A method and system for baseline construction and monitoring calibration of multilayer embedded FBG for composite hydrogen storage cylinders

By embedding a multi-layer FBG network in a composite hydrogen storage cylinder, establishing measurement point identification codes and matching peak wavelengths before and after curing, calculating manufacturing residual response parameters, and generating a baseline dataset, the problem of aliasing of FBG wavelength changes during the manufacturing stage was solved, enabling accurate load response calibration and structural health monitoring.

CN123104100APending Publication Date: 2026-11-17SPECIAL EQUIP SAFETY SUPERVISION INSPECTION INST OF JIANGSU PROVINCE
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611126106.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-07-28
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

Existing fiber optic sensing technology for composite hydrogen storage cylinders fails to effectively record wavelength changes before and after FBG curing during the manufacturing stage, resulting in the overlap of manufacturing residual response and pressure load response, affecting strain evaluation and life assessment, and making it impossible to distinguish between initial non-uniformity in the manufacturing process and new responses caused by subsequent loads or damage.

Method used

A multi-layer embedded FBG network is adopted. By embedding sensing optical fibers in the cylinder winding layer, the measurement point identification code is established, the peak wavelengths before and after curing are collected, the manufacturing residual response parameters are calculated, the manufacturing baseline dataset is generated, and the load response calibration is performed with the cured wavelength as the zero reference.

Benefits of technology

It achieves separation of manufacturing residual response and load response, provides traceable quality assessment, ensures the accuracy and reliability of subsequent monitoring, and can provide reliable structural condition monitoring during hydrostatic testing, burst testing, pressure cycling, and service.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123104100A_ABST
    Figure CN123104100A_ABST
Patent Text Reader

Abstract

This invention discloses a method and system for constructing and monitoring a manufacturing baseline for multilayer embedded fiber optic cables (FBGs) used in composite hydrogen storage cylinders. The method embeds multiple sets of FBG sensing fibers within the cylinder's winding layer, establishing an identification code that includes the cylinder number, fiber group number, path type, thickness direction layer position, and measurement point sequence number. Peak wavelengths before curing and baseline peak wavelengths after curing are collected and paired according to the identification code. Curing wavelength migration, temperature correction, curing equivalent strain, and layer / path statistical parameters are calculated to generate a manufacturing baseline dataset bound to a single cylinder and evaluate baseline reliability. Subsequent pressure monitoring uses the peak wavelength of the cured baseline as a zero reference, subtracting manufacturing residual response and temperature drift to obtain the load response. This invention can separate the manufacturing residual response from the subsequent load response, improving the traceability and calibration accuracy of embedded FBG monitoring data.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention belongs to the field of composite material pressure vessel manufacturing process monitoring, fiber optic sensing and structural health monitoring technology, specifically relating to a method and system for constructing and calibrating a manufacturing baseline for a multilayer embedded FBG used in composite material hydrogen storage cylinders. Background Technology

[0002] With the development of hydrogen-powered transportation, mobile hydrogen supply equipment, and high-pressure storage and transportation equipment, Type IV composite material hydrogen storage cylinders, due to their lightweight, corrosion resistance, and high hydrogen storage efficiency, are gradually becoming important pressure-bearing components in on-board hydrogen storage systems with pressure ratings of 70 MPa and higher. These cylinders typically consist of a plastic inner liner, a metal neck, and a carbon fiber / resin composite winding layer. The composite winding layer bears the main internal pressure load, and its manufacturing quality and internal residual condition directly affect subsequent hydrostatic tests, burst tests, pressure cycling tests, and structural response during service.

[0003] Existing fiber optic monitoring technologies for composite hydrogen storage cylinders primarily focus on strain monitoring during hydrogen filling and discharging, pressure cycling, bursting, or service operations, typically using the nominal initial wavelength of the embedded FBG, the pre-implantation wavelength, or the wavelength at the start of the test as a zero reference. However, the embedded FBG undergoes resin crosslinking shrinkage, thermal expansion coefficient mismatch, interlayer compaction, localized resin flow, and sensor-matrix interface coupling changes during cylinder winding, curing, and cooling. The peak wavelength of the cured FBG is not a strain-free state, but rather a state after the manufacturing residual response has been written. If this manufacturing stage baseline is ignored, the manufacturing residual spectral response will be mixed with the pressure-induced spectral response in subsequent pressure load response interpretation, thus affecting strain evaluation, anomaly identification, and lifespan determination.

[0004] On the other hand, Type IV composite hydrogen storage cylinders are not homogeneous laminated structures; their winding layers include circumferential layers, low-angle spiral layers, end cap turning regions, and cylinder-end cap transition regions. The FBG curing response may vary significantly depending on the path and thickness direction. Simply recording the wavelength change before and after curing at a single measurement point cannot establish a data system suitable for overall cylinder quality evaluation and subsequent monitoring and calibration; similarly, simply judging the strain threshold during service cannot distinguish between initial non-uniformity formed during manufacturing and the additional response caused by subsequent loads or damage.

[0005] Therefore, there is an urgent need for a manufacturing baseline construction method and system for multilayer embedded FBG networks in Type IV composite hydrogen storage cylinders. This method should be able to complete the following during the manufacturing stage: measurement point identification encoding, peak value pairing before and after curing, abnormal jump identification, layer and path statistics, and baseline reliability evaluation. Furthermore, it should use the cured wavelength as a zero reference in subsequent pressure monitoring, thereby achieving separation of manufacturing residual response and load response. Summary of the Invention

[0006] To address the aforementioned issues, this invention discloses a method and system for constructing and monitoring a manufacturing baseline for multilayer embedded FBG used in composite material hydrogen storage cylinders.

[0007] To achieve the above objectives, the technical solution of the present invention is as follows:

[0008] A method for establishing and monitoring calibration of a manufacturing baseline for multilayer embedded FBG for composite hydrogen storage cylinders includes the following steps:

[0009] S1. Multiple sets of sensing optical fibers containing FBG measurement points are embedded in the composite material winding layer of the gas cylinder;

[0010] S2. Establish measurement point identification codes for each FBG measurement point of the sensing fiber;

[0011] S3. Acquire the peak wavelength λ of each FBG before curing, after the sensing fiber is wound and cured. pre,j And the baseline peak wavelength λ of each FBG after curing and cooling to room temperature. base,j ;

[0012] S4. Pair the same measurement point's λ according to the FBG measurement point identity code. pre,j and λ base,j And calculate the manufacturing residual response parameters;

[0013] S5. Generate a manufacturing baseline dataset that is bound to the cylinder number and FBG measurement point identification code, and evaluate the baseline reliability.

[0014] S6. Call λ in subsequent pressure monitoring. base,j The load response is calibrated using the zero reference wavelength as the corresponding FBG measurement point.

[0015] As a preferred technical solution of the present invention, in step S1, the sensing optical fiber is arranged in at least two types of regions among the circumferential layer, spiral layer, end cap region and cylinder-end cap transition region of the gas cylinder, and covers at least two thickness direction layers.

[0016] As a preferred embodiment of the present invention, the measurement point identification code in step S2 is ID. j For: ID j ={B, G, R, L, k}; where B is the gas cylinder number, G is the fiber group, R is the winding path type, L is the thickness direction layer, and k is the measurement point number.

[0017] As a preferred technical solution of the present invention, the residual response parameters in step S4 include at least the curing wavelength migration and the curing equivalent strain.

[0018] When the room temperature is the same before and after the sensing fiber is cured, the curing wavelength shift Δλ cure,jfor:

[0019]

[0020] At this time, the equivalent strain of curing ε cure,j for:

[0021]

[0022] When the room temperature is different before and after the sensing fiber is cured, the curing wavelength shift Δλ cure,j* for:

[0023]

[0024] At this time, the equivalent strain of curing ε cure,j for:

[0025]

[0026] K ε,j Let K be the strain sensitivity coefficient of the j-th FBG measuring point. T,j Let T be the temperature sensitivity coefficient of the j-th FBG measuring point. base The temperature measured after curing, T pre The temperature was measured before curing.

[0027] As a preferred technical solution of the present invention, in step S4, the missing λ is addressed... pre,j or λ base,j The measurement points and the measurement points whose wavelength migration exceeds the abnormal threshold are marked with an abnormal status. Measurement points marked with an abnormal status do not participate in the layer or path statistics, but are retained in the manufacturing baseline dataset.

[0028] like The measurement point where the wavelength migration exceeds the abnormal threshold is the measurement point where the wavelength migration exceeds the abnormal threshold. This represents the median of the wavelength shift for all candidate paired measurement points. γ is the median absolute deviation, and γ is the anomaly detection factor. The minimum anomaly threshold is denoted by 'max', which indicates that the larger of the two thresholds is used as the anomaly threshold.

[0029] As a preferred embodiment of the present invention, the baseline dataset generated in step S5 includes single-point baseline records, and the single-point baseline records include at least the measurement point identification code and λ. pre,j , λ base,j The residual response parameters, acquisition temperature, measurement point status identifier, and baseline recall identifier are generated.

[0030] As a preferred embodiment of the present invention, in step S5, the effective measurement points retained after anomaly identification are statistically analyzed for layer position and path, wherein the average curing equivalent strain μ of the layer position is calculated according to the thickness direction. L Stratigraphic standard deviation σ L and thickness gradient index G th The path-average curing equivalent strain μ is calculated according to the winding path type or fiber group. R Path standard deviation σ R and path variation coefficient CV R And use them as the layer baseline parameters and path baseline parameters in the manufacturing baseline dataset; where μ outer To compensate for the equivalent strain of the outer layer after average curing, μ inner This is to compare the equivalent deformation of the inner layer's average curing.

[0031] in μ R and σ R These represent the mean and standard deviation of the curing response for the same path type or the same fiber group.

[0032] As a preferred embodiment of the present invention, the baseline reliability coefficient in step S5 when evaluating baseline reliability is:

[0033]

[0034] in, N is the baseline reliability coefficient for the b-th gas cylinder; out N represents the number of measurement points with abnormal jumps. paired CV represents the number of pairable measurement points before and after curing. global Let be the coefficient of variation of the solidification response at all effective measurement points in the bottle, where α, β, and χ are non-negative weighting coefficients and α+β+χ=1.

[0035] As a preferred embodiment of the present invention, the load response after manufacturing baseline calibration is obtained in the subsequent pressure monitoring stage of step S6 according to the following formula:

[0036]

[0037]

[0038] in, Let T(t) be the peak wavelength of the j-th FBG measuring point at time t, and let T(t) be the temperature at time t, K. ε,j Let K be the strain sensitivity coefficient of the j-th FBG measuring point. T,j Let T be the temperature sensitivity coefficient of the j-th FBG measuring point. base The temperature measured after curing is Δλ. load,j(t)*ε is the load-induced wavelength change after accounting for the effects of manufacturing baseline and temperature. load,j(t) To manufacture the equivalent strain of the load after baseline calibration.

[0039] A system for constructing and monitoring the manufacturing baseline of a multilayer embedded FBG for composite hydrogen storage cylinders includes a composite hydrogen storage cylinder body 1, an embedded FBG sensor network 2, a spectral demodulation module 3, a measurement point identification management module 4, a pre-curing acquisition module 5, a post-curing acquisition module 6, a peak pairing module 7, an anomaly identification module 8, a manufacturing baseline generation module 9, a baseline reliability evaluation module 10, and a load response calibration module 11. The manufacturing baseline generation module 9 is used to generate the manufacturing baseline dataset, and the load response calibration module 11 is used to... base,j To produce a baseline-calibrated load response for zero-reference wavelength output.

[0040] The beneficial effects of this invention are as follows:

[0041] First, the peak wavelength after curing is defined as the manufacturing baseline for the measurement point to avoid the overlap of manufacturing residual response and subsequent pressure response. Second, traceable management of multi-layer, multi-path, and multi-measurement point FBG data is achieved through identity coding and pairing tracking. Third, curing quality and sensor network effectiveness are quantified through layer gradient, path consistency, and baseline reliability indicators. Fourth, the manufacturing baseline dataset is bound to the unique number of the gas cylinder, which can be used throughout the entire process of water pressure, burst, pressure cycle, and service monitoring.

[0042] This invention can separate the residual spectral response caused by resin curing shrinkage, thermal shrinkage mismatch, interlayer compaction, and sensor-matrix coupling during the manufacturing stage from the subsequent load response, and can provide traceable evaluation basis for curing quality, sensor network effectiveness, batch consistency, and subsequent structural condition monitoring. Attached Figure Description

[0043] Figure 1 This is a flowchart of the baseline construction and monitoring calibration method in an embodiment of the present invention;

[0044] Figure 2 This is a schematic diagram of a multilayer embedded FBG sensing structure in an embodiment of the present invention;

[0045] Figure 3 This is an example diagram illustrating the relationship between the peak wavelength before curing and the migration amount of the curing wavelength in an embodiment of the present invention;

[0046] Figure 4 This is an example diagram of the equivalent strain distribution of different fiber groups during curing in an embodiment of the present invention;

[0047] Figure 5 This is an example diagram of the equivalent strain distribution of curing in different thickness directions in an embodiment of the present invention. Detailed Implementation

[0048] The present invention will be further illustrated below with reference to the accompanying drawings and specific embodiments. It should be understood that the following specific embodiments are for illustrative purposes only and are not intended to limit the scope of the invention.

[0049] Please see Figure 1-5 This invention is based on the peak wavelength changes of embedded FBG before and after curing and in the subsequent pressure monitoring stage. It sequentially completes the identification tracking of measuring points, peak pairing before and after curing, calculation of manufacturing residual response, statistics of layer and path, generation of manufacturing baseline and calibration of load response, thereby forming a manufacturing baseline construction and monitoring calibration method suitable for composite material hydrogen storage cylinders.

[0050] like Figure 1 As shown, a method for establishing and monitoring the manufacturing baseline of a multilayer embedded FBG for composite hydrogen storage cylinders includes the following steps:

[0051] S1. Multiple sets of sensing optical fibers containing FBG measurement points are embedded in the composite material winding layer of the gas cylinder;

[0052] S2. Establish measurement point identification codes for each FBG measurement point of the sensing fiber;

[0053] S3. Acquire the peak wavelength λ of each FBG before curing, after the sensing fiber is wound and cured. pre,j And the baseline peak wavelength λ of each FBG after curing and cooling to room temperature. base,j ;

[0054] S4. Pair the same measurement point's λ according to the FBG measurement point identity code. pre,j and λ base,j And calculate the manufacturing residual response parameters;

[0055] S5. Generate a manufacturing baseline dataset that is bound to the cylinder number and FBG measurement point identification code, and evaluate the baseline reliability.

[0056] S6. Call λ in subsequent pressure monitoring. base,j The load response is calibrated using the zero reference wavelength as the corresponding FBG measurement point.

[0057] In step S1, the sensing optical fiber is arranged in at least two types of regions of the gas cylinder, including the circumferential layer, spiral layer, end cap region, and cylinder body-end cap transition region, and covers at least two thickness direction layers.

[0058] In step S2, the identification code of the measurement point is ID. j For: ID j={B, G, R, L, k}; where B is the gas cylinder number, G is the fiber group, R is the winding path type, L is the thickness direction layer, and k is the measurement point number. This identification code allows for unique tracking of data from the same measurement point before, after, and during subsequent pressure monitoring stages.

[0059] The residual response parameters manufactured in step S4 include at least the curing wavelength shift and the curing equivalent strain;

[0060] When the room temperature is the same before and after the sensing fiber is cured, the curing wavelength shift Δλ cure,j for:

[0061]

[0062] At this time, the equivalent strain of curing ε cure,j for:

[0063]

[0064] When the room temperature is different before and after the sensing fiber is cured, the curing wavelength shift Δλ cure,j* for:

[0065]

[0066] At this time, the equivalent strain of curing ε cure,j for:

[0067]

[0068] K ε,j Let K be the strain sensitivity coefficient of the j-th FBG measuring point. T,j Let T be the temperature sensitivity coefficient of the j-th FBG measuring point. base The temperature measured after curing, T pre This is the temperature measured before curing. (K) ε,j Obtained through strain calibration, several levels of known axial strain were applied to the j-th FBG under isothermal conditions, and the change in peak wavelength was recorded. The slope of the linear fitting of Δλ-ε was used as K. ε,j The unit is usually pm / με; verified manufacturer or batch calibration values ​​can also be used. In the embodiments of this invention, K... ε,j A value of 1.2 pm / με was used, and calibration batches or calibration records were written into the manufacturing baseline dataset. K T,j Obtained from temperature calibration, the j-th FBG was subjected to graded temperature increases and decreases under no external load or constant strain conditions. The change in peak wavelength with temperature was recorded, and the slope of the linear fitting of Δλ-T was used as K. T,j The unit is usually pm / ℃. For embedded FBGs, it is preferable to use the calibration value after packaging or the calibration value of the same structure compensation grating, and specify the calibration temperature range and fitting error.

[0069] If Δλ cure,j or Δλ cure,j* If <0, it indicates that the peak value shifts towards the shortwave direction after solidification.

[0070] In step S4, for the missing λ pre,j or λ base,j An abnormal status indicator is set for the measurement points and the measurement points whose wavelength migration exceeds the abnormal threshold. Measurement points with abnormal status indicators are not included in the layer or path statistics, but are retained in the manufacturing baseline dataset.

[0071] Step S5 involves creating a baseline dataset that includes single-point baseline records. Each single-point baseline record includes at least the measurement point identification code and λ. pre,j , λ base,j The data includes manufacturing residual response parameters, acquisition temperature, measurement point status indicators, and baseline recall indicators. Manufacturing residual response parameters include Δλ. cure,j Temperature-corrected Δλ cure,j* and ε cure,j The status identifiers for measurement points should include at least valid, unpaired, abnormal jump, and invalid; the baseline retrieval identifiers should include at least directly retrievalable, requiring verification, and retrieval not retrievalable. The status identifiers are used for single-point data quality management, and the retrieval identifiers are used to determine whether the baseline of the measurement point or the gas cylinder can proceed to subsequent monitoring and calibration.

[0072] In step S5, the effective measurement points retained after anomaly identification are statistically analyzed for layer and path. Among them, the average curing equivalent strain μ of the layer is calculated according to the thickness direction. L Stratigraphic standard deviation σ L and thickness gradient index G th The path-average curing equivalent strain μ is calculated according to the winding path type or fiber group. R Path standard deviation σ R and path variation coefficient CV R They are used as the layer baseline parameters and path baseline parameters in the manufacturing baseline dataset.

[0073] The mean curing equivalent strain and standard deviation of L / P at any layer are as follows:

[0074]

[0075]

[0076] Where, n L This represents the number of valid measurement points within layer L. (Using μ) L and σ L It can evaluate the mean and dispersion of manufacturing residual response at different thickness directions.

[0077] The stratigraphic gradient index is:

[0078]

[0079] Where, μ outer To compensate for the equivalent strain of the outer layer after average curing, μ inner The average curing equivalent strain of the innermost layer is considered. The outermost layer between any two fiber layers is considered the outermost layer, and the innermost layer is considered the innermost layer. G th Used to represent the trend and magnitude of the change in the solidification residual response from the inside out.

[0080] The path consistency metric is:

[0081]

[0082] Where, μ R and σ R These represent the mean and standard deviation of the curing response for the same path type or the same fiber group, respectively. CV R CV is a path consistency or intra-path dispersion index, representing the ratio of the standard deviation of the solidified equivalent strain to the absolute value of its mean within the same path type or fiber group. R A larger value indicates that the residual response is more unevenly generated within the path or fiber group.

[0083] In step S5, the baseline confidence coefficient is used to evaluate baseline confidence.

[0084]

[0085] Where, N out N represents the number of measurement points with abnormal jumps. paired CV represents the number of pairable measurement points before and after curing. global The coefficient of variation of the curing response at all effective measurement points in the bottle is calculated by measuring the CV between multiple sets of optical fibers. R The arithmetic mean was then taken to obtain α, β, and χ, which are non-negative weighting coefficients and α+β+χ=1.

[0086] In step S6, the subsequent pressure monitoring stage obtains the load response after manufacturing baseline calibration using the following formula:

[0087]

[0088]

[0089] in, Let T(t) be the peak wavelength of the j-th FBG measuring point at time t, and let T(t) be the temperature at time t, K. ε,j Let K be the strain sensitivity coefficient of the j-th FBG measuring point. T,j Let T be the temperature sensitivity coefficient of the j-th FBG measuring point. baseThe temperature measured after curing is Δλ. load,j(t)* ε is the load-induced wavelength change after accounting for the effects of manufacturing baseline and temperature. load,j(t) To manufacture the equivalent strain of the load after baseline calibration.

[0090] like Figure 2 As shown, a system for constructing and monitoring the manufacturing baseline of a multilayer embedded FBG for composite hydrogen storage cylinders includes a composite hydrogen storage cylinder body 1, an embedded FBG sensor network 2, a spectral demodulation module 3, a measurement point identification management module 4, a pre-curing acquisition module 5, a post-curing acquisition module 6, a peak pairing module 7, an anomaly identification module 8, a manufacturing baseline generation module 9, a baseline reliability evaluation module 10, and a load response calibration module 11. The manufacturing baseline generation module 9 is used to generate the manufacturing baseline dataset, and the load response calibration module 11 is used to... base,j A baseline-calibrated load response is generated for the zero-reference wavelength output. The embedded FBG sensor network 2 comprises at least two layers of optical fiber, for example... Figure 2 In the embodiment, an outer optical fiber 12, a middle optical fiber 13, and an inner optical fiber 14 are provided.

[0091] The Type IV composite hydrogen storage cylinder body 1 includes a plastic inner liner, a metal nozzle, and a carbon fiber / epoxy resin composite winding layer. An embedded FBG sensor network 2 is simultaneously implanted during the fiber winding process, preferably including sensing optical fibers located in the circumferential layer of the cylindrical section, sensing optical fibers located in the spiral layer, and sensing optical fibers located in the relevant area of ​​the end cap.

[0092] In one specific embodiment, each gas cylinder is equipped with 46 FBG measurement points, and the three gas cylinders are equipped with a total of 138 FBG measurement points; the optical fibers in the circumferential layer, spiral layer, and end cap region are represented by C, H, and F, respectively, and each measurement point is numbered P1 to P6 or P1 to P5 along the optical fiber path. According to the position in the thickness direction, some optical fiber groups can be classified into inner layer, middle layer, and outer layer.

[0093] The measurement point identity management module 4 generates an identity ID for each measurement point. j ={B, G, R, L, k}. For example, B represents the gas cylinder number such as KJ002, KJ003, or KJ004; G represents the fiber optic group such as C1, H1, C2, H2, F1, C3, H3, or F2; R represents the circumferential path C, the spiral path H, or the end cap path F; L represents the innermost, middle, or outermost layer, numbered sequentially from the inside out, starting from L1; k represents the measurement point number, numbered along the fiber optic path starting from P1. For example, in this embodiment, a certain measurement point is ID1={KJ002, C1, C, L1, P1}.

[0094] like Figure 3As shown, the pre-curing acquisition module 5 acquires the first peak wavelength λ after the winding is completed and before entering the heat curing stage. pre,j After curing, the acquisition module 6 acquires the second peak wavelength λ after the gas cylinder has completed thermal curing and cooled to room temperature. base,j Peak pairing module 7 only pairs λ with consistent identity codes. pre,j and λ base,j Pairing is performed to avoid mismatch between different channels or different gratings. Figure 3 This shows the majority of the curing wavelength migration Δλ. cure A negative Δλ indicates that the peak value shifts towards the shorter wavelength direction after curing relative to before curing. If the temperature effect has been subtracted or the sampling temperature is consistent, then the negative Δλ... cure Typically, this corresponds to the equivalent compressive response along the FBG axis, with a positive Δλ. cure This usually corresponds to the equivalent tensile response; however, in embedded composite materials, this symbol reflects the "equivalent strain" under the combined effects of resin curing shrinkage, thermal shrinkage mismatch, interlayer compaction, and interfacial coupling, and should not be simply expressed as a single mechanical extrusion or tension.

[0095] The anomaly detection module 8 identifies missing data and anomalous data jumps. If the measurement point lacks λ... pre,j or λ base,j If, then mark as an unpaired measuring point; if

[0096]

[0097] These are marked as anomalous jump measurement points. Anomalous jump measurement points are not included in the mean, standard deviation, and stratigraphic gradient statistics, but are retained as manufacturing process records. This represents the median of the wavelength shift for all candidate paired measurement points. γ is the median absolute deviation, and γ is the anomaly detection factor. The minimum anomaly threshold is defined as γ, where γ represents the larger of the two thresholds. γ is the anomaly detection factor based on MAD (Magnitude of Determination), which is not directly obtained from a single measurement point but is preferably based on Δλ from historical qualified batches. cure The distribution is determined through verification using the false positive / false negative rate. When historical samples are unavailable, a robust statistical empirical value can be initially selected and adjusted during process validation. Approximately 4.5 can be used as an initial example value close to the 3σ criterion, but it should not be written as a unique fixed value when validation data is lacking. λ min It is determined by the lowest resolvable change of the detection system and by the demodulator model.

[0098] Manufacturing baseline generation module 9 calculation and according to Convert the curing equivalent strain. For data acquisition conditions with different temperature requirements, first... Perform temperature correction, and according to Convert the equivalent strain of curing. For example... Figure 4 As shown, after obtaining the equivalent strain of single-point curing, the baseline distribution can be statistically analyzed according to the fiber groups or winding paths such as C, H, and F, and the path consistency index can be calculated.

[0099] In one embodiment, among the 138 designed FBG measurement points across the three gas cylinders, effective peak values ​​could be identified after curing. After removing missing data and abnormal jumps, 131 effective paired measurement points before and after curing were obtained, accounting for 94.9% of the total designed measurement points. The average curing wavelength shift of all effective measurement points was -0.395 nm, with a standard deviation of 0.214 nm; according to 1.2 The strain sensitivity conversion yielded an average curing equivalent effect of -329 με and a standard deviation of 178 με. The original data for the above processed data are sourced from Table 1.

[0100] Table 1 Original Data Table of Implementation Examples

[0101]

[0102] Furthermore, such as Figure 5 As shown, each valid measurement point is assigned to the inner layer (e.g., inner fiber 14), middle layer (e.g., middle fiber 13), or outer layer (e.g., outer fiber 12) according to the layer L in the identity code, and the calculations for each group are performed accordingly. The arithmetic mean of the values ​​shows that the average equivalent curing strains of the inner, middle, and outer layers are approximately -213 με, -354 με, and -389 με, respectively, indicating that the residual curing response increases from the inside to the outside along the thickness direction. Some spiral fiber assemblies exhibit strong responses and low dispersion, making them a preferred path for evaluating manufacturing consistency. However, some circumferential layers or end cap-related paths show greater dispersion, which can be used to indicate localized resin enrichment, fiber microbending, lead-out disturbances, or uneven interlayer compaction. The bluer the color, the greater the equivalent compressive strain amplitude, and regions such as C2, F1, C3, H3, and F2 exhibit stronger negative responses or localized differences.

[0103] Baseline credibility assessment module 10 based on Calculate the baseline reliability coefficient. When Q b When the value is above the available threshold, the gas cylinder manufacturing baseline can be directly used for subsequent monitoring; when Q b When in the verification range, a prompt will appear to verify abnormal measurement points or local paths; when Q b If the value falls below the unretrievable threshold, it indicates that the cylinder manufacturing baseline is unsuitable as a zero reference for subsequent monitoring. Based on engineering experience, the retrievable threshold is 0.8, the review range is 0.6-0.8, and the unretrievable threshold is 0.6.

[0104] The load response calibration module 11 reads the manufacturing baseline dataset corresponding to the unique number of the gas cylinder during subsequent hydrostatic loading, burst test, pressure cycle test, or service monitoring, and uses λ base,j As the first Zero reference for each FBG measurement point. Monitoring wavelength λ at time t. j(t) ,calculate , If temperature compensation is set, further calculations are required. .

[0105] Through the above process, the subsequent load response no longer mixes with the manufacturing residual spectral response formed during the curing stage, thus enabling a more accurate evaluation of structural strain under pressure load, differences in load transfer between different layers, and local anomalies.

[0106] In other embodiments, the number of sensing optical fibers, the number of FBG measurement points on each optical fiber, the winding angle, the layer division method, and the temperature compensation method can be adjusted according to the cylinder volume, pressure rating, winding process, and monitoring target; the method is not limited to 62 L and 70 MPa cylinders, but can also be used for other types of Type IV composite material pressure vessels.

[0107] In other embodiments, the free-state wavelength λ of the FBG can also be acquired before curing. free This is used to compare spectral changes in three states: free state, pre-curing constrained state, and post-curing baseline state; however, subsequent load response calibration still uses λ. base,j Used as a zero reference.

[0108] It should be noted that the above content merely illustrates the technical concept of the present invention and should not be construed as limiting the scope of protection of the present invention. For those skilled in the art, various improvements and modifications can be made without departing from the principle of the present invention, and all such improvements and modifications fall within the scope of protection of the claims of the present invention.

Claims

1. A method for baseline construction and monitoring calibration of multilayer embedded FBG manufacturing for composite hydrogen storage cylinders, characterized in that, Includes the following steps: S1. Multiple sets of sensing optical fibers containing FBG measurement points are embedded in the composite material winding layer of the gas cylinder; S2. Establish measurement point identification codes for each FBG measurement point of the sensing fiber; S3. Acquire the peak wavelength λ of each FBG before curing, after the sensing fiber is wound and cured. pre,j And the baseline peak wavelength λ of each FBG after curing and cooling to room temperature. base,j ; S4. Pair the same measurement point's λ according to the FBG measurement point identity code. pre,j and λ base,j And calculate the manufacturing residual response parameters; S5. Generate a manufacturing baseline dataset that is bound to the cylinder number and FBG measurement point identification code, and evaluate the baseline reliability. S6. Call λ in subsequent pressure monitoring. base,j The load response is calibrated using the zero reference wavelength as the corresponding FBG measurement point.

2. The method for baseline construction and monitoring calibration of multilayer embedded FBG manufacturing for composite hydrogen storage cylinders according to claim 1, characterized in that, In step S1, the sensing optical fiber is arranged in at least two types of regions of the gas cylinder, including the circumferential layer, spiral layer, end cap region, and cylinder body-end cap transition region, and covers at least two thickness direction layers.

3. The method for baseline construction and monitoring calibration of multilayer embedded FBG manufacturing for composite hydrogen storage cylinders according to claim 1, characterized in that, In step S2, the identification code of the measurement point is ID. j For: ID j ={B, G, R, L, k}; Where B is the gas cylinder number, G is the fiber group, R is the winding path type, L is the layer in the thickness direction, and k is the measurement point number.

4. The method for baseline construction and monitoring calibration of multilayer embedded FBG manufacturing for composite hydrogen storage cylinders according to claim 1, characterized in that, The residual response parameters manufactured in step S4 include at least the curing wavelength shift and the curing equivalent strain; When the room temperature is the same before and after the sensing fiber is cured, the curing wavelength shift Δλ cure,j for: At this time, the equivalent strain of curing ε cure,j for: When the room temperature is different before and after the sensing fiber is cured, the curing wavelength shift Δλ cure,j* for: At this time, the equivalent strain of curing ε cure,j for: K ε,j Let K be the strain sensitivity coefficient of the j-th FBG measuring point. T,j Let T be the temperature sensitivity coefficient of the j-th FBG measuring point. base The temperature measured after curing, T pre The temperature was measured before curing.

5. The method for baseline construction and monitoring calibration of multilayer embedded FBG manufacturing for composite hydrogen storage cylinders according to claim 4, characterized in that, In step S4, for the missing λ pre,j or λ base,j The measurement points and the measurement points whose wavelength migration exceeds the abnormal threshold are marked with an abnormal status. Measurement points marked with an abnormal status do not participate in the layer or path statistics, but are retained in the manufacturing baseline dataset. like The measurement point where the wavelength migration exceeds the abnormal threshold is the measurement point where the wavelength migration exceeds the abnormal threshold. This represents the median of the wavelength shift for all candidate paired measurement points. γ is the median absolute deviation, and γ is the anomaly detection factor. The minimum anomaly threshold is denoted by 'max', which indicates that the larger of the two thresholds is used as the anomaly threshold.

6. The method for baseline construction and monitoring calibration of multilayer embedded FBG manufacturing for composite hydrogen storage cylinders according to claim 1, characterized in that, Step S5 involves creating a baseline dataset that includes single-point baseline records. Each single-point baseline record includes at least the measurement point identification code and λ. pre,j , λ base,j The residual response parameters, acquisition temperature, measurement point status identifier, and baseline recall identifier are generated.

7. The method for baseline construction and monitoring calibration of multilayer embedded FBG manufacturing for composite hydrogen storage cylinders according to claim 1, characterized in that, In step S5, the effective measurement points retained after anomaly identification are statistically analyzed for layer and path. Among them, the average curing equivalent strain μ of the layer is calculated according to the thickness direction. L Stratigraphic standard deviation σ L and thickness gradient index G th The path-average curing equivalent strain μ is calculated according to the winding path type or fiber group. R Path standard deviation σ R and path variation coefficient CV R And use them as the layer baseline parameters and path baseline parameters in the manufacturing baseline dataset; in μ outer To compensate for the equivalent strain of the outer layer after average curing, μ inner This is to compare the equivalent deformation of the inner layer's average curing. in μ R and σ R These represent the mean and standard deviation of the curing response for the same path type or the same fiber group.

8. The method for baseline construction and monitoring calibration of multilayer embedded FBG manufacturing for composite hydrogen storage cylinders according to claim 1, characterized in that, In step S5, the baseline confidence coefficient is used to evaluate baseline confidence. in, N is the baseline reliability coefficient for the b-th gas cylinder; out N represents the number of measurement points with abnormal jumps. paired CV represents the number of pairable measurement points before and after curing. global Let be the coefficient of variation of the solidification response at all effective measurement points in the bottle, and α, β, and χ be non-negative weighting coefficients with α+β+χ=1.

9. The method for baseline construction and monitoring calibration of multilayer embedded FBG manufacturing for composite hydrogen storage cylinders according to claim 1, characterized in that, In step S6, the subsequent pressure monitoring stage obtains the load response after manufacturing baseline calibration using the following formula: in, Let T(t) be the peak wavelength of the j-th FBG measuring point at time t, and let T(t) be the temperature at time t, K. ε,j Let K be the strain sensitivity coefficient of the j-th FBG measuring point. T,j Let T be the temperature sensitivity coefficient of the j-th FBG measuring point. base The temperature measured after curing is Δλ. load,j(t) * ε is the load-induced wavelength change after accounting for the effects of manufacturing baseline and temperature. load,j(t) To manufacture the equivalent strain of the load after baseline calibration.

10. A system for baseline construction and monitoring calibration of multilayer embedded FBG manufacturing for composite hydrogen storage cylinders as described in any one of claims 1 to 9, characterized in that, The system includes a composite material hydrogen storage cylinder body 1, an embedded FBG sensor network 2, a spectral demodulation module 3, a measurement point identification management module 4, a pre-curing acquisition module 5, a post-curing acquisition module 6, a peak pairing module 7, an anomaly identification module 8, a manufacturing baseline generation module 9, a baseline reliability evaluation module 10, and a load response calibration module 11. The manufacturing baseline generation module 9 is used to generate the manufacturing baseline dataset, and the load response calibration module 11 is used to... base,j To manufacture a baseline-calibrated load response for zero-reference wavelength output.