A method and system for detecting and managing post-fire liquid hydrogen cylinders

Abstract

The application relates to the technical field of liquid hydrogen cylinder detection, in particular to a fire post-liquid hydrogen cylinder detection management method and system, wherein the method comprises the following steps: a preliminary judgment step, a fire-affected part detection step, a toughness prediction step, an adiabatic evaluation step, a safety determination step and a traceability archiving step; wherein the system comprises a processor and a memory in communication connection with the processor. The application realizes independent monitoring in three dimensions and dangerous determination based on a logical or relationship, so that the detection result can comprehensively reflect the real safety state of the cylinder in a super-low-temperature service environment.

Description

Technical Field

[0001] This application relates to the technical field of liquid hydrogen cylinder detection, and in particular to a method and system for the detection and management of liquid hydrogen cylinders after a fire. Background Technology

[0002] Liquid hydrogen, with its high density and efficient storage and transportation, has become an important development direction for on-board hydrogen storage technology. Liquid hydrogen cylinders typically employ a high-vacuum, multi-layered insulation structure to maintain an ultra-low temperature environment of -253°C, ensuring stable storage of liquid hydrogen. However, in extreme situations such as traffic accidents or fires, the cylinders may suffer localized high-temperature impacts, leading to insulation layer failure, degradation of the inner liner material, and even hydrogen embrittlement risks, seriously threatening the structural integrity and operational safety of the cylinders. Therefore, establishing dedicated safety testing methods for liquid hydrogen cylinders after fires is of great significance for ensuring the normal use of liquid hydrogen cylinders.

[0003] In existing technologies, such as patent CN114544749A, a method and system for safety testing of long-tube trailer gas cylinders after a fire is disclosed. This method uses metallographic examination to determine whether a phase change has occurred in the fire-affected area. If no phase change has occurred, magnetic multi-parameter detection combined with a multiple linear regression equation is used to evaluate the mechanical properties. If a phase change has occurred, the mechanical properties are determined by comparing metallographic images with standard values ​​to determine the cylinder's safety. This method has good applicability in the post-fire testing of high-pressure gas cylinders at room temperature.

[0004] However, existing technologies primarily target high-pressure gas cylinders at room temperature, neglecting the unique failure modes of liquid hydrogen cylinders after a fire. First, the inner liner of a liquid hydrogen cylinder is subjected to extremely low temperatures for extended periods. After a fire, it's necessary to consider not only material degradation caused by high temperatures but also the material's ability to recover its low-temperature toughness. Existing magnetic multi-parameter detection models lack correlation with low-temperature performance and cannot be directly applied to the safety assessment of liquid hydrogen cylinders. Second, the insulation system of liquid hydrogen cylinders is highly susceptible to damage in a fire. Existing technologies lack the ability to detect the integrity of the insulation layer and the degree of vacuum, making it impossible to determine whether the cylinder's cold-keeping performance has failed, resulting in blind spots in safety assessments. Furthermore, the hydrogen embrittlement sensitivity of materials increases significantly in a liquid hydrogen environment. Existing technologies do not consider the assessment of hydrogen embrittlement risk after a fire, making it difficult to comprehensively reflect the true safety status of the cylinder. Summary of the Invention

[0005] This application provides a method and system for detecting and managing liquid hydrogen cylinders after a fire, which can at least partially solve the above-mentioned technical problems.

[0006] Firstly, this application provides a method for detecting and managing liquid hydrogen cylinders after a fire, which adopts the following technical solution: A method for the detection and management of liquid hydrogen cylinders after a fire includes the following steps: Preliminary assessment: Metallographic examination of the liquid hydrogen cylinder after the fire was conducted to determine whether the metallographic structure of the fire-affected and non-fire-affected areas was consistent, thus obtaining the first assessment result; Fire-affected area detection: If the first judgment result is yes, then perform magnetic multi-parameter detection on the fire-affected area to obtain magnetic multi-parameter characteristic signals; Toughness prediction: The magnetic multi-parameter characteristic signal is input into a pre-established low-temperature mechanical property multiple linear regression model to obtain the predicted value of the low-temperature impact toughness of the inner liner material. The low-temperature mechanical property multiple linear regression model is established based on the low-temperature impact toughness test data of damaged test blocks at -196℃ under different fire temperatures and the corresponding magnetic multi-parameter characteristic signal regression. Insulation assessment: Vacuum degree detection and infrared thermal imaging analysis are performed on the gas cylinder to obtain the vacuum loss rate as the result of the insulation layer integrity assessment; Safety determination: The predicted value of low temperature impact toughness is compared with the preset low temperature toughness threshold, and combined with the vacuum loss rate. If the predicted value of low temperature impact toughness is lower than the threshold or the vacuum loss rate exceeds the insulation failure threshold, the gas cylinder is determined to be in a dangerous state, and the safety handling procedure is initiated. Traceability and Filing: The test data, intermediate calculation results and judgment results obtained during the test are used to generate a test traceability report, which is stored to establish a full life cycle test file for gas cylinders.

[0007] By adopting the above technical solutions and employing a graded judgment strategy that combines metallographic inspection and magnetic multi-parameter detection, the technical blind spot of being unable to assess material performance degradation by directly relying on metallographic inspection when no phase change has occurred is reduced. A multiple linear regression model established based on low-temperature impact toughness test data at -196 degrees Celsius enables the magnetic multi-parameter detection signal to accurately map the low-temperature toughness of the inner liner material in the liquid hydrogen temperature range, solving the deficiency of existing technologies where magnetic multi-parameter detection is only applicable to room-temperature mechanical properties and cannot be used in ultra-low-temperature service environments. The introduction of vacuum loss rate makes the integrity of the insulation layer an independent assessment dimension, compensating for the safety assessment blind spot of existing technologies that only focus on the cylinder body material and ignore the damage to the insulation system. The determination of hazardous states uses a logical OR relationship between the low-temperature toughness threshold and the insulation failure threshold; if either dimension exceeds the safe range, the handling procedure is triggered, reflecting the safety management characteristics of multi-system coupled failures in liquid hydrogen cylinders. The establishment of inspection traceability reports and full life-cycle inspection archives provides a data foundation for subsequent re-inspection, accident tracing, and inspection strategy optimization.

[0008] Optionally, a hydrogen embrittlement assessment step may also be included; Hydrogen embrittlement assessment: Electrochemical hydrogen permeation test or thermal desorption spectrum analysis is performed on the fire-affected part to determine the hydrogen diffusion coefficient and hydrogen trap density, and the hydrogen embrittlement sensitivity index is obtained; If the hydrogen embrittlement sensitivity index exceeds the preset hydrogen embrittlement threshold, the gas cylinder is determined to have a risk of hydrogen embrittlement, and this risk is recorded in the detection traceability report.

[0009] By adopting the above technical solutions, hydrogen embrittlement sensitivity of materials in a liquid hydrogen environment is a unique failure mode of liquid hydrogen cylinders, and changes in material structure during a fire may exacerbate the risk of hydrogen embrittlement. Electrochemical hydrogen permeation testing can quantitatively determine the diffusion behavior of hydrogen in materials, and thermal desorption spectroscopy analysis can characterize the density and distribution of hydrogen traps. Both of these detection methods are direct quantitative methods for hydrogen embrittlement sensitivity. The introduction of the hydrogen embrittlement sensitivity index expands the safety assessment from a single material strength degradation to a comprehensive evaluation of the material's compatibility with the hydrogen environment, further improving the comprehensiveness of the test results. Recording the hydrogen embrittlement risk in the test traceability report allows this risk information to be stored in the same file as information such as low-temperature toughness and thermal insulation integrity, providing data support for subsequent comprehensive judgment.

[0010] Optionally, it also includes a sub-health status assessment step; Sub-health status assessment: Sub-health ranges were set for the predicted low-temperature impact toughness value, the vacuum loss rate, and the hydrogen embrittlement sensitivity index, respectively. When the predicted value of low temperature impact toughness, vacuum loss rate and hydrogen embrittlement sensitivity index are not in the danger range, but at least one of them enters the corresponding sub-health range, the gas cylinder is judged to be in a sub-healthy state. The results of the sub-health status assessment will be recorded in the testing and traceability report.

[0011] By adopting the above technical solution, the setting of the sub-health range enables the detection method to identify gas cylinders that have not yet reached the danger threshold but have shown a trend of performance degradation, reducing the potential for missed critical state assessments caused by the binary division of safety and danger. Using the sub-health ranges of the three indicators as the judgment criteria reflects the independent monitoring of the three dimensions of low-temperature toughness, thermal insulation integrity, and hydrogen embrittlement risk. Entering the sub-health range in any dimension triggers the sub-health state judgment, which conforms to the basic principle of the "weakest link" in safety management. Recording the sub-health state judgment results allows subsequent steps such as setting dynamic detection cycles, fault prediction, and life prediction to initiate a refined evaluation process for gas cylinders in a sub-health state.

[0012] Optionally, it also includes a period setting step; Periodic setting: When the gas cylinder is in a sub-healthy state, calculate the predicted value of low temperature impact toughness, vacuum loss rate, and hydrogen embrittlement sensitivity index relative to their respective danger boundaries. The minimum value among the remaining safety margins is taken as the dominant margin, and the dynamic detection period is set according to the interval in which the dominant margin is located; Record the set testing cycle in the testing traceability report.

[0013] By adopting the above technical solution, the detection cycle is set according to the dominant margin range, so that the detection frequency matches the degree of deterioration of the gas cylinder: the smaller the remaining safety margin and the closer to the dangerous state, the shorter the detection cycle. Compared with the fixed cycle detection strategy, this state-based dynamic cycle setting method can allocate detection resources more effectively and avoid over-detection or under-detection while ensuring safety. The detection cycle is recorded in the detection traceability report, so that subsequent re-inspection can be performed according to the preset cycle.

[0014] Optionally, it also includes a fault prediction step; Fault prediction: When the gas cylinder is in a sub-healthy state, the current and historical detection data are input into the pre-established fault prediction model to predict the time when each indicator will enter a dangerous state within a preset time window in the future. The minimum predicted time to enter a dangerous state for each indicator is taken as the predicted dangerous time for the gas cylinder. ; when When the risk level is below the preset risk threshold, an early warning message is output and the detection cycle is shortened accordingly. The prediction results are recorded in the detection traceability report.

[0015] By adopting the above technical solution, the fault prediction model extrapolates the future trends of various indicators based on historical detection data, and can predict the time when each indicator reaches the danger boundary. Taking the minimum value of the predicted time of entering the dangerous state for each indicator as the predicted danger time reflects the risk management idea of ​​identifying the earliest possible danger among multiple deterioration paths. When the predicted danger time is less than the preset risk threshold, it indicates that the gas cylinder may enter a dangerous state in the short term. At this time, the warning information is output and the detection cycle is shortened accordingly, realizing the transformation from passive detection to active warning. The prediction results are recorded in the detection traceability report, so that the warning information and the detection data are stored together in the same file, which facilitates the subsequent verification of the prediction accuracy and optimization of the prediction model.

[0016] Optionally, when the gas cylinder is in a sub-healthy state, based on the current values ​​of each indicator and its historical decay rate, the remaining time for each indicator to reach the danger boundary is predicted, and the minimum value is taken as the remaining safe service life of the gas cylinder. The calculation formula is as follows: ; in , , These are the predicted values ​​of low-temperature impact toughness, vacuum loss rate, and average decay rates of hydrogen embrittlement sensitivity index, respectively, obtained by fitting historical test data. Record the remaining safe service life prediction results in the inspection traceability report.

[0017] By adopting the above technical solutions, the introduction of historical decay rates allows life prediction to be based on the actual degradation trend of the gas cylinder rather than a static threshold, which can more accurately reflect the remaining safe service time of the gas cylinder; the use of a minimum value function to take the minimum value among the remaining time of the three indicators to reach the danger boundary reflects the safety management principle that the gas cylinder's remaining life is determined by the indicator that reaches the danger boundary first; the remaining safe service life is quantified into specific time values, providing a quantitative basis for subsequent inspection cycle adjustments, replacement plan formulation, and scrapping decisions; the prediction results are recorded in the inspection traceability report, so that life prediction information can be stored in the same file as historical inspection data.

[0018] Optionally, it may also include adding adjustment steps; New adjustment: In subsequent re-examinations, if a new indicator enters the sub-health range for the first time, or if an existing indicator moves from a lower-level sub-health range to a higher-level sub-health range, dynamic adjustment of life expectancy prediction will be triggered. Recalculate the remaining safety margin and decay rate of each current indicator, and update the predicted remaining safe service life. ; like Then Update the remaining safe service life of the gas cylinder, and according to Reset the testing cycle; The adjustment process and the updated prediction results will be recorded in the testing traceability report.

[0019] By adopting the above technical solution, when the condition of the gas cylinder deteriorates (new indicators enter the sub-healthy range or existing indicators deteriorate further), the system automatically triggers a recalculation of the lifespan prediction, ensuring that the predicted remaining safe service life always reflects the latest gas cylinder condition. Updates are only performed when the updated prediction value is lower than the original prediction value, reflecting a conservative prediction principle and avoiding overestimation of the safety margin due to fluctuations in the prediction value. The detection cycle is reset based on the updated remaining safe service life, allowing the detection frequency to dynamically adjust in accordance with the deterioration of the gas cylinder condition. The recording of the adjustment process and update results provides data support for subsequent analysis of the gas cylinder degradation trajectory.

[0020] Optionally, it also includes a lifetime prediction correction step; Life expectancy prediction correction: When multiple indicators simultaneously enter or are within the sub-healthy range, a coupling correction coefficient is introduced. The remaining safe service life is adjusted using the following formula: ;in , The number of indicators required to enter the sub-healthy zone, when hour , For the first The relative deviation of each indicator within its sub-healthy range; The relative deviation is calculated as follows: Relative deviation of predicted low-temperature impact toughness ; Relative deviation of vacuum loss rate ; Relative deviation of the hydrogen embrittlement sensitivity index ; Record the revised remaining safe service life in the inspection traceability report.

[0021] By adopting the above technical solution, when multiple indicators simultaneously enter the sub-healthy zone, the synergistic deterioration effect among the indicators may accelerate the degradation of the overall performance of the gas cylinder. The independent prediction of a single indicator cannot reflect this coupled effect. The coupling correction coefficient considers both the number of indicators entering the sub-healthy zone and the relative deviation of each indicator within the sub-healthy zone: the more indicators there are and the farther each indicator deviates from the starting point of the sub-healthy zone, the smaller the correction coefficient and the greater the reduction in the remaining safe service life. The formula for calculating the relative deviation standardizes the position of each indicator within its respective sub-healthy zone to a value between zero and one, enabling indicators of different dimensions to be compared and combined on the same scale. This correction mechanism makes the life prediction results more consistent with the actual gas cylinder safety status under the condition of synergistic deterioration of multiple indicators.

[0022] Optionally, it also includes a step for constructing composite health indicators; Construction of a composite health index: When a gas cylinder is in a sub-healthy state, a composite health index C is constructed as a quantitative representation of the overall health status of the gas cylinder. The calculation formula is as follows:

[0023] A dynamic covariate V(t) is introduced as an external environmental influencing factor. The dynamic covariate includes the rate of change of ambient temperature, the cumulative value of the service life of the gas cylinder, or the number of filling and depressurization cycles. The dynamic covariates are incorporated into the decay rate correction model to construct covariate correction factors. , where β is the covariate sensitivity coefficient, obtained by fitting historical detection data; The composite health index C is dynamically compensated using the aforementioned covariate correction factor: The overall health status value after compensation is obtained; when When the health status falls below the preset health threshold, an early warning message is output and the detection cycle is shortened accordingly. The compensated comprehensive health status value is recorded in the detection traceability report.

[0024] By adopting the above technical solution, the composite health index integrates the current values, danger boundary values, and decay rates of three indicators—low-temperature toughness, vacuum loss rate, and hydrogen embrittlement sensitivity index—as well as coupling correction coefficients, forming a single quantitative characterization of the overall health status of the gas cylinder. The introduction of dynamic covariates allows external environmental factors to be incorporated into the health status assessment. The rate of change in ambient temperature reflects the severity of the service environment, the cumulative value of the gas cylinder's service years reflects the cumulative effect over time, and the number of pressurization and depressurization cycles reflects the cumulative impact of fatigue loads. The exponential function form of the covariate correction factor causes the influence of the dynamic covariate to be exponentially amplified or attenuated, consistent with the nonlinear influence characteristics of environmental factors during material degradation. The dynamically compensated comprehensive health status value can more accurately reflect the safety status of the gas cylinder under actual service conditions. When it falls below a preset health threshold, an early warning is triggered and the detection cycle is shortened, realizing dynamic detection management driven by the external environment.

[0025] Secondly, the post-fire liquid hydrogen cylinder detection and management system provided in this application adopts the following technical solution: A post-fire liquid hydrogen cylinder detection and management system includes: a processor, and a memory communicatively connected to the processor; The memory is provided with a computer-readable storage medium, and a computer program is stored on the computer-readable storage medium. When the processor processes the computer program stored on the computer-readable storage medium, it implements the post-fire liquid hydrogen cylinder detection and management method.

[0026] In summary, this application includes at least one of the following beneficial technical effects: 1. Targeting the unique failure modes after liquid hydrogen cylinder fires, three independent and complementary detection dimensions were constructed: low-temperature impact toughness, vacuum loss rate, and hydrogen embrittlement sensitivity index. The low-temperature impact toughness prediction is based on a magnetic multi-parameter regression model established using measured data at -196 degrees Celsius, which solves the technical deficiency of existing technologies where magnetic multi-parameter detection is only applicable to room temperature mechanical properties and cannot assess the toughness of materials in the liquid hydrogen temperature range. The introduction of vacuum loss rate compensates for the blind spot of existing technologies in detecting the lack of insulation layer integrity. The determination of hydrogen embrittlement sensitivity index incorporates the compatibility of materials with the hydrogen environment into the evaluation system. The independent monitoring of the three dimensions and the hazard judgment based on logical OR relationships enable the test results to comprehensively reflect the true safety status of the cylinder in the ultra-low temperature service environment. 2. A dynamic detection cycle setting method based on the minimum remaining safety margin as the dominant margin enables the detection frequency to adaptively match the degree of gas cylinder deterioration. The smaller the remaining safety margin and the closer to the dangerous state, the shorter the detection cycle. This avoids over-detection or under-detection that may be caused by fixed-cycle detection, and also achieves optimized allocation of detection resources. 3. The establishment of inspection traceability reports and full life cycle inspection archives forms a complete management closed loop from data collection, status assessment, risk prediction to dynamic optimization of inspection strategies, providing a complete technical solution for the refined and intelligent safety management of liquid hydrogen cylinders after a fire. Attached Figure Description

[0027] Figure 1 This is a flowchart of the detection management method in Embodiment 1 of this application; Figure 2 This is a flowchart of the detection management method in Embodiment 2 of this application; Figure 3 This is a flowchart of the detection management method in Embodiment 3 of this application; Figure 4 This is a flowchart of the detection management method in Embodiment 4 of this application; Figure 5 This is a flowchart of the detection management method in Embodiment 5 of this application; Figure 6 This is a flowchart of the detection management method in Embodiment 6 of this application. Detailed Implementation

[0028] The following combination Figures 1 to 6 This application will be described in further detail.

[0029] This embodiment discloses a method for detecting and managing liquid hydrogen cylinders after a fire.

[0030] Example 1: Refer to Figure 1 The post-fire inspection and management methods for liquid hydrogen cylinders include the following steps: Preliminary assessment: Metallographic examination of the liquid hydrogen cylinder after the fire was conducted to determine whether the metallographic structure of the fire-affected and non-fire-affected areas was consistent, thus obtaining the first assessment result; Specifically, a metallographic microscope is used to observe the metallographic structure of the fire-exposed and non-fire-exposed parts of the gas cylinder. The metallographic images of the fire-exposed part and the non-fire-exposed part are compared to determine whether their metallographic structures are consistent. If they are consistent, it indicates that the fire temperature has not reached the material phase transformation temperature (about 740°C), and the subsequent testing process is initiated. If they are inconsistent, the mechanical properties are directly evaluated based on the degree of change in the metallographic structure.

[0031] Fire-affected area detection: If the first judgment result is yes, then perform magnetic multi-parameter detection on the fire-affected area to obtain magnetic multi-parameter characteristic signals; Specifically, a multi-parameter electromagnetic non-destructive testing instrument was used. The probe was attached to the surface of the fire-affected area, and four physical signals were collected simultaneously: magnetic Barkhausen noise signal, multi-frequency eddy current signal, incremental permeability signal, and tangential magnetic field harmonic signal. During the test, the probe scanned at a constant speed, the sampling frequency was set to 1 kHz, and at least three valid data points were collected at each measurement point. The average value was taken as the raw signal. The raw signal was then filtered for noise reduction, signal separation, and feature extraction to obtain a set of magnetic multi-parameter feature signals, including the amplitude of the magnetic Barkhausen noise signal. and envelope features Impedance amplitude of multi-frequency eddy current signal and phase The width of the incremental permeability signal curve and amplitude The second harmonic amplitude of the tangential magnetic field harmonic signal and third harmonic phase These characteristic signals constitute a multidimensional input vector.

[0032] Toughness prediction: The magnetic multi-parameter characteristic signal is input into a pre-established low-temperature mechanical property multiple linear regression model to obtain the predicted value of the low-temperature impact toughness of the inner liner material. The low-temperature mechanical property multiple linear regression model is established based on the low-temperature impact toughness test data of damaged test blocks at -196℃ under different fire temperatures and the corresponding magnetic multi-parameter characteristic signal regression. Specifically, the aforementioned magnetic multi-parameter characteristic signals are input into a pre-established low-temperature mechanical property multiple linear regression model to obtain the predicted value of the low-temperature impact toughness of the inner liner material. (Unit: J); The process of establishing this regression model is as follows: First, samples were cut from the same batch of gas cylinder materials and heat-treated at different firing temperatures (200℃, 300℃, 400℃, 500℃, 600℃, 650℃, 700℃, 730℃, 750℃, 770℃, 800℃, 850℃) to simulate heat treatment. After holding at each temperature for 2 hours, the samples were air-cooled to obtain 12 types of damaged test blocks. Each damaged test block was subjected to a Charpy V-notch impact test at -196℃ (liquid nitrogen environment), and the measured low-temperature impact toughness was obtained according to GB / T 229-2020 standard. Simultaneously, magnetic multi-parameter detection was performed on each damaged specimen to extract the same characteristic signals as described above. Using these characteristic signals as independent variables, Using the dependent variable, a multiple linear regression method is used to establish the equation: ; Regression coefficient to The goodness of fit was determined by the least squares method. It should be no less than 0.95 to ensure the accuracy of model predictions.

[0033] Insulation assessment: Vacuum degree detection and infrared thermal imaging analysis are performed on the gas cylinder to obtain the vacuum loss rate as the result of the insulation layer integrity assessment; Specifically, the gas cylinders undergo vacuum testing and infrared thermal imaging analysis. Vacuum testing utilizes a high-precision vacuum gauge, measuring the absolute pressure of the interlayer space after a fire via a pre-installed vacuum measurement interface within the gas cylinder interlayer. (Unit: Pa). Retrieve the initial vacuum level of the gas cylinder at the time of manufacture. (Unit: Pa), Calculate the vacuum loss rate: ; Simultaneously, an infrared thermal imager was used to scan the outer surface of the gas cylinder to obtain images of the surface temperature distribution, focusing on observing localized overheated areas of the insulation layer to help determine the degree of damage to the insulation layer. Vacuum loss rate. It serves as a core quantitative indicator for assessing the integrity of insulation layers.

[0034] Safety determination: The predicted value of low temperature impact toughness is compared with the preset low temperature toughness threshold, and combined with the vacuum loss rate. If the predicted value of low temperature impact toughness is lower than the threshold or the vacuum loss rate exceeds the insulation failure threshold, the gas cylinder is determined to be in a dangerous state, and the safety handling procedure is initiated. Specifically, the predicted value of low-temperature impact toughness Compared with the preset low temperature toughness threshold Compare and simultaneously consider vacuum loss rate With the thermal failure threshold Comparison. Preset low-temperature toughness threshold. The value is determined according to the gas cylinder design standard (such as GB / T 34510-2017), and is taken as the lower limit of the material's impact absorption energy at -196℃, for commonly used austenitic stainless steel (such as S30408). Take 60J. Insulation failure threshold. Based on the design requirements of the gas cylinder insulation system, when the vacuum loss rate exceeds 20% or the absolute pressure of the interlayer is higher than 10 Pa, the insulation performance is considered insufficient to meet the requirements for liquid hydrogen storage. Therefore, the following settings are established: .like or If the gas cylinder is found to be in a dangerous state, the safety handling procedure should be initiated immediately, including moving the gas cylinder to a safe area, emptying the residual medium, attaching a hazard sign, and recording the handling process.

[0035] Traceability and Filing: The test data, intermediate calculation results and judgment results obtained during the test are used to generate a test traceability report, which is stored to establish a full life cycle test file for gas cylinders.

[0036] Specifically, all data acquired during the testing process, including metallographic images, raw magnetic multi-parameter data, vacuum measurements, infrared thermal images, intermediate calculation results (such as characteristic signals and predicted values), final judgment results, and handling records, are used to generate a test traceability report in a structured form. The report includes the cylinder's unique identification code, testing time, testing personnel, and testing equipment information. A hash algorithm is used to calculate the report summary, which is then stored on a blockchain platform to ensure data immutability. Simultaneously, the report is stored in a local database, forming a full lifecycle test file for the cylinder, which can be used for subsequent re-inspection, accident tracing, and testing strategy optimization.

[0037] For example, a certain model of liquid hydrogen cylinder has a volume of 500L, a design pressure of 2.5MPa, an inner liner made of S30408 ​​stainless steel, and an initial vacuum level of [missing information]. During a transport operation, the rear of the gas cylinder suffered a localized fire, with the temperature in the affected area reaching approximately 650°C. After the fire was extinguished, inspectors used a portable metallographic microscope to conduct metallographic examinations of the affected area (rear of the cylinder) and the unaffected area (front of the cylinder). The results showed that both areas had austenitic microstructures without pearlite or martensite transformation, indicating consistent microstructures. Therefore, the initial assessment was positive. Subsequently, a magnetic multi-parameter detector was used to scan the affected area and extract characteristic signals. , , , , , , , Substituting these values ​​into the established regression model, the predicted values ​​of low-temperature impact toughness are calculated. Meanwhile, the interlayer pressure was measured using a vacuum gauge. Calculate the vacuum loss rate Preset low-temperature toughness threshold Insulation failure threshold .because If the conditions for danger are met, the gas cylinder is determined to be in a dangerous state, and the safety handling procedure is initiated. An inspection report is automatically generated and stored on the blockchain, and the entire lifecycle inspection record of the gas cylinder is also entered into the database.

[0038] This embodiment uses metallographic examination to screen out gas cylinders that have not undergone phase change, and then uses a magnetic multi-parameter regression model based on -196℃ low-temperature impact toughness test data to accurately predict the low-temperature toughness of the inner liner material in the liquid hydrogen temperature range. At the same time, it introduces vacuum degree loss rate to quantify the damage to the insulation layer, and uses a logical OR relationship between the two to determine the danger. This effectively solves the problem that existing technologies cannot assess the coupled failure of low-temperature toughness and insulation performance after a liquid hydrogen gas cylinder fire, and improves the comprehensiveness and reliability of the detection.

[0039] Example 2: Refer to Figure 2 The difference between this embodiment and Embodiment 1 is that it also includes a hydrogen embrittlement assessment step; Hydrogen embrittlement assessment: Electrochemical hydrogen permeation test or thermal desorption spectrum analysis is performed on the fire-affected part to determine the hydrogen diffusion coefficient and hydrogen trap density, and the hydrogen embrittlement sensitivity index is obtained; If the hydrogen embrittlement sensitivity index exceeds the preset hydrogen embrittlement threshold, the gas cylinder is determined to have a risk of hydrogen embrittlement, and this risk is recorded in the detection traceability report.

[0040] Specifically, after the fire-affected area is inspected, a non-destructive testing method using a portable electrochemical hydrogen permeation testing device is employed. During electrochemical hydrogen permeation testing, the surface of the fire-affected area is cleaned, an electrolytic cell is installed, a constant cathode current is applied, the steady-state permeation current density is measured, and the hydrogen diffusion coefficient is calculated using Fick's first law. (Unit: m² / s). When using thermal desorption spectroscopy, the sample is placed in a vacuum chamber and heated to 500°C at a constant heating rate (e.g., 10°C / min). The hydrogen desorption rate is recorded using a mass spectrometer, and the total hydrogen content and hydrogen trap density are obtained by integration. (Unit: m⁻³). Combining the hydrogen diffusion coefficient and hydrogen trap density, a hydrogen embrittlement sensitivity index H is constructed, defined as: ; in The reference value for the hydrogen diffusion coefficient of unfired materials. The reference value for hydrogen trap density of unfired materials. The weighting factor is set to 0.5. The preset hydrogen embrittlement critical value is... Based on the material's resistance to hydrogen embrittlement in a liquid hydrogen environment, the following criteria are adopted: .like If the gas cylinder is found to have a hydrogen embrittlement risk, this risk will be recorded in the detection traceability report, and the hydrogen embrittlement sensitivity index will be recorded. It is included in the safety assessment file as an independent indicator.

[0041] For example, based on the testing in Example 1, the testing personnel used a portable electrochemical hydrogen permeation device to test the burned area (650°C region). The hydrogen diffusion coefficient was measured. Known baseline values ​​for unfired materials Simultaneously, the hydrogen trap density was measured through on-site thermal desorption analysis. benchmark value Calculate the hydrogen embrittlement sensitivity index: ; Preset hydrogen embrittlement critical value ,because The gas cylinder is determined to have a risk of hydrogen embrittlement; this risk information is recorded in the inspection traceability report and stored in the gas cylinder's full life cycle inspection file along with the predicted value of low temperature impact toughness and the vacuum loss rate.

[0042] This embodiment, building upon Embodiment 1, adds an independent assessment of hydrogen embrittlement risk. It directly quantifies the material's hydrogen embrittlement sensitivity after a fire through electrochemical hydrogen permeation or thermal desorption spectroscopy, overcoming the shortcomings of existing technologies that do not adequately consider hydrogen embrittlement failure in liquid hydrogen environments. The introduction of the hydrogen embrittlement sensitivity index expands the safety assessment from a single mechanical property to the material's compatibility with the hydrogen environment, further enhancing the comprehensiveness and safety of the detection.

[0043] Example 3: Reference Figure 3 The difference between this embodiment and embodiment 2 is that it also includes a sub-health state determination step; Sub-health status assessment: Sub-health ranges were set for the predicted low-temperature impact toughness value, the vacuum loss rate, and the hydrogen embrittlement sensitivity index, respectively. When the predicted value of low temperature impact toughness, vacuum loss rate and hydrogen embrittlement sensitivity index are not in the danger range, but at least one of them enters the corresponding sub-health range, the gas cylinder is judged to be in a sub-healthy state. The results of the sub-health status assessment will be recorded in the testing and traceability report.

[0044] Specifically, sub-healthy ranges are defined for three indicators. For the predicted value of low-temperature impact toughness... The danger zone is The sub-healthy range is The safe interval is Regarding vacuum loss rate The danger zone is The sub-healthy range is The safe interval is Regarding the hydrogen embrittlement sensitivity index The danger zone is The sub-healthy range is The safe interval is When all three indicators are below the danger zone, but at least one indicator is below its corresponding sub-health zone, the gas cylinder is determined to be in a sub-healthy state, and the sub-healthy state determination result is recorded in the detection traceability report.

[0045] It also includes a cycle setting step; Periodic setting: When the gas cylinder is in a sub-healthy state, calculate the predicted value of low temperature impact toughness, vacuum loss rate, and hydrogen embrittlement sensitivity index relative to their respective danger boundaries. The minimum value among the remaining safety margins is taken as the dominant margin, and the dynamic detection period is set according to the interval in which the dominant margin is located; Record the set testing cycle in the testing traceability report.

[0046] Specifically, when the gas cylinder is in a sub-healthy state, the remaining safety margins of the three indicators relative to their respective danger boundaries are calculated: ; ; ; The minimum of the three residual safety margins is taken as the dominant margin. The dynamic detection period is set according to the range in which the dominant margin is located: when At that time, the testing cycle was set at 12 months; when At that time, the testing cycle was set at 6 months; when At that time, the testing cycle was set at 3 months; when At that time, the testing cycle is set at 1 month, and a detailed evaluation is recommended immediately; when When this occurs, it indicates that a dangerous state has been entered, and the set cycle is no longer set; the safety handling process is initiated directly. The set testing cycle is recorded in the testing traceability report as the basis for subsequent re-testing.

[0047] For example, using the data from Examples 1 and 2, it is known that , , , , , First, determine the range of each indicator: The gas cylinder has entered a dangerous zone, therefore it is directly classified as a dangerous state, and the sub-health state assessment is not applicable. Now, let's assume the test data for another gas cylinder is: , , , , , Therefore, none of the indicators have entered the danger zone. and They have entered the sub-healthy zone. and They have entered the sub-healthy zone. Furthermore, if the value is less than 1.2, it enters the sub-healthy range. If all three indicators fall into the sub-healthy range, the gas cylinder is determined to be in a sub-healthy state. Calculate the remaining safety margin: ; ; ; Dominance margin The concentration is between 5% and 15%, and the testing cycle is set at 3 months. This cycle is recorded in the testing traceability report.

[0048] This embodiment introduces a three-level classification of sub-healthy states and a dynamic cycle setting based on the dominant margin, which solves the problem that binary judgment cannot reflect the critical state. The setting of the sub-healthy range enables the detection method to identify gas cylinders that have not yet reached the danger threshold but have shown a trend of performance degradation. The cycle setting based on the dominant margin realizes the adaptive matching between the detection frequency and the degree of degradation, which avoids both over-detection and under-detection, and significantly improves the precision and economy of detection management.

[0049] Example 4: Reference Figure 4 The difference between this embodiment and embodiment 3 is that it also includes a fault prediction step; Fault prediction: When the gas cylinder is in a sub-healthy state, the current and historical detection data are input into the pre-established fault prediction model to predict the time when each indicator will enter a dangerous state within a preset time window in the future. The minimum predicted time to enter a dangerous state for each indicator is taken as the predicted dangerous time for the gas cylinder. ; when When the risk level is below the preset risk threshold, an early warning message is output and the detection cycle is shortened accordingly. The prediction results are recorded in the detection traceability report.

[0050] Specifically, when a gas cylinder is in a sub-optimal state, current and historical monitoring data are input into a pre-established fault prediction model. This model employs a time-series-based exponential smoothing method to extrapolate the historical values ​​of each indicator and predict its future changes over time. Specifically, for the predicted value of low-temperature impact toughness... Its trend can be described by a first-order exponential smoothing model: ; in The average decay rate is obtained through linear regression of historical data. Solve the equation... ,get Time to reach the danger boundary Similarly, for Solve get ,right Solve get The minimum value among the three is taken as the predicted hazardous time for the gas cylinder. ; Preset risk threshold: when When, issue an emergency warning; when When, issue a short-term warning; when In such cases, a medium-term early warning will be issued. Simultaneously, the detection cycle will be shortened accordingly based on the early warning level: 1 month for emergency warnings, 3 months for short-term warnings, and 6 months for medium-term warnings. The prediction results and early warning information will be recorded in the detection traceability report.

[0051] When a gas cylinder is in a sub-healthy state, based on the current values ​​of each indicator and its historical decay rate, the remaining time for each indicator to reach the danger boundary is predicted, and the minimum value is taken as the remaining safe service life of the gas cylinder. The calculation formula is as follows: ; in , , These are the predicted values ​​of low-temperature impact toughness, vacuum loss rate, and average decay rates of hydrogen embrittlement sensitivity index, respectively, obtained by fitting historical test data. Record the remaining safe service life prediction results in the inspection traceability report.

[0052] Specifically, based on the current values ​​and historical decay rates of each indicator, the remaining time for each indicator to reach the danger boundary is predicted, and the minimum value is taken as the remaining safe service life of the gas cylinder. : ; in , , The average decay rates (units: J / month, % / month, dimensionless / month) for the three indicators were obtained using linear regression fitting with historical test data (at least two test values). If historical test data is insufficient, accelerated aging test data from the same batch of gas cylinders or test blocks of the same material can be used as a substitute. The prediction results are recorded in the testing traceability report.

[0053] For example, a gas cylinder underwent three re-inspections, and the index values ​​at three time points were recorded: First time (t=0 months): , , ; Second time (t=June): , , ; The third time (t=December): , , ; Perform linear regression on each indicator to obtain the decay rate: Take the absolute value of 0.667 J / month; ; ; Current (t=12 months) values ​​of each indicator: , , threshold , , .

[0054] Calculate the time it takes for each indicator to reach the danger boundary: ; ; ; Then predict the dangerous time. . The value equals the short-term warning threshold, triggering a short-term warning, and the detection cycle is adjusted to 3 months. Remaining safe service life: .

[0055] This embodiment adds fault prediction and life prediction functions, extending inspection management from current status assessment to future risk prediction. By quantifying the degradation trend of indicators through historical decay rates, it is possible to predict in advance the time window when gas cylinders enter a dangerous state and dynamically adjust the inspection cycle accordingly to achieve proactive early warning. The quantitative prediction of remaining safe service life provides a scientific basis for formulating replacement plans and optimizing maintenance strategies, significantly improving the foresight and decision support capabilities of inspection management.

[0056] Example 5: Refer to Figure 5 The difference between this embodiment and embodiment 4 is that it also includes an additional adjustment step; New adjustment: In subsequent re-examinations, if a new indicator enters the sub-health range for the first time, or if an existing indicator moves from a lower-level sub-health range to a higher-level sub-health range, dynamic adjustment of life expectancy prediction will be triggered. Recalculate the remaining safety margin and decay rate of each current indicator, and update the predicted remaining safe service life. ; like Then Update the remaining safe service life of the gas cylinder, and according to Reset the testing cycle; The adjustment process and the updated prediction results will be recorded in the testing traceability report.

[0057] Specifically, in subsequent re-examinations, if a new indicator enters the sub-health range for the first time, or if an existing indicator moves from a lower-level sub-health range to a higher-level sub-health range, dynamic adjustments to life expectancy prediction will be triggered.

[0058] The sub-health range of each indicator is divided into three parts: the front, middle and back. The position of the existing indicator in the sub-health range moves towards the danger side, that is, from the front to the middle, from the middle to the back, or directly from the front to the back.

[0059] Specifically, current detection data is merged with historical data to recalculate the remaining safety margin, decay rate, and predicted remaining safe service life for each indicator. ,like (in (If the value is the predicted value after the last adjustment), then use Update the remaining safe service life of the gas cylinder, and according to The testing cycle was reset according to the rules in Example 4. The adjustment process and the updated prediction results were recorded in the testing traceability report.

[0060] For example, following the scenario in Example 4, the gas cylinder's inspection cycle was set to 3 months after the 12-month inspection, meaning a re-inspection should be conducted in the 15th month. In the 14th month, due to a change in the gas cylinder's operating status (such as encountering a second minor fire), an earlier, temporary inspection was performed. The inspection data is as follows: , , Current values ​​for each indicator: It has entered the sub-healthy zone and is approaching the danger zone. Entering the sub-healthy zone and with increasing deviation, Still in a sub-healthy state. During this test, The percentage has risen from 18% to 19.5%, indicating a shift from the lower-level sub-health range (18% ≤ 20% and > 14%) to the higher-level sub-health range (closer to 20%), triggering dynamic adjustment. The decay rate is recalculated: the current data is combined with three historical data sets, and a new decay rate is obtained through refitting. , , Calculate the new remaining safe service life: (Approximately 21 days).

[0061] Original predicted value , Therefore, the remaining safe service life is updated to 0.694 months, and the testing cycle is reset based on the new value: because (in fact) The testing cycle is set to 1 month, and the adjustment process is recorded in the testing traceability report.

[0062] Example 6: Refer to Figure 6 The difference between this embodiment and embodiment 5 is that it also includes a lifetime prediction correction step; Life expectancy prediction correction: When multiple indicators simultaneously enter or are within the sub-healthy range, a coupling correction coefficient is introduced. The remaining safe service life is adjusted using the following formula: ;in , The number of indicators required to enter the sub-healthy zone, when hour , For the first The relative deviation of each indicator within its sub-healthy range; The relative deviation is calculated as follows: Relative deviation of predicted low-temperature impact toughness ; Relative deviation of vacuum loss rate ; Relative deviation of the hydrogen embrittlement sensitivity index ; Record the revised remaining safe service life in the inspection traceability report.

[0063] Specifically, when multiple indicators simultaneously enter or are within the sub-healthy range, a coupling correction coefficient is introduced. The remaining safe service life is adjusted. First, the relative deviation of each indicator within its sub-healthy range is calculated. : ; ; ; When the indicator value is in the sub-healthy range The range of values ​​is A larger value indicates that the index is closer to the danger boundary. Let the number of indicators entering the sub-healthy zone be n (1≤n≤3), then the coupling correction coefficient is defined as: ; when hour, Take the minimum value of 0.6. Record the corrected remaining safe service life as: ; in The remaining safe service life is calculated according to Example 4. Record it in the testing traceability report.

[0064] It also includes the steps for constructing composite health indicators; Construction of a composite health index: When a gas cylinder is in a sub-healthy state, a composite health index C is constructed as a quantitative representation of the overall health status of the gas cylinder. The calculation formula is as follows:

[0065] A dynamic covariate V(t) is introduced as an external environmental influencing factor. The dynamic covariate includes the rate of change of ambient temperature, the cumulative value of the service life of the gas cylinder, or the number of filling and depressurization cycles. The dynamic covariates are incorporated into the decay rate correction model to construct covariate correction factors. , where β is the covariate sensitivity coefficient, obtained by fitting historical detection data; The composite health index C is dynamically compensated using the aforementioned covariate correction factor: The overall health status value after compensation is obtained; when When the health status falls below the preset health threshold, an early warning message is output and the detection cycle is shortened accordingly. The compensated comprehensive health status value is recorded in the detection traceability report.

[0066] Specifically, when a gas cylinder is in a sub-healthy state, a composite health index C is constructed as a quantitative representation of the overall health status of the gas cylinder: ; This indicator integrates the current margin, decay rate, and coupling correction of three indicators. A dynamic covariate V(t) is introduced as an external environmental influence factor; V(t) can be selected as the rate of change of ambient temperature. (Unit: ℃ / month) Cumulative value of gas cylinder service life (Unit: Month) or Number of pressurization / depressurization cycles (Unit: times). Constructing covariate correction factors: ; in The sensitivity coefficients for covariates are obtained by fitting historical detection data, for example... When V(t) represents the service life, dynamic compensation for the composite health indicators is performed using covariate correction factors. The compensated comprehensive health status value is obtained, and a preset health threshold is set. ,Pick (Unit is the same as C). When When necessary, an early warning message is output and the detection cycle is shortened accordingly, with the shortening rules being the same as in Example 4. The compensated comprehensive health status value is recorded in the detection traceability report.

[0067] For example, after a temporary test in the 14th month, the gas cylinder was in a sub-healthy state, and all three indicators entered the sub-healthy range (n=3). Calculate the relative deviation of each indicator: ; ; ; Coupling correction coefficient: ; Since n=3, according to the rules... (Conservative value). Corrected remaining safe service life: One month (approximately 12.5 days).

[0068] Composite health indicators: ; Introducing dynamic covariates: Assuming the gas cylinder has been in service for 36 months, take V(t) = 36, and the sensitivity coefficient β = 0.02, then... Compensated overall health status value: .

[0069] Preset health threshold , No warning was triggered. However, considering more severe environments, such as drastic temperature fluctuations leading to a higher β value, a warning might be triggered. The compensated overall health status value of 0.855 was recorded in the testing traceability report.

[0070] This embodiment, based on Embodiment 5, adds multi-index coupling correction and dynamic covariate compensation, further improving the accuracy of lifespan prediction. The coupling correction coefficient considers both the number of indices and the degree of deviation, reflecting the accelerated impact of multi-index synergistic degradation on the overall lifespan of the cylinder. The introduction of dynamic covariates allows external environmental factors (such as service life and cycle count) to be included in the health status assessment. The exponential form of the covariate correction factor conforms to the nonlinear influence characteristics of environmental factors in material degradation. The composite health index integrates information from multiple dimensions into a single quantitative value, facilitating intuitive monitoring and early warning triggering. This represents a leap from independent assessment of single indices to dynamic assessment of coupled multi-indexes, providing support for refined and intelligent detection and management of liquid hydrogen cylinders after a fire.

[0071] This application also discloses a post-fire liquid hydrogen cylinder detection and management system.

[0072] A post-fire liquid hydrogen cylinder detection and management system includes: a processor, and a memory communicatively connected to the processor; The memory is provided with a computer-readable storage medium, and a computer program is stored on the computer-readable storage medium. When the processor processes the computer program stored on the computer-readable storage medium, it implements a method for detecting and managing liquid hydrogen cylinders after a fire.

[0073] The above are all preferred embodiments of this application, and are not intended to limit the scope of protection of this application. Therefore, all equivalent changes made in accordance with the structure, shape and principle of this application should be covered within the scope of protection of this application.

Claims

1. A method for detecting and managing liquid hydrogen cylinders after a fire, characterized in that: Includes the following steps: Preliminary assessment: Metallographic examination of the liquid hydrogen cylinder after the fire was conducted to determine whether the metallographic structure of the fire-affected and non-fire-affected areas was consistent, thus obtaining the first assessment result; Fire-affected area detection: If the first judgment result is yes, then perform magnetic multi-parameter detection on the fire-affected area to obtain magnetic multi-parameter characteristic signals; Toughness prediction: The magnetic multi-parameter characteristic signal is input into a pre-established low-temperature mechanical property multiple linear regression model to obtain the predicted value of the low-temperature impact toughness of the inner liner material. The low-temperature mechanical property multiple linear regression model is established based on the low-temperature impact toughness test data of damaged test blocks at -196℃ under different fire temperatures and the corresponding magnetic multi-parameter characteristic signal regression. Insulation assessment: Vacuum degree detection and infrared thermal imaging analysis are performed on the gas cylinder to obtain the vacuum loss rate as the result of the insulation layer integrity assessment; Safety determination: The predicted value of low temperature impact toughness is compared with the preset low temperature toughness threshold, and combined with the vacuum loss rate. If the predicted value of low temperature impact toughness is lower than the threshold or the vacuum loss rate exceeds the insulation failure threshold, the gas cylinder is determined to be in a dangerous state, and the safety handling procedure is initiated. Traceability and Filing: The test data, intermediate calculation results and judgment results obtained during the test are used to generate a test traceability report, which is stored to establish a full life cycle test file for gas cylinders.

2. The method for detecting and managing liquid hydrogen cylinders after a fire according to claim 1, characterized in that: It also includes a hydrogen embrittlement assessment step; Hydrogen embrittlement assessment: Electrochemical hydrogen permeation test or thermal desorption spectrum analysis is performed on the fire-affected part to determine the hydrogen diffusion coefficient and hydrogen trap density, and the hydrogen embrittlement sensitivity index is obtained; If the hydrogen embrittlement sensitivity index exceeds the preset hydrogen embrittlement threshold, the gas cylinder is determined to have a risk of hydrogen embrittlement, and this risk is recorded in the detection traceability report.

3. The method for detecting and managing liquid hydrogen cylinders after a fire according to claim 2, characterized in that: It also includes steps for determining sub-health status; Sub-health status assessment: Sub-health ranges were set for the predicted low-temperature impact toughness value, the vacuum loss rate, and the hydrogen embrittlement sensitivity index, respectively. When the predicted value of low temperature impact toughness, vacuum loss rate and hydrogen embrittlement sensitivity index are not in the danger range, but at least one of them enters the corresponding sub-health range, the gas cylinder is judged to be in a sub-healthy state. The results of the sub-health status assessment will be recorded in the testing and traceability report.

4. The method for detecting and managing liquid hydrogen cylinders after a fire, as described in claim 3, is characterized in that: It also includes a cycle setting step; Periodic setting: When the gas cylinder is in a sub-healthy state, calculate the predicted value of low temperature impact toughness, vacuum loss rate, and hydrogen embrittlement sensitivity index relative to their respective danger boundaries. The minimum value among the remaining safety margins is taken as the dominant margin, and the dynamic detection period is set according to the interval in which the dominant margin is located; Record the set testing cycle in the testing traceability report.

5. The method for detecting and managing liquid hydrogen cylinders after a fire according to claim 3, characterized in that: It also includes a fault prediction step; Fault prediction: When the gas cylinder is in a sub-healthy state, the current and historical detection data are input into the pre-established fault prediction model to predict the time when each indicator will enter a dangerous state within a preset time window in the future. The minimum predicted time to enter a dangerous state for each indicator is taken as the predicted dangerous time for the gas cylinder. ; when When the risk level is below the preset risk threshold, an early warning message is output and the detection cycle is shortened accordingly. The prediction results are recorded in the detection traceability report.

6. The method for detecting and managing liquid hydrogen cylinders after a fire according to claim 3, characterized in that: When a gas cylinder is in a sub-healthy state, based on the current values ​​of each indicator and its historical decay rate, the remaining time for each indicator to reach the danger boundary is predicted, and the minimum value is taken as the remaining safe service life of the gas cylinder. The calculation formula is as follows: ； in , , These are the predicted values ​​of low-temperature impact toughness, vacuum loss rate, and average decay rates of hydrogen embrittlement sensitivity index, respectively, obtained by fitting historical test data. Record the remaining safe service life prediction results in the inspection traceability report.

7. The method for detecting and managing liquid hydrogen cylinders after a fire according to claim 6, characterized in that: It also includes new adjustment steps; New adjustment: In subsequent re-examinations, if a new indicator enters the sub-health range for the first time, or if an existing indicator moves from a lower-level sub-health range to a higher-level sub-health range, dynamic adjustment of life expectancy prediction will be triggered. Recalculate the remaining safety margin and decay rate of each current indicator, and update the predicted remaining safe service life. ; like Then Update the remaining safe service life of the gas cylinder, and according to Reset the testing cycle; The adjustment process and the updated prediction results will be recorded in the testing traceability report.

8. The method for detecting and managing liquid hydrogen cylinders after a fire according to claim 7, characterized in that: It also includes a lifetime prediction correction step; Life expectancy prediction correction: When multiple indicators simultaneously enter or are within the sub-healthy range, a coupling correction coefficient is introduced. The remaining safe service life is adjusted using the following formula: ;in , The number of indicators required to enter the sub-healthy zone, when hour , For the first The relative deviation of each indicator within its sub-healthy range; The relative deviation is calculated as follows: Relative deviation of predicted low-temperature impact toughness ; Relative deviation of vacuum loss rate ; Relative deviation of the hydrogen embrittlement sensitivity index ; Record the revised remaining safe service life in the inspection traceability report.

9. The method for detecting and managing liquid hydrogen cylinders after a fire according to claim 4, characterized in that: It also includes the steps for constructing composite health indicators; Construction of a composite health index: When a gas cylinder is in a sub-healthy state, a composite health index C is constructed as a quantitative representation of the overall health status of the gas cylinder. The calculation formula is as follows: ； A dynamic covariate V(t) is introduced as an external environmental influencing factor. The dynamic covariate includes the rate of change of ambient temperature, the cumulative value of the service life of the gas cylinder, or the number of filling and depressurization cycles. The dynamic covariates are incorporated into the decay rate correction model to construct covariate correction factors. , where β is the covariate sensitivity coefficient, obtained by fitting historical detection data; The composite health index C is dynamically compensated using the aforementioned covariate correction factor: The overall health status value after compensation is obtained; when When the health status falls below the preset health threshold, an early warning message is output and the detection cycle is shortened accordingly. The compensated comprehensive health status value is recorded in the detection traceability report.

10. A post-fire liquid hydrogen cylinder detection and management system, characterized in that: include: A processor, and a memory communicatively connected to the processor; The memory is provided with a computer-readable storage medium, and a computer program is stored on the computer-readable storage medium. When the processor processes a computer program stored on the computer-readable storage medium, it implements the method as described in any one of claims 1-9.

Images