure response value, and the calculation expression of the accuracy abnormal degree is: ; If the accuracy abnormal degree is greater than the preset accuracy abnormal degree reference threshold, it means that the accuracy abnormal degree is high, that is, the failure time and failure condition in the simulation deviate from the true situation, so the failure time and the failure condition need to be dynamically adjusted.

[0022] Preferably, is a dynamically adjusted failure time, is a dynamically adjusted failure condition, the failure time and the failure condition are adjusted based on the abnormality degree, and the formula for dynamic adjustment is: ; wherein, is a theoretical failure time, is a theoretical failure condition, , is an adjustment factor, indicating the influence degree of the abnormality degree on the failure time and the failure condition.

[0023] The present application also provides a virtual reality inspection emergency system for safe operation of a pressure vessel, comprising a data acquisition module, a test module, a comprehensive analysis module, a grade division module and a dynamic adjustment module.

[0024] The data acquisition module: according to the physical properties of the equipment material and the use conditions of the pressure vessel, the equipment material fatigue data in several time periods are obtained respectively, and the obtained equipment material fatigue data are integrated into the VR system;

[0025] The test module: a fault test environment is created in the VR system, the equipment material fatigue data in the VR system are dynamically loaded to simulate the real failure response, and the abnormal fluctuation of the equipment material fatigue data and the emergency response ability of the operator during the simulation process are recorded;

[0026] The comprehensive analysis module: the abnormal fluctuation of the equipment material fatigue data and the emergency response ability of the operator during the simulation process are comprehensively analyzed to evaluate the accuracy of the VR system simulation equipment under the physical failure mode;

[0027] The grade division module: according to the evaluation results, the accuracy of the VR system simulation equipment under the physical failure mode is divided into different grades, which is divided into high accuracy simulation and low accuracy simulation, and corresponding processing is carried out;

[0028] The dynamic adjustment module: for low accuracy simulation, the abnormality degree of the accuracy of the VR system simulation equipment under the physical failure mode is predicted and analyzed, if the abnormality degree is high, the failure time and the failure condition of the equipment are dynamically adjusted to improve the emergency simulation precision.

[0029] In the above technical solution, the present application provides technical effects and advantages:

[0030] ​1、The present application provides an innovative emergency training method by integrating device material fatigue data and dynamic simulation of pressure vessels under extreme failure modes in a virtual reality (VR) system, effectively overcoming the lack of extreme physical failure modes in VR simulations in the prior art. By obtaining fatigue data based on the physical properties of the device and the operating conditions of the pressure vessel, and combining real-time fault data and the emergency response capabilities of the operators, the system can accurately simulate severe failure conditions of the device under overpressure or fatigue. The method evaluates and predicts the accuracy of the device failure simulation by calculating the fatigue crack growth rate anomaly index and the error operation frequency variation drift index, and adjusts the device failure mode in the virtual environment according to the evaluation results to improve the simulation accuracy.

[0031] 2、The present application not only improves the emergency response capability of VR simulated devices, but also significantly reduces the possibility of operator error caused by the inconsistency between virtual training and real device behavior. In the case of low accuracy simulation, the system will analyze the abnormality degree and dynamically adjust the device failure time and conditions to ensure high precision of VR simulation training, ensuring that operators can make timely and accurate decisions in real emergency situations, avoiding wasting valuable time or exacerbating the spread of accidents due to reliance on false assumptions in virtual training. Through these technical innovations, the present application effectively improves the operator's ability to identify device failure modes and emergency response efficiency, enhances the operator's safety response capability when facing complex emergency events, and significantly improves the safety and accident prevention level of pressure vessels. BRIEF DESCRIPTION OF DRAWINGS

[0032] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed in the embodiments will be briefly introduced as follows. Obviously, the drawings described below are only some embodiments described in the present application, and other drawings can be obtained by those skilled in the art based on these drawings.

[0033] Figure 1 The method flowchart of the present application.

[0034] Figure 2 The system module diagram of the present application. DETAILED DESCRIPTION

[0035] In order to make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0036] Embodiment 1, please refer to Figure 1 and Figure 2 The virtual reality inspection emergency method for pressure vessel safety operation in the embodiment includes the following steps:

[0037] S1: According to the physical properties of the equipment material and the use conditions of the pressure vessel, obtain the equipment material fatigue data in several time periods respectively, and integrate the obtained equipment material fatigue data into the VR system;

[0038] S2: Create a failure test environment in the VR system, simulate the real failure response by dynamically loading the equipment material fatigue data in the VR system, and record the abnormal fluctuation of the equipment material fatigue data and the emergency response ability of the operator during the simulation process;

[0039] S3: Comprehensive analysis of the abnormal fluctuation of the equipment material fatigue data and the emergency response ability of the operator during the simulation process, evaluate the accuracy of the VR system simulation equipment under the physical failure mode;

[0040] S4: According to the evaluation results, the accuracy of the VR system simulation equipment under the physical failure mode is divided into different levels, which is divided into high accuracy simulation and low accuracy simulation, and the corresponding processing is carried out;

[0041] S5: For low accuracy simulation, the abnormal degree of the accuracy of the VR system simulation equipment under the physical failure mode is predicted and analyzed, if the abnormal degree is high, the failure time and failure condition of the equipment are dynamically adjusted to improve the emergency simulation precision.

[0042] In S1, according to the physical properties of the equipment material and the use conditions of the pressure vessel, the equipment material fatigue data in several time periods is obtained respectively, and the obtained equipment material fatigue data is integrated into the VR system, which is specifically:

[0043] First of all, it is necessary to understand the composition and material properties of the pressure vessel, and these physical properties are crucial for predicting the fatigue state of the equipment. Common equipment materials may include carbon steel, stainless steel, aluminum alloy, etc., and the fatigue behavior of these materials will be different under different use conditions. Material mechanical properties: such as yield strength, tensile strength, fatigue limit, elastic modulus, etc. Fatigue life: the fatigue life curve of the material, which is usually drawn according to the fatigue life data of the material under different stress levels. Fatigue crack propagation rate: how cracks propagate and the rate of propagation during the fatigue process of the material, which affects the safe service life of the material. Stress-strain relationship: understand the deformation behavior of the material under different stress states.

[0044] The service conditions of a device can affect its material fatigue state. Common service conditions include: operating pressure: the maximum operating pressure the device is subjected to in actual operation, including static and dynamic pressure. Temperature fluctuations: the impact of high and low temperature environments on material fatigue. The expansion and contraction of materials under temperature fluctuations can cause fatigue damage. Vibration and impact loads: pressure vessels may be subjected to periodic vibration and impact loads during transportation, operation, or external impact forces, which can also exacerbate material fatigue damage. Corrosion and environmental factors: corrosive environments (such as acidic, alkaline gases, seawater) can accelerate the fatigue process of materials. By combining the physical properties of materials and the service conditions of the device, we can obtain relevant fatigue data, such as the fatigue life of each material under different operating pressures, temperatures, and time periods.

[0045] Obtaining fatigue data for device materials typically involves the following aspects:

[0046] By conducting actual fatigue experiments on sample materials, we can obtain the fatigue life curve and related data of the materials. These experiments are usually conducted in controlled environments, including high and low temperature tests, corrosion environment tests, etc., to obtain fatigue life under different loads and stresses. Fatigue testing: using standard fatigue testing equipment (such as fatigue testing machines) to load the material samples multiple times, recording the damage of the material at different cycle times. Crack propagation testing: monitoring the crack propagation rate of the material through crack propagation testing (such as through CT scanning or acoustic emission technology).

[0047] Using theoretical models and computational methods of materials science to predict the fatigue behavior of materials under different service conditions. Common models include: S-N curve (stress-life curve): through experiments on the stress range of materials, the fatigue life prediction of materials is obtained. Miner's fatigue accumulation method: according to the weighted sum of different cyclic stresses, the fatigue damage of materials under complex loads is predicted. Models based on fracture mechanics: predict the propagation of fatigue cracks under actual working conditions through crack propagation rate. These data can be obtained through material databases, engineering standards, or through finite element analysis (FEA) simulation calculations.

[0048] Using historical data or on-site monitoring data of pressure vessels in actual use, such as the running time of the device, the actual working pressure it bears, temperature changes, etc., to assess the fatigue state of the device material. These data usually come from: online monitoring systems: real-time monitoring of the working state of the device through the installation of sensors (such as pressure sensors, temperature sensors). Maintenance and inspection records: through regular maintenance and inspection records, the fatigue condition and possible damage signs of the device are obtained.

[0049] By machine learning algorithms and big data analysis, based on the historical operation data, failure data and material properties of the equipment, a fatigue prediction model is established. By analyzing past data to predict the fatigue life of equipment materials in the future.

[0050] The obtained equipment material fatigue data is formatted and classified according to different time periods, working conditions and material properties. For example: time period data: according to the cycle of equipment operation (such as year, month, week), assign corresponding values to the fatigue data of each cycle. Condition variables: including different pressures, temperatures, loads, etc., as input variables to generate material fatigue life data. Failure mode label: in the simulation process, mark the possible failure modes of the equipment under different conditions, such as valve jamming, material crack propagation, etc.

[0051] Develop a data interface with the VR system to connect material fatigue data with the VR simulation engine, ensuring that these data can be dynamically loaded during simulation. This interface can be: database interface: store material fatigue data in the database and call through API to extract data in real time during simulation. Data flow interface: transmit equipment fatigue state data through real-time data flow technology (such as MQTT or WebSocket) to ensure that the VR system can update the fatigue state in real time.

[0052] During the VR simulation process, the system will dynamically load the corresponding fatigue data according to the real-time use state of the equipment (such as working pressure, temperature). The specific methods include: environmental condition adaptation: according to the current operating conditions of the equipment (such as the pressure vessel in the overpressure state), load the fatigue data matching the condition. Dynamic fatigue state update: when the equipment state changes (such as temperature rise or pressure increase), the system will update the fatigue state of the material in real time, affecting the performance of the virtual equipment.

[0053] Through the integration of fatigue data, the VR system can simulate the fatigue response of materials under specific conditions. For example: stress / fatigue accumulation simulation: simulate the stress accumulation of the equipment under working pressure, and gradually cause fatigue damage to the material over time. Fatigue crack propagation: when the equipment material appears fatigue damage, the system will simulate crack propagation according to fatigue data, leading to performance degradation or failure of the equipment. Failure mode presentation: when the material fatigue reaches a certain degree, failure events such as valve jamming, crack propagation to rupture, etc. occur.

[0054] In this application, by obtaining the physical properties, working conditions and fatigue data of equipment materials, and integrating these data into the VR system, the accuracy and realism of virtual reality training can be significantly improved. In VR training,

[0055] S2: Create a fault test environment in the VR system, load the device material fatigue data in the VR system dynamically to simulate the real failure response, and record the abnormal fluctuation of the device material fatigue data and the emergency response ability of the operator during the simulation process.

[0056] First, a virtual pressure vessel environment matching the actual device structure needs to be established in the VR system. This process includes: using CAD models or scanning technology to obtain the three-dimensional geometry of the actual device, and converting it into an interactive model in the VR environment. Set the material properties (such as density, elastic modulus, yield strength, fatigue limit, etc.) and physical behavior (such as thermal expansion, stress-strain response, etc.) for each device component (such as valves, pipes, pressure vessel shells, etc.).

[0057] To simulate failure response, multiple common failure modes need to be defined, including but not limited to: overpressure failure: simulate the failure mode that may occur in the pressure vessel under overpressure conditions, such as valve rupture, pipe leakage, container shell deformation, etc. Leakage failure: simulate the gas or liquid leakage caused by the sealing failure or joint loosening of the pressure vessel. Material fatigue failure: simulate the structural damage caused by material fatigue, stress concentration or crack propagation. Control system failure: simulate the failure of the automatic control system caused by sensor, actuator or power system failure. These failure modes not only change in time dimension, but also need to be triggered in real time according to environmental variables (such as temperature, pressure, vibration, etc.).

[0058] Operators can interact with the virtual environment through VR devices, including: handsets or controllers: operators use VR controllers for input, such as starting and stopping valves, adjusting pressure, etc. Operation prompts and feedback: provide real-time visual and audio feedback to help operators understand the current device status and emergency measures.

[0059] Integrate the fatigue data of the device (such as material fatigue life, crack propagation, stress-life relationship, etc.) into the VR system to ensure that the simulation reflects the actual material fatigue behavior. The specific methods include: fatigue data input interface: through the interface or API, input the fatigue data obtained from laboratory tests, field monitoring and historical data into the VR system. These data should include time period, working conditions (pressure, temperature, etc.) and corresponding fatigue damage model. Data update mechanism: according to the current operating state of the device (such as working pressure, temperature, load, etc.), dynamically load the relevant fatigue data and update the fatigue state of the material in real time. This process can use real-time databases or cloud computing for data transmission.

[0060] Simulate the fatigue response of materials in a VR environment based on loaded fatigue data. For example: Fatigue crack propagation: Simulate how materials accumulate damage under load over time, eventually leading to crack propagation. As the crack propagates, the system simulates the failure response of the device, such as pressure leakage or structural rupture. Fatigue life curve application: Simulate the remaining life of the device under specific conditions and fatigue accumulation, predicting its performance in the virtual environment.

[0061] During simulation, monitor the fatigue state of the device in real time and generate real-time monitoring results based on fatigue data, such as: Stress concentration: Simulate the stress concentration that may occur in high pressure areas of the device, affecting the fatigue damage process. Strain feedback: Monitor the strain of the device, especially the strain changes in areas with severe fatigue damage, and feed back to the operator. Fatigue data fluctuation: As the device is used over time, changes in material fatigue data will cause changes in virtual device behavior, and the system will adjust the simulation results in real time based on these fluctuations.

[0062] To ensure the authenticity of the simulation, it is necessary to monitor and record the fluctuation of the fatigue data of the device material. Common abnormal fluctuations include: Abnormally high fatigue crack propagation rate: If the fatigue crack propagation rate of a certain part is much higher than expected, it indicates that there is a structural weakness that needs to be recorded and reported. Inconsistent stress-strain response: If the stress or strain values of some components deviate from the normal range under the same working conditions, it may indicate that fatigue damage is experiencing abnormal fluctuations. Sudden change in failure mode: If the physical failure mode of the device (such as from leakage to explosion, valve jam to fracture) experiences significant fluctuations, the abnormal situation needs to be recorded.

[0063] After analyzing the case of abnormally high fatigue crack propagation rate, generate the fatigue crack propagation rate anomaly index, and the method for obtaining the fatigue crack propagation rate anomaly index is:

[0064] Collect and organize crack propagation rate data X = [x1, x2,..., xn], which should be time series data or experimental data. Calculate the quartiles based on the data set: Q1 (first quartile): Sort the data and find the 25th percentile. Q3 (third quartile): Sort the data and find the 75th percentile. Q2 (median): Sort the data and find the middle data point (50% position). IQR (interquartile range): According to the rules of the box plot, calculate the lower and upper boundaries of the outliers: Lower Bound: Upper Bound: According to the calculated lower and upper boundaries, identify outliers: Any crack propagation rate xi if or , then it is considered an outlier. The outlier index can reflect the degree of outliers by measuring the deviation between each outlier and Q1 or Q3, and then calculating the fatigue crack growth rate outlier index, expressed as: ; wherein: The i-th outlier, is the total number of outliers, n is the total number of all data points, is the deviation of each data point from the median, and HK is the fatigue crack growth rate outlier index.

[0065] The size of the fatigue crack growth rate outlier index directly reflects the simulation accuracy of the physical failure mode of the device under extreme conditions. The larger the outlier index, the more serious the abnormal increase in crack growth rate, which usually means that the VR system fails to accurately reflect the real physical behavior in simulating the failure process of the device, and may ignore the more subtle or more complex failure mechanisms in the crack propagation process. In this case, the device in the VR simulation may be too idealized and fail to exhibit the weaknesses or failure modes of the real device under extreme conditions, leading to a possible delay in the operator's response to the serious failure of the device and reliance on incorrect emergency steps. This misleading can affect the timeliness and correctness of the emergency response, thereby increasing the potential danger.

[0066] On the contrary, the smaller the outlier index, the more consistent the change in crack growth rate with the actual physical failure mode during simulation, and the VR system can more accurately reflect the failure characteristics of the device under extreme conditions such as overpressure and fatigue. A smaller outlier index means that the VR system has effectively simulated the key failure processes of the device when it faces material fatigue or overload, such as the acceleration of crack propagation speed, potential failure of emergency valves or control systems, etc., and the operator can make more accurate judgments and emergency responses based on this simulation data. In summary, the smaller the outlier index, the more consistent the VR system's simulation with the actual failure mode of the device, thereby improving the accuracy and safety of the operator's emergency handling.

[0067] By recording the operator's reaction process during the simulation of the failure, the operator's emergency response ability can be evaluated. Important indicators include: decision-making time: the time taken by the operator to identify the failure and make a decision. Reaction speed: the speed at which the operator takes emergency measures during the simulation, such as closing valves, starting emergency systems, etc. Error decision record: the operator's incorrect operation (such as repeatedly starting the failed emergency valve, incorrect use of tools, etc.) in the face of abnormal situations will be recorded by the system to provide a reference for subsequent training.

[0068] After analyzing the frequency changes of the operator's incorrect operations in the face of abnormal situations, the error operation frequency change drift index is generated, and the error operation frequency change drift index is obtained as follows:

[0069] Collecting the error operation frequency data of the operation personnel and arranging them in time sequence, setting the error operation frequency time series data as where t represents the time point, determining the order of the ARIMA model, where: p is the autoregressive order, indicating the number of past values included in the model, d is the difference number, indicating the number of differences required for data stationarity, q is the moving average order, indicating the number of past values of error terms. These parameters are usually determined by ACF (autocorrelation function) and PACF (partial autocorrelation function) selection. The common method is: determine q (moving average order) by ACF graph, observe the first significant peak in the graph. Determine p (autoregressive order) by PACF graph, observe the first significant peak in the graph. d is determined by difference operation, usually confirmed by unit root test. By selecting the (p, d, q) parameters, use least squares or maximum likelihood estimation method to fit the ARIMA model, the expression is: ; where α is the constant term, is the parameter of the autoregressive term, is the parameter of the moving average term, is the white noise error term, indicating the part that the model fails to explain, and the estimated value of the parameter after fitting the model ; use the fitted ARIMA model to calculate the fitted value of the model , the expression is: ; the residual is the difference between the actual data and the fitted data, indicating the prediction error of the model: ; by calculating the residual, analyze the fitting effect of the model, ensure that the model reasonably captures the change of error operation frequency, the calculation of error operation frequency change drift index is through the analysis of the change of model residual, to reflect the change and drift of error operation frequency, the calculation expression is: ; where GH is the error operation frequency change drift index, is the jth value in the residual sequence, is the average value of the residual.

[0070] The greater the error operation frequency change drift index, the worse the accuracy of the VR system simulating the device in the physical failure mode. When the drift index is high, it means that the frequency of error operations changes greatly when the operator faces extreme situations in virtual reality training, reflecting the operator's inability to effectively identify and respond to the failure mode of the virtual device. This large drift usually indicates that the simulation environment fails to fully reflect the complex reactions of the device in the physical failure state, such as nonlinear failure of the device under conditions of overpressure or material fatigue, which can lead the operator to rely on ideal emergency steps in virtual training, but fail to respond correctly in actual emergencies. This not only affects the training effect, but also may increase the operator's error rate in real crises, even exacerbating the danger.

[0071] The smaller the error operation frequency change drift index, the higher the accuracy of the VR system simulating the device in the physical failure mode. When the drift index is low, it means that the operator can adapt and correctly respond to emergency handling steps of the device under abnormal conditions such as overpressure, leakage or material fatigue after training in the virtual environment. This indicates that the virtual reality system successfully simulates the failure mode of the device under extreme conditions, and the training process effectively improves the operator's ability to identify device failures and abnormal states. The operator's emergency response in the virtual environment is accurate and does not rely on ineffective emergency steps, which improves their emergency handling ability in real crises, effectively reducing safety hazards and operational errors.

[0072] S3: Comprehensive analysis of abnormal fluctuations in material fatigue data of the device during simulation and emergency response capabilities of the operator to evaluate the accuracy of the VR system simulating the device in the physical failure mode.

[0073] The fatigue crack propagation rate anomaly index and the error operation frequency change drift index are converted into a comprehensive feature vector, the comprehensive feature vector is taken as the input of the machine learning model, the machine learning model takes the accuracy value label of the VR system simulating the device in the physical failure mode as the prediction target, and the training target is to minimize the sum of prediction errors of the accuracy value labels of all VR systems simulating the device in the physical failure mode, the machine learning model is trained until the sum of prediction errors converges, and the model training is stopped, and the accuracy value of the VR system simulating the device in the physical failure mode is determined according to the model output result, wherein the machine learning model is a polynomial regression model.

[0074] The method for obtaining the accuracy value of the VR system simulating the device in the physical failure mode is: obtaining the corresponding function expression from the comprehensive feature vector training data of the trained machine learning model: ; in the formula, is the output function of the model, HK is the fatigue crack growth rate anomaly index, GH is the error operation frequency variation drift index, is the accuracy value of the VR system simulation device in the physical failure mode.

[0075] S4: According to the evaluation result, the accuracy of the VR system simulation device in the physical failure mode is divided into different levels, which is divided into high accuracy simulation and low accuracy simulation, and corresponding processing is carried out.

[0076] The accuracy value of the VR system simulation device in the physical failure mode obtained is compared with the accuracy value reference threshold value set in the historical data. If the accuracy value of the VR system simulation device in the physical failure mode is greater than or equal to the accuracy value reference threshold value set in the historical data, it means that the accuracy of the VR system simulation device in the physical failure mode is high, at this time, a high accuracy simulation signal is generated, and it is divided into high accuracy simulation. If the accuracy value of the VR system simulation device in the physical failure mode is less than the accuracy value reference threshold value set in the historical data, it means that the accuracy of the VR system simulation device in the physical failure mode is low, at this time, a low accuracy simulation signal is generated, and it is divided into low accuracy simulation.

[0077] When the accuracy value of the VR system simulation device is higher than or equal to the set reference threshold value, it means that the simulation system performs well in the physical failure mode. If the evaluation result shows that the accuracy of the simulation system meets the expected standard, the current model can be used for training and emergency drill to ensure that the operator obtains accurate failure response in the training process. The results obtained in the training are fed back to the system, which further optimizes the training model and data set, improves the precision and response ability of the simulation. The system will generate a high accuracy simulation signal, which is used to indicate that this simulation environment meets the requirements of safety training and can be used in daily training, device operation procedure development and other scenes.

[0078] When the accuracy value of the VR system simulation device is lower than the set reference threshold value, it means that the simulation system may not be able to fully simulate the behavior of the device in the physical failure state, which may lead to misleading reactions of the operator. The model needs to be optimized, the training data needs to be adjusted, and the precision of the simulation needs to be improved. It may be necessary to strengthen the simulation of the failure mode of the device under extreme conditions, add more physical failure scenes, failure modes, etc., to ensure the accuracy of the training. At this time, the system will generate a low accuracy simulation signal, indicating that the simulation does not meet the requirements of safety training and needs to be further adjusted and optimized.

[0079] S5: For low accuracy simulation, the abnormal degree of the accuracy of the VR system simulation device in the physical failure mode is predicted and analyzed. If the abnormal degree is high, the failure time and failure condition of the device are dynamically adjusted to improve the accuracy of the emergency simulation.

[0080] The accuracy abnormality degree is calculated by comparing the difference between the actual failure response of the VR simulation device and the theoretical failure response is the response value of the actual device in the failure mode, is the expected device failure response value, and the calculation expression of the accuracy abnormality degree is: If the accuracy abnormality degree is greater than the preset accuracy abnormality degree reference threshold, it means that the accuracy abnormality degree is high, that is, the failure time and failure condition in the simulation deviate from the true situation, so it is necessary to dynamically adjust the failure time and the failure condition.

[0081] is the dynamically adjusted failure time, is the dynamically adjusted failure condition, and the formula for dynamically adjusting the failure time and the failure condition based on the abnormality degree is: ; wherein, is the theoretical failure time, is the theoretical failure condition, , is an adjustment factor, indicating the influence degree of the abnormality degree on the failure time and the failure condition.

[0082] After adjusting the failure time and the failure condition, the simulation will become more accurate. The new failure time and the failure condition will make the failure response of the device closer to the behavior of the actual device under extreme conditions, thereby improving the accuracy of the emergency simulation. The simulation result at this time will be closer to the response of the real device: according to the adjusted simulation result, the accuracy of the VR system simulation device can be re-evaluated. If the new simulation result still has a significant gap with the theoretical response, the adjustment will continue until the accuracy of the simulation reaches the predetermined standard.

[0083] ​​​​In this embodiment, the safety operation of a pressure vessel is simulated for emergency training through a virtual reality (VR) system to ensure that the operator can respond accurately in real failure conditions. First, fatigue data in different time periods is obtained according to the physical properties of the equipment material and the use conditions, and these data are integrated into the VR system for simulation. Then, a failure test environment is created in the VR system, the failure response of the equipment is simulated by dynamically loading the material fatigue data, and the abnormal fluctuations of the equipment data and the emergency response ability of the operator are recorded. Through comprehensive analysis of these data, the accuracy of the VR system in simulating the physical failure mode of the equipment is evaluated, and it is divided into high-accuracy and low-accuracy simulation. For low-accuracy simulation, the accuracy abnormality degree is predicted and analyzed, and if the abnormality degree is high, the failure time and failure conditions of the equipment are dynamically adjusted to improve the accuracy of the emergency simulation, make the virtual environment more realistic and credible, and ultimately enhance the emergency response ability of the operator to ensure safe operation.

[0084] In this embodiment, the safety operation of a pressure vessel is simulated for emergency training through a virtual reality (VR) system to ensure that the operator can respond accurately in real failure conditions. First, fatigue data in different time periods is obtained according to the physical properties of the equipment material and the use conditions, and these data are integrated into the VR system for simulation. Then, a failure test environment is created in the VR system, the failure response of the equipment is simulated by dynamically loading the material fatigue data, and the abnormal fluctuations of the equipment data and the emergency response ability of the operator are recorded. Through comprehensive analysis of these data, the accuracy of the VR system in simulating the physical failure mode of the equipment is evaluated, and it is divided into high-accuracy and low-accuracy simulation. For low-accuracy simulation, the accuracy abnormality degree is predicted and analyzed, and if the abnormality degree is high, the failure time and failure conditions of the equipment are dynamically adjusted to improve the accuracy of the emergency simulation, make the virtual environment more realistic and credible, and ultimately enhance the emergency response ability of the operator to ensure safe operation.

[0085] Data acquisition module: according to the physical properties of the equipment material and the use conditions of the pressure vessel, the fatigue data of the equipment material in several time periods is obtained respectively, and the obtained fatigue data of the equipment material is integrated into the VR system;

[0086] Test module: a failure test environment is created in the VR system, the real failure response is simulated by dynamically loading the fatigue data of the equipment material in the VR system, and the abnormal fluctuations of the fatigue data of the equipment material and the emergency response ability of the operator during simulation are recorded;

[0087] Comprehensive analysis module: the abnormal fluctuations of the fatigue data of the equipment material and the emergency response ability of the operator during simulation are comprehensively analyzed, and the accuracy of the VR system in simulating the physical failure mode of the equipment is evaluated;

[0088] Grade division module: according to the evaluation results, the accuracy of the VR system in simulating the physical failure mode of the equipment is divided into different grades, which is divided into high-accuracy simulation and low-accuracy simulation, and corresponding processing is performed;

[0089] Dynamic adjustment module: for low-accuracy simulation, the accuracy abnormality degree of the VR system in simulating the physical failure mode of the equipment is predicted and analyzed, and if the abnormality degree is high, the failure time and failure conditions of the equipment are dynamically adjusted to improve the accuracy of the emergency simulation.

[0090] The above formulas are all dimensionless values calculated, the formula is obtained by collecting a large amount of data to simulate the recent real situation, and the preset parameters in the formula are set by a person skilled in the art according to the actual situation.

[0091] It should be understood that the term "and / or" in this document is only a description of the association relationship between the associated objects, which means that there can be three relationships, for example, A and / or B can represent: A exists alone, A and B exist together, and B exists alone, where A and B can be singular or plural. In addition, the character " / " in this document generally represents an "or" relationship between the associated objects before and after it, but it can also represent an "and / or" relationship. The specific meaning can be understood according to the context before and after it.

[0092] Those skilled in the art can realize that the units and algorithm steps of the examples described in combination with the embodiments disclosed herein can be realized by electronic hardware or a combination of computer software and electronic hardware. Whether the functions are realized in hardware or software depends on the specific application and design constraints of the technical solution. A person skilled in the art can use different methods to realize the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0093] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto, any person skilled in the art can easily think of changes or replacements within the technical scope disclosed in the present application, which should be covered within the protection scope of the present application.

Claims

1. A virtual reality inspection emergency method for safe operation of a pressure vessel, characterized by: Comprising the following steps: S1: According to the physical properties of the equipment material and the use conditions of the pressure vessel, the fatigue data of the equipment material in several time periods are obtained respectively, and the obtained fatigue data of the equipment material are integrated into the VR system; S2: A failure test environment is created in the VR system, the fatigue data of the equipment material in the VR system are dynamically loaded to simulate the real failure response, and the abnormal fluctuation of the fatigue data of the equipment material and the emergency response ability of the operator during the simulation process are recorded; After analyzing the abnormal increase of the fatigue crack propagation rate, a fatigue crack propagation rate abnormality index is generated, and the method for obtaining the fatigue crack propagation rate abnormality index is: Collect and organize the crack growth rate data X = [x1, x2,..., xn], calculate the quartiles according to the data set: Q1: sort the data and find the 25th percentile of the data points, Q3: the third quartile: sort the data and find the 75th percentile of the data points, Q2: sort the data and find the middle data point, i.e. the 50th percentile, the interquartile range IQR: ; according to the rules of the box plot, calculate the upper and lower boundaries of the outliers: Lower Bound: ; Upper Bound: ; according to the calculated upper and lower boundaries, identify outliers: any crack growth rate xi is considered an outlier if or , calculate the degree of deviation between each outlier and Q1 or Q3, and then calculate the fatigue crack growth rate anomaly index, expressed as: ; where: the i-th outlier, the total number of outliers, n is the total number of data points, the degree of deviation of each data point from the median, HK is the fatigue crack growth rate anomaly index; S3: The abnormal fluctuation of the fatigue data of the equipment material and the emergency response ability of the operator during the simulation process are comprehensively analyzed to evaluate the accuracy of the VR system simulation equipment in the physical failure mode; S4: According to the evaluation result, the accuracy of the VR system simulation equipment in the physical failure mode is divided into different levels, which is divided into high accuracy simulation and low accuracy simulation, and corresponding processing is carried out; S5: For low accuracy simulation, the abnormal degree of the accuracy of the VR system simulation equipment in the physical failure mode is predicted and analyzed, and if the abnormal degree is high, the failure time and failure condition of the equipment are dynamically adjusted to improve the emergency simulation precision.

2. The virtual reality inspection emergency method for safe operation of a pressure vessel according to claim 1, characterized in that: In S2, the error operation frequency change drift index is generated after analyzing the error operation frequency change of the operator in the face of abnormal situation, and the method for obtaining the error operation frequency change drift index is: Collect the error operation frequency data of the operation personnel and arrange them in time sequence, set the error operation frequency time series data as where t represents the time point, determine the order of the ARIMA model, where: p is the autoregressive order, represents the number of past values included in the model, d is the difference number, represents the difference number required for data stabilization, q is the moving average order, represents the number of past values of the error term, fit the ARIMA model, the expression is: ; where α is the constant term, is the parameter of the autoregressive term, is the parameter of the moving average term, is the white noise error term, after fitting the model, the estimated value of the parameter ; use the fitted ARIMA model to calculate the fitted value of the model , the expression is: ; the residual is the difference between the actual data and the fitted data, which represents the prediction error of the model: ; calculate the error operation frequency change drift index, the calculation expression is: ; in the formula, GH is the error operation frequency change drift index, is the jth value in the residual sequence, is the average value of the residual.

3. The virtual reality inspection emergency method for safe operation of a pressure vessel according to claim 2, characterized in that: In S3, the fatigue crack propagation rate abnormality index and the error operation frequency change drift index are converted into a comprehensive feature vector, the comprehensive feature vector is taken as the input of the machine learning model, the machine learning model takes the accuracy value label of the VR system simulation equipment in the physical failure mode as the prediction target, and the sum of the prediction errors of all VR system simulation equipment in the physical failure mode is minimized as the training target. The machine learning model is trained until the sum of the prediction errors converges, and the model training is stopped, and the accuracy value of the VR system simulation equipment in the physical failure mode is determined according to the model output result, wherein the machine learning model is a polynomial regression model.

4. The virtual reality inspection emergency method for safe operation of a pressure vessel according to claim 3, characterized in that: In S4, according to the evaluation result, the accuracy of the VR system simulation equipment in the physical failure mode is divided into different levels, which is: The accuracy value of the acquired VR system simulation device in the physical failure mode is compared with the accuracy value reference threshold set in the historical data, if the accuracy value of the VR system simulation device in the physical failure mode is greater than or equal to the accuracy value reference threshold set in the historical data, it indicates that the accuracy of the VR system simulation device in the physical failure mode is high, at this time, a high-accuracy simulation signal is generated, and it is divided into high-accuracy simulation; if the accuracy value of the VR system simulation device in the physical failure mode is less than the accuracy value reference threshold set in the historical data, it indicates that the accuracy of the VR system simulation device in the physical failure mode is low, at this time, a low-accuracy simulation signal is generated, and it is divided into low-accuracy simulation.

5. The virtual reality inspection emergency method for safe operation of a pressure vessel according to claim 1, characterized in that: In S5, for low-accuracy simulation, the abnormality degree of the accuracy of the VR system simulation device in the physical failure mode is predicted and analyzed, if the abnormality degree is high, the failure time and failure condition of the device are dynamically adjusted to improve the emergency simulation accuracy, specifically: calculating the degree of accuracy anomaly by comparing the difference between the actual failure response of the VR simulation device and the theoretical failure response ; Set is the actual device response value in the failure mode, is the expected device failure response value, and the calculation expression of the accuracy anomaly degree is: ; If the accuracy abnormality degree is greater than the preset accuracy abnormality degree reference threshold value, it indicates that the accuracy abnormality degree is high, that is, the failure time and failure condition in the simulation deviate from the real situation, and therefore it is necessary to dynamically adjust the failure time and failure condition .

6. The virtual reality inspection emergency method for safe operation of a pressure vessel according to claim 5, characterized in that: Set is the dynamically adjusted failure time, is the dynamically adjusted failure condition, the failure time and the failure condition are adjusted based on the abnormality degree, and the formula for dynamic adjustment is: ; ; wherein, is the theoretical failure time, is the theoretical failure condition, , is an adjustment factor, indicating the influence degree of the abnormality degree on the failure time and the failure condition.

7. Virtual reality inspection emergency system for safe operation of pressure vessels, for implementing the virtual reality inspection emergency method for safe operation of pressure vessels according to any one of claims 1 to 6, characterized in that: It includes a data acquisition module, a test module, a comprehensive analysis module, a grade division module and a dynamic adjustment module; The data acquisition module: according to the physical properties of the equipment material and the use conditions of the pressure vessel, the equipment material fatigue data in several time periods are acquired respectively, and the acquired equipment material fatigue data are integrated into the VR system; The test module: a fault test environment is created in the VR system, the equipment material fatigue data in the VR system are dynamically loaded to simulate the real failure response, and the abnormal fluctuation of the equipment material fatigue data and the emergency response ability of the operator during the simulation process are recorded; The comprehensive analysis module: the abnormal fluctuation of the equipment material fatigue data and the emergency response ability of the operator during the simulation process are comprehensively analyzed to evaluate the accuracy of the VR system simulation device in the physical failure mode; The grade division module: according to the evaluation result, the accuracy of the VR system simulation device in the physical failure mode is divided into different grades, which is divided into high-accuracy simulation and low-accuracy simulation, and corresponding processing is carried out; The dynamic adjustment module: for low-accuracy simulation, the abnormality degree of the accuracy of the VR system simulation device in the physical failure mode is predicted and analyzed, if the abnormality degree is high, the failure time and failure condition of the device are dynamically adjusted to improve the emergency simulation accuracy.

Images