4D printing shape memory alloy karst cave hydrogen storage liner tube and predictive sealing method thereof

Abstract

The invention relates to the technical field of shape memory alloy hydrogen storage liner tubes, in particular to a 4D printing shape memory alloy karst cave hydrogen storage liner tube and a predictive sealing method of the 4D printing shape memory alloy karst cave hydrogen storage liner tube. Regional controllable generation of friction heat is achieved, the friction heat is accurately conducted to a coating activation area, the proportion of heat and the target gap variation is kept consistent, the utilization efficiency of the heat-force conversion process is improved, and through loaded curvature radius detection, difference comparison with matched curvature radius combined data and hot spot expansion pressure adjustment records, the friction heat can be accurately generated. A tight closed loop among geometric matching, strain monitoring, deformation screening, segmented thermal adjustment and final state confirmation is formed, the sealing reliability and durability in the high-pressure hydrogen storage environment can be maintained for a long time, and the leakage risk caused by uneven thermal stress and deformation deviation is reduced.

Description

Technical Field

[0001] This invention relates to the field of shape memory alloy hydrogen storage liner technology, and more particularly to a 4D printed shape memory alloy cavity hydrogen storage liner and its predictive sealing method. Background Technology

[0002] In the field of hydrogen storage liner technology, specifically, it refers to tubular or cylindrical liner installed inside hydrogen storage containers or cavern hydrogen storage devices. It is used to store hydrogen under high pressure, low temperature or cyclic pressurization conditions, and can isolate hydrogen from direct contact with the outer container material to prevent material performance degradation caused by hydrogen embrittlement or permeation.

[0003] A shape memory alloy liner for karst cave hydrogen storage, manufactured using 4D printing, and its predictive sealing method aim to enable the liner to have controllable shape change capabilities, meeting the stress adaptation requirements of karst cave hydrogen storage environments. The shape memory alloy has the characteristic of undergoing a phase transition and returning to its original shape under specific external conditions, allowing the liner to automatically restore to a preset shape when sealing is required. The purpose is to improve the sealing reliability and durability of the hydrogen storage liner, reduce sealing failures caused by temperature and pressure changes, ensure the safety and stability of the hydrogen storage system during long-term cyclic use, and reduce the risk of hydrogen leakage.

[0004] In practical applications, existing hydrogen storage liners rely heavily on fixed geometry and single-piece molding precision to fit the walls of cavities or containers. Under long-term high pressure, low temperature, and cyclic pressurization conditions, gaps easily accumulate between the liner surface and the inner wall, making it difficult to proactively adjust the shape according to temperature and stress changes during operation. Current processes rely on passively adapting to the elastic recovery of the material, the degree of which is greatly affected by temperature, pressure fluctuations, and hydrogen permeation. Once changes in the external environment exceed the material's own stable range, it can easily lead to insufficient local sealing pressure. Current technologies mostly rely on periodic inspections or single-parameter alarms for operational monitoring, failing to correlate strain and temperature changes. This results in potential shape deviations not being corrected in time. When the cyclical changes in hydrogen temperature are superimposed with fluctuations in stress distribution, the minute deformations of the liner contact surface will gradually expand, eventually causing loosening or uneven contact of the sealing surface, increasing the risk of hydrogen leakage and maintenance frequency, and potentially forming stress concentration zones in local areas, accelerating material fatigue aging. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of existing technologies and propose a 4D-printed shape memory alloy cavitary hydrogen storage liner and its predictive sealing method.

[0006] To achieve the above objectives, the present invention employs the following technical solution: a 4D-printed shape memory alloy cavity hydrogen storage liner: The matrix layer is a NiTiNbTa quaternary alloy formed by laser powder bed melting, with the composition Ni46.5Ti43Nb9.5Ta1, a phase transformation temperature As of 28℃ and Af of 35℃, and a fatigue life of more than 10^4 cycles. The sensing layer consists of a fiber grating array disposed within the liner wall, with an array spacing of 5 mm. The functional layer, located on the outer surface of the substrate layer, is a PHB / ZnO nanocomposite coating with a thickness of 80±5μm formed by plasma spraying. The coating contains microencapsulated silane coupling agent KH-570. The liner tube wall has a negative Poisson's ratio structure with an inward concavity angle of 60°. Under axial compression, the radial expansion rate is not less than 25%. The surface has a biomimetic shark skin microgroove structure with a ridge height of 20μm, which can reduce flow resistance by 42%. When the liner is decommissioned, the PHB / ZnO nanocomposite coating can be recovered by acid dissolution and used as 3D printing raw material.

[0007] A 4D-printed shape memory alloy cavity hydrogen storage liner and its predictive sealing method include the following steps: S1: Based on the radius of curvature of the inner wall of the karst cave, the thickness of the cylinder wall, the outer diameter of the cylinder wall and the negative Poisson's ratio structural parameters, measure the difference between the inner wall and the outer diameter, compare the difference with the radius corresponding to the elastic limit stress of the alloy, screen out the set of matching points, and establish the combination data of the curvature radius of the fitting. S2: Based on the combined curvature radius data, match the curvature radius with the circumferential strain value and temperature cycle range monitored by the fiber optic grating, and use Kalman filtering to map the data to the curvature record in chronological order, detect the phase transition start and end temperatures, eliminate out-of-limit combinations, and generate a set of phase transition-adaptable combinations. S3: Based on the phase-change adaptable combination set, mark the curvature position number, compare the deformation of the corresponding number with the pressure of the sealing contact surface node and the thermal conductivity area of ​​the PHB coating, generate a number table for positions with positive differences, and obtain the number list of positions to be adjusted. S4: Based on the numbered list of positions to be adjusted, read the radius of curvature and deformation of the numbered positions, apply segmented increasing radial pressure according to the number using wavelet decomposition to generate frictional heat, and conduct it to the PHB coating microbial community activation area. The process terminates when the change is proportional to the difference in the target gap, and a segmented loading pressure distribution table is established. S5: Based on the segmented loading pressure distribution table, load the pressure at the set position, measure the radius of curvature after completion, compare the difference between the data and the combined data of the radius of curvature, and associate it with the hot spot expansion pressure adjustment record to obtain the set of sealing completion morphological parameters.

[0008] As a further embodiment of the present invention, the matching curvature radius combination data includes curvature radius values, position numbers, and continuity identifiers; the phase-change adaptable combination set includes curvature radius group numbers, strain value ranges, and temperature ranges; the adjustment position number sheet includes position numbers, pressure values, and difference records; the segmented loading pressure distribution table includes position numbers, loading pressures, and termination conditions; and the sealing completion morphology parameter set includes curvature radius distribution, difference comparison results, and pressure adjustment information.

[0009] As a further aspect of the present invention, the specific steps for generating the combined curvature radius data are as follows: Based on the curved surface radius, cylinder wall thickness, cylinder wall outer diameter, and negative Poisson's ratio structural parameters of the cave's inner wall, the difference between the curved surface and the cylinder wall outer diameter at each corresponding measuring point is measured. The measurements are read and converted into numerical values ​​using contact measuring instruments, recorded in a coordinate scale table, and the measuring point number is marked to generate a set of measurement difference coordinates. Based on the measured difference coordinate set, the difference of each measuring point is compared with the radius corresponding to the elastic limit stress of the alloy. Measuring points that exceed the allowable radius are eliminated. During the comparison, the measuring point number is used as the primary key, and the curvature radius value of the measuring point is bound to the same recording unit to establish a combination of curvature radius data.

[0010] As a further aspect of the present invention, the specific steps for generating the phase-change adaptable combination set are as follows: Based on the combined curvature radius data, the curvature radius is matched with the circumferential strain value and temperature cycle range monitored by the fiber optic grating. During matching, Kalman filtering is used to predict and correct the strain value and temperature data and remove abnormal fluctuations. By setting a time index for each set of curvature radii and binding the corrected strain value and temperature range, and storing the mapping in the recording unit, a curvature mapping data table is generated. Based on the curvature mapping data table, the temperature value of each node in the time series is detected and the recorded curvature radius number is recorded. The critical temperature points for the start and end of the phase transition are detected by comparing the temperature value change range and locking them. The critical values ​​are then bound to the curvature radius numbers to generate a set of critical temperature points for phase transition. Based on the set of critical phase transition temperatures, combinations with critical temperatures exceeding the set range are eliminated. Data units that meet the conditions but are filtered by curvature radius number and corresponding strain range are also eliminated. The data units are then sorted from high to low according to the curvature change amplitude to generate a set of phase transition adaptable combinations.

[0011] As a further aspect of the present invention, the Kalman filter, when matching the radius of curvature with the circumferential strain value and temperature cycle range monitored by the fiber optic grating, arranges the collected strain and temperature data into an observation sequence in chronological order. The initial state is set as the state estimate value by the measured value at the first time point. At each subsequent time point, a predicted value is calculated based on the previous state estimate value and the state transition relationship. The predicted value is subtracted from the current measured value to obtain the residual. The residual is weighted and calculated with a preset noise variance to obtain the update gain. The update gain is multiplied by the residual and added to the predicted value to form a new state estimate value. At the same time, the state estimate value replaces the original observation data as the corrected strain and temperature data.

[0012] As a further aspect of the present invention, the specific steps for generating the adjustment position number sheet are as follows: Based on the phase-change adaptable combination set, the position number of the radius of curvature within the combination is marked. The marking is achieved by reading the spatial coordinates of the position and mapping them to the node index table, and the relationship between the number and the coordinates is stored synchronously in the mapping data table to generate a node number coordinate set. Based on the node number coordinate set, the node coordinates are matched with the node pressure value of the sealing contact surface and the coordinates of the heat conduction area of ​​the PHB coating. During matching, the node deformation value is calculated simultaneously, and the numbers with positive differences between the deformation and the target sealing gap are arranged in sequence to obtain the number list of positions to be adjusted.

[0013] As a further aspect of the present invention, the specific steps for generating the segmented loading pressure allocation table are as follows: Based on the numbered list of positions to be adjusted, the radius of curvature and deformation of the numbered positions are read, the data stored by matching the numbers is read and the corresponding deformation values ​​are calculated, and the two are bound to the position coordinates to form a set of data records to generate a loading preparation dataset; Based on the loading preparation dataset, segmented increasing radial pressure is applied in numerical order. The pressure value is gradually increased through the loading device while the frictional heat is recorded. During the recording process, wavelet decomposition is used to decompose the pressure curve and frictional heat signal into low-frequency trend and high-frequency fluctuation components. The decomposed data is stored in correspondence with the location number and associated with the PHB coating microbial community activation area to generate a loading pressure and heat table. Based on the loading pressure heat meter, the difference between the change in curvature at the monitoring position and the target gap is monitored. The curvature radius is measured after each loading and the ratio of the change is calculated. Loading is stopped when the ratio is equal to the target value, and the number and corresponding pressure data are stored to establish a segmented loading pressure distribution table.

[0014] As a further aspect of the present invention, the wavelet decomposition, when recording the pressure curve and friction heat signal, constructs a data vector by arranging the continuously acquired signals in the order of sampling time, selects a preset signal length as the decomposition window, performs a decomposition operation on the signal within the window to separate multi-level scale coefficients, recombines the decomposed low-frequency coefficients to form a low-frequency trend signal, recombines the high-frequency coefficients to form a high-frequency fluctuation signal, and establishes a corresponding index relationship between the low-frequency trend signal and the high-frequency fluctuation signal and the original position number, respectively, and stores them as a set of decomposed pressure and friction heat values.

[0015] As a further aspect of the present invention, the specific steps for generating the sealing completion morphology parameter set are as follows: Based on the segmented loading pressure allocation table, a set pressure is applied to the corresponding node position according to the number. The loading start and end time and position number are recorded during the application process, and a corresponding record form for time and number is generated to generate a loading position record table. Based on the loading position record table, the radius of curvature of the loaded node position is measured. The measurement results are compared one by one with the combined data of the radius of curvature. The comparison data is then bound to the hot spot expansion pressure adjustment record field to obtain the set of sealing completion morphological parameters.

[0016] Compared with the prior art, the advantages and positive effects of the present invention are as follows: In this invention, by matching the radius of curvature with the circumferential strain value and temperature cycle range monitored by the fiber optic grating, and using Kalman filtering to perform time series mapping on the data, the starting and ending temperature points of the phase transition can be stably extracted. This ensures the continuity and accuracy of the morphological recovery process under environmental loading changes. The difference between the pressure at the sealing contact surface node and the deformation of the thermally conductive area of ​​the PHB coating is compared and processed so that the judgment result only retains the area that helps to form a stable seal, avoids invalid positions from participating in the adjustment, and reduces the risk of mechanical interference. In this invention, by reading the radius of curvature and deformation at the numbered position, a wavelet decomposition method is used to apply segmented incremental radial pressure, thereby achieving regional controllable generation of frictional heat and precise conduction to the coating activation zone. This ensures that the heat is proportional to the change in the target gap, thus improving the utilization efficiency of the heat-to-mechanical conversion process. In this invention, the curvature radius is detected after loading, and the difference is compared with the combined curvature radius data. Combined with the hot spot expansion pressure adjustment record, the geometric and mechanical parameters of the final sealing state are ensured to be stable. This forms a tight closed loop between geometric matching, strain monitoring, deformation screening, segmented thermal adjustment and final state confirmation. This can maintain the sealing reliability and durability under high pressure hydrogen storage environment for a long time and reduce the leakage risk caused by uneven thermal stress and deformation deviation. Attached Figure Description

[0017] Figure 1This is a schematic diagram of the main steps of the present invention. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0019] In the description of this invention, it should be understood that the terms "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientation or positional relationships, are based on the orientation or positional relationships shown in the accompanying drawings and are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, in the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0020] Example 1 Please see Figure 1 The present invention provides a technical solution: A 4D-printed shape memory alloy cavity hydrogen storage liner: The matrix layer is a NiTiNbTa quaternary alloy formed by laser powder bed melting, with the composition Ni46.5Ti43Nb9.5Ta1, a phase transformation temperature As of 28℃ and Af of 35℃, and a fatigue life of more than 10^4 cycles. The sensing layer consists of a fiber grating array disposed within the liner wall, with an array spacing of 5 mm. The functional layer, located on the outer surface of the substrate layer, is a PHB / ZnO nanocomposite coating with a thickness of 80±5μm formed by plasma spraying. The coating contains microencapsulated silane coupling agent KH-570. The liner tube wall has a negative Poisson's ratio structure with an inward concavity angle of 60°. Under axial compression, the radial expansion rate is not less than 25%. The surface has a biomimetic shark skin microgroove structure with a ridge height of 20μm, which can reduce flow resistance by 42%. When the liner is decommissioned, the PHB / ZnO nanocomposite coating can be recovered by acid dissolution and used as 3D printing raw material.

[0021] A 4D-printed shape memory alloy cavity hydrogen storage liner and its predictive sealing method include the following steps: S1: Based on the radius of curvature of the inner wall of the karst cave, the thickness of the cylinder wall, the outer diameter of the cylinder wall and the negative Poisson's ratio structural parameters, measure the difference between the inner wall and the outer diameter, compare the difference with the radius corresponding to the elastic limit stress of the alloy, screen out the set of matching points, and establish the combination data of the curvature radius of the fitting. S2: Based on the combined curvature radius data, the curvature radius is matched with the circumferential strain value and temperature cycle range monitored by the fiber optic grating. Kalman filtering is used to map the data to the curvature record in chronological order, detect the phase transition start and end temperatures, eliminate out-of-limit combinations, and generate a set of phase transition-adaptable combinations. S3: Based on the phase change adaptable combination set, mark the curvature position number, compare the deformation of the corresponding number with the sealing contact surface node pressure and the thermal conductive area of ​​the PHB coating, generate a number table for positions with positive differences, and obtain the number list of positions to be adjusted. S4: Based on the numbered list of positions to be adjusted, read the radius of curvature and deformation of the numbered positions, apply segmented increasing radial pressure according to the number using wavelet decomposition to generate frictional heat, and conduct it to the PHB coating microbial community activation area. The process terminates when the change is proportional to the difference in the target gap, and a segmented loading pressure distribution table is established. S5: Based on the segmented loading pressure distribution table, load the pressure at the set position, measure the radius of curvature after completion, compare the difference between the data and the combined data of the radius of curvature, and associate it with the hot spot expansion pressure adjustment record to obtain the set of sealing completion morphological parameters.

[0022] The curvature radius combination data includes curvature radius values, location numbers, and continuity indicators; the phase change adaptable combination set includes curvature radius group numbers, strain value ranges, and temperature ranges; the adjustment location number sheet includes location numbers, pressure values, and difference records; the segmented loading pressure distribution table includes location numbers, loading pressures, and termination conditions; and the sealing completion morphology parameter set includes curvature radius distribution, difference comparison results, and pressure adjustment information.

[0023] The specific steps for generating the combination data of curvature radii are as follows: Based on the curved surface radius, cylinder wall thickness, cylinder wall outer diameter, and negative Poisson's ratio structural parameters of the cave's inner wall, the difference between the curved surface and the cylinder wall outer diameter at each corresponding measuring point is measured. The measurements are read and converted into numerical values ​​using contact measuring instruments, recorded in a coordinate scale table, and the measuring point number is marked to generate a set of measurement difference coordinates. Based on the coordinate set of measurement difference, the difference of each measuring point is compared with the radius corresponding to the elastic limit stress of the alloy. Measuring points that exceed the allowable radius are eliminated. During the comparison, the measuring point number is used as the primary key, and the curvature radius value of the measuring point is bound to the same recording unit to establish the matching curvature radius combination data. Based on the surface radius, wall thickness, outer diameter, and negative Poisson's ratio structural parameters of the cave's inner wall, a three-dimensional surface least squares fitting algorithm is employed. The input is a matrix data file containing the three-dimensional coordinates of each measuring point. The iteration convergence tolerance is set to 0.0001, the maximum number of iterations is set to 500, and the fitting order is a second-order polynomial surface. The difference between the surface and the outer diameter of the cylinder at each corresponding measuring point is calculated. This calculation is accomplished by registering the discrete point cloud data obtained from the inner wall scan with the theoretical outer diameter model of the cylinder in the same coordinate system. The registration step uses an iterative nearest-neighbor registration algorithm. The input is the actual point cloud of the inner wall and the theoretical point cloud of the cylinder. The number of nearest-neighbor search points is set to... 1. The matching distance threshold is 0.5 mm, the maximum number of convergence iterations is 200. After registration is completed, the difference between the surface and the outer diameter is calculated from the normal direction according to the numbered measurement points, and the difference is recorded synchronously with the spatial coordinates of the measurement points. The measurement data of the contact gauge is read through the serial data interface. The interface baud rate is set to 115200, the data frame length is fixed at 64 bytes, and each frame contains the measurement point number, three-dimensional coordinate value and measurement difference field. All data are uniformly converted to millimeter units and four decimal places are retained. The data is recorded in the coordinate scale table according to the measurement point number order. The numbering method adopts a six-digit integer encoding in spiral order, and finally the measurement difference coordinate set is generated. Based on the measurement difference coordinate set, a radius threshold comparison algorithm is adopted. The input is the radius value of each measurement point and the allowable radius range corresponding to the elastic limit stress of the alloy. The radius of each measurement point is compared with the corresponding allowable radius one by one. The calculation precision used for comparison is double precision floating point. When an equality condition is encountered, it is classified as a qualified item. The measurement point number that meets the condition is extracted and a qualified number list is generated. In the comparison process, the measurement point number is used as the primary key. Each number and the corresponding curvature radius value are bound and stored in the same record. A hash mapping table structure is used to store the data. The number of buckets allocated in the mapping table is set to 512. The conflict resolution method is linked list. After all the measurement point data is bound and stored, the matching curvature radius combination data is generated.

[0024] The specific steps for generating a set of phase-change adaptable combinatorial sets are as follows: Based on the combined curvature radius data, the curvature radius is matched with the circumferential strain value and temperature cycle range monitored by the fiber optic grating. During the matching process, Kalman filtering is used to predict and correct the strain value and temperature data and remove abnormal fluctuations. By setting a time index for each set of curvature radii and binding the corrected strain value and temperature range, the mapping is stored in the recording unit to generate a curvature mapping data table. Based on the curvature mapping data table, the temperature value of each node in the time series is detected and the curvature radius number is recorded. The critical temperature points for the start and end of the phase transition are detected by comparing the range of temperature value changes and binding the critical values ​​with the curvature radius numbers to generate a set of critical temperature points for phase transition. Based on the set of critical phase transition temperatures, combinations with critical temperatures exceeding the set range are eliminated. Data units that meet the conditions are eliminated by filtering the radius of curvature number and corresponding strain range. The data units are then sorted from high to low according to the magnitude of curvature change to generate a set of phase transition adaptable combinations. Based on the combined curvature radius data, a Kalman filter algorithm is used to match the curvature radius with the circumferential strain value and temperature cycle range monitored by the fiber optic grating. The state transition matrix values ​​of the Kalman filter are configured as an identity matrix, and the observation matrix is ​​a row-three-column structure corresponding to the strain value, temperature value and time index. The time step of state prediction is set to 1 second, the prediction noise covariance is set to 0.002, the observation noise covariance is set to 0.005, the initial state vector is filled with zero vectors, and the initial covariance matrix is ​​a diagonal matrix with all diagonal values ​​of 0.1. During the prediction stage, the strain value and temperature value are updated in time order, and the residual between the prediction result and the actual monitoring value is calculated at each step. Abnormal fluctuation data with residuals greater than 0.02 are removed. A time index is established for each set of curvature radii and the corrected strain value and temperature cycle range are bound together. The mapped data is stored in a two-dimensional array structure with the curvature radius number as the index to generate a curvature mapping data table. Based on the curvature mapping data table, a time series threshold comparison algorithm is used to detect the temperature value of each node in the time series and the recorded curvature radius number. The temperature value change threshold is set to 0.1 degrees Celsius. When the change of adjacent nodes is greater than the threshold, they are judged as candidate boundaries. The starting point of the increasing interval is judged as the critical temperature point for the phase change initiation, and the ending point of the decreasing interval is judged as the critical temperature point for the phase change termination. The detection process traverses the entire time series in the order of curvature radius number, and binds each critical temperature point with the corresponding curvature radius number in the same storage unit. The binding structure is in the form of key-value pairs, where the key value is the curvature radius number and the key content is the floating-point temperature value. After the binding is completed, the data is summarized and output as a temporary dataset, forming a set of critical temperatures for phase change. Based on the critical temperature set for phase transition, a range filtering and sorting algorithm is adopted. The filtering step removes temperature value combinations that exceed the set range. The set temperature range is the starting critical temperature between 20 and 50 degrees Celsius and the ending critical temperature between 5 and 15 degrees Celsius. During the removal, the corresponding radius of curvature number and strain range are removed simultaneously. The remaining data units that meet the conditions are retained in a newly created array structure. The data in the array are sorted in descending order according to the curvature change amplitude. The curvature change amplitude is calculated as the absolute value of the difference between the radius of curvature before and after the phase transition, and the numerical unit is millimeters. The sorted data are output to the new structural unit in sequence to generate a set of phase transition adaptable combinations.

[0025] Kalman filtering, when matching the radius of curvature with the circumferential strain value and temperature cycle range monitored by fiber optic grating, arranges the collected strain and temperature data into an observation sequence in chronological order. The initial state is set as the state estimate by the measured value at the first time point. At each subsequent time point, the predicted value is calculated based on the previous state estimate and the state transition relationship. The predicted value is subtracted from the current measured value to obtain the residual. The residual is weighted with the preset noise variance to obtain the update gain. The update gain is multiplied by the residual and added to the predicted value to form a new state estimate. At the same time, the state estimate replaces the original observation data as the corrected strain and temperature data.

[0026] The specific steps for generating the list of locations to be adjusted are as follows: Based on the phase-change adaptable combination set, the position number of the curvature radius within the combination is labeled. The labeling is achieved by reading the spatial coordinates of the position and mapping them to the node index table, and the relationship between the number and the coordinates is synchronously stored in the mapping data table to generate a node number coordinate set. Based on the node number coordinate set, the node coordinates are matched with the node pressure value of the sealing contact surface and the coordinates of the heat conduction area of ​​the PHB coating. During the matching, the node deformation value is calculated simultaneously, and the numbers with positive differences between the deformation and the target sealing gap are arranged in sequence to obtain the number list of positions to be adjusted. Based on the phase-change adaptable combinatorial set, a node position labeling and spatial mapping algorithm is used to label the position numbers of the curvature radii within the combinatorial set. During execution, the three-dimensional spatial coordinates corresponding to each curvature radius within the combinatorial set are first read. The spatial coordinates are stored in the structure array in millimeters, and the array index is consistent with the curvature radius number. The read coordinates are mapped to the node index table through nearest neighbor interpolation. During interpolation, the neighborhood search radius is set to 2 millimeters, and the weight decay coefficient is 0.5. After the mapping is completed, a correspondence between node numbers and spatial coordinates is formed, and the mapping data table is stored synchronously in key-value pairs. The key is a six-digit integer of the node number, and the value is a ternary floating-point number of the coordinate. Finally, a set of node number coordinates is generated. Based on the node number coordinate set, a node coordinate and physical parameter matching algorithm is adopted. The matching step compares the node coordinates with the array of node pressure values ​​on the sealing contact surface and the array of coordinates of the thermally conductive area of ​​the PHB coating one by one. During the comparison, the node number is used as the primary key for association. The tolerance of coordinate matching is set to 0.2 mm. The pressure value is in MPa and the coordinate of the thermally conductive area is in millimeters. After the matching is completed, the node deformation is calculated synchronously. The deformation calculation adopts the vector difference method. The square root of the sum of the squares of the differences in the three directions of the node before and after loading is taken as the deformation value. After obtaining the deformation, it is subtracted from the target difference of the sealing gap. The node number with a positive difference is retained. After arranging the nodes in ascending order, the number list of positions to be adjusted is obtained. Only records that meet the conditions are retained in the data unit. These records are sorted in descending order of curvature change range. The curvature change range is calculated based on the absolute value of the difference in curvature radius before and after the phase change, in millimeters. After sorting, the set of phase-change adaptable combinations is output.

[0027] The specific steps for generating the segmented load pressure distribution table are as follows: Based on the numbered list of positions to be adjusted, the radius of curvature and deformation of the numbered positions are read, the data stored by matching the numbers is read and the corresponding deformation values ​​are calculated, and the two are bound to the position coordinates to form a set of data records to generate a loading preparation dataset; Based on the loading preparation dataset, segmented increasing radial pressure is applied in numerical order. The pressure value is gradually increased through the loading device while the frictional heat is recorded. During the recording process, wavelet decomposition is used to decompose the pressure curve and frictional heat signal into low-frequency trend and high-frequency fluctuation components. The decomposed data is stored in correspondence with the location number and associated with the PHB coating microbial community activation area to generate a loading pressure and heat table. Based on the loading pressure heat meter, the difference between the change in curvature at the monitoring position and the target gap is monitored. The monitoring is carried out by measuring the radius of curvature after each loading and calculating the ratio of the change. Loading is stopped when the ratio is equal to the target value, and the number and corresponding pressure data are stored to establish a segmented loading pressure distribution table. Based on the numbered list of positions to be adjusted, a numbered data reading and calculation algorithm is used to read the radius of curvature and deformation of the numbered position. During execution, the number is matched with the radius of curvature records stored in the database and the three-dimensional coordinate data of the nodes before and after loading. During the matching process, the primary key index is used to locate the record row. When calculating the deformation, the square difference of the coordinates in the three directions is summed and the square root is taken. The calculation precision is set to four decimal places. The radius of curvature value, deformation value and position coordinates are stored together in a structured array. Each cell of the array contains a triplet of number, radius of curvature, deformation and spatial coordinates. Finally, the loading preparation dataset is generated. Based on the loading preparation dataset, a step-by-step loading and signal decomposition algorithm is adopted to apply segmented increasing radial pressure in numerical order. During execution, the loading device controller increases the applied pressure value from an initial pressure of 0.5 MPa in increments of 0.1 MPa to a maximum pressure of 2 MPa within each segment cycle. The pressure sensor collects the current pressure value and the heat signal generated by friction every 0.1 seconds in real time. The heat unit is degrees Celsius, and the sampling data length is not less than 1000 points. In the data decomposition stage, the wavelet decomposition algorithm is used to decompose the pressure curve and heat signal. The db4 wavelet basis is selected, and the number of decomposition layers is set to 4. The low-frequency component in the decomposition result is used as the trend signal, and the high-frequency component is used as the fluctuation signal. The low-frequency and high-frequency data are recorded one by one with the location number in a two-dimensional data table structure. A PHB coating bacterial community activation zone identification field is added to the record, and finally, a loading pressure heat table is generated. Based on the loading pressure heat meter, a real-time ratio monitoring and control algorithm is adopted to monitor the change in curvature at the position and the difference between the target gap. During execution, after each loading cycle, the latest curvature radius is measured by a CNC displacement probe. The change in curvature is calculated as the difference between the initial radius before loading and the current radius. The change is then compared with the target gap value, and the ratio calculation result is retained to three decimal places. When the ratio and the difference between the target gap reach the equal ratio condition, the loading operation is stopped immediately, and the corresponding position number and the pressure value at this time are recorded in the data table. Finally, a segmented loading pressure distribution table is established.

[0028] Wavelet decomposition, when recording pressure curves and frictional heat signals, constructs a data vector from continuously acquired signals in the order of sampling time. A preset signal length is selected as the decomposition window, and the decomposition operation is performed on the signals within the window to separate multi-level scaling coefficients. The low-frequency coefficients obtained from the decomposition are recombined to form a low-frequency trend signal, and the high-frequency coefficients are recombined to form a high-frequency fluctuation signal. The low-frequency trend signal and the high-frequency fluctuation signal are respectively linked to the original position number and stored as a set of decomposed pressure and frictional heat values.

[0029] The specific steps for generating the complete sealing morphology parameter set are as follows: Based on the segmented loading pressure allocation table, the set pressure is applied to the corresponding node position according to the number. The loading start and end time and position number are recorded during the application process, and a corresponding record form for time and number is generated to generate a loading position record table. Based on the loading position record table, the radius of curvature of the loaded node position is measured. The measurement results are compared with the combined data of the radius of curvature of the mating nodes one by one, and the comparison data is bound to the hot spot expansion pressure adjustment record field to obtain the set of sealing completion morphological parameters. Based on a segmented loading pressure allocation table, a number-driven loading control algorithm is adopted. The set pressure is applied to the corresponding node position according to the number. During execution, the output pressure is adjusted to the corresponding value range in the allocation table by the pressure controller. The adjustment step of the pressure controller is set to 0.05 MPa, the stabilization time is 3 seconds, and the accuracy threshold is ±0.01 MPa. The loading process records the start time and end time of loading and binds them to the node position number. The time record is obtained using a high-precision real-time clock module with a time resolution of milliseconds. The number and time binding data are stored in a two-dimensional array structure. The array index corresponds to the node number, and the value is a timestamp pair of start and end times. After loading is completed, the number and start and end time data are output in node order and saved as text format, finally generating a loading position record table. Based on the loading position record table, a curvature measurement and data comparison algorithm is used to measure the curvature radius of the loaded node position. During execution, at least 500 surface sampling points are collected within the three-dimensional coordinate range of the node using a laser scanning rangefinder, with a sampling interval of 0.5 mm. The node curvature radius is calculated using the least squares curvature fitting method, with a calculation accuracy of four decimal places. The obtained measured radius is compared one by one with the combined data of the matching curvature radius. The comparison rule is that the absolute value of the numerical difference does not exceed 0.02 mm to be considered a match. The matching result is bound to the corresponding node number and measured radius, and a hot spot expansion pressure adjustment record field is added. The field unit is megapascal, and the recording accuracy is ±0.01 megapascal. All numbers, measured radii, and adjustment pressure records are stored in a structured data table, finally obtaining the sealing completion morphological parameter set.

[0030] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A 4D printed shape memory alloy cavern hydrogen storage liner, characterized in that: a base layer is a NiTiNbTa quaternary alloy formed by laser powder bed fusion, with a composition of Ni46.5Ti43Nb9.5Ta1, a phase transition temperature As of 28℃ and an Af of 35℃, and a fatigue life of more than 10^4 cycles; a sensing layer is an optical fiber grating array arranged in the liner wall, with an array spacing of 5mm; and a functional layer is a PHB / ZnO nanocomposite coating formed by plasma spraying on the outer surface of the base layer, with a thickness of 80±5μm, and containing microencapsulated silane coupling agent KH-570; wherein the liner wall is a negative Poisson's ratio structure with an inner concave angle of 60°, a radial expansion rate of not less than 25% under axial compression, and a surface having a bionic sharkskin microgroove structure with a ridge height of 20μm, which can reduce the flow resistance by 42%, and the liner can recycle the PHB / ZnO nanocomposite coating by acid dissolution at retirement and use it as a 3D printing raw material. The method comprises the following steps: S1: Based on the inner wall curvature radius, the thickness of the cylinder wall, the outer diameter of the cylinder wall and the negative Poisson's ratio structure parameters, the difference between the inner wall and the outer diameter is measured, the difference is compared with the corresponding radius of the elastic limit stress of the alloy, the point set that meets the requirements is screened out, and the matching curvature radius combination data is established; S2: Based on the matching curvature radius combination data, the curvature radius is matched with the hoop strain value and temperature cycle interval monitored by the optical fiber grating, and the data is mapped to the curvature record in time sequence by using Kalman filtering, the phase transition start and end temperatures are detected, the out-of-limit combination is removed, and the phase transition adaptive combination set is generated; 2. The 4D printed shape memory alloy cavern hydrogen storage liner predictive sealing method of claim 1, wherein, S3: Based on the phase transition adaptive combination set, the curvature position number is labeled, the number is compared with the node pressure of the sealing contact surface and the deformation of the PHB coating heat conduction area, the number table is generated for the position with a positive difference, and the position number single to be adjusted is obtained; S4: Based on the position number single to be adjusted, the curvature radius and deformation of the numbered position are read, the wavelet decomposition is applied to the numbered position to generate friction heat by applying segmented incremental radial pressure, the change amount is transmitted to the PHB coating microbial activation area when the target gap difference is equal, and the segmented loading pressure distribution table is established; S5: Based on the segmented loading pressure distribution table, the position is loaded according to the set pressure, the curvature radius is measured after completion, the data is compared with the difference of the matching curvature radius combination data and is associated with the hot spot expansion pressure adjustment record to obtain the sealing completed morphology parameter set. The matching curvature radius combination data includes curvature radius value, position number and continuity identifier, the phase transition adaptive combination set includes curvature radius group number, strain value interval and temperature range, the position number single to be adjusted includes position number, pressure value and difference record, the segmented loading pressure distribution table includes position number, loading pressure and termination condition, and the sealing completed morphology parameter set includes curvature radius distribution, difference comparison result and pressure adjustment information. The specific steps for generating the matching curvature radius combination data are: ​ 3. The 4D printed shape memory alloy cavern hydrogen storage liner predictive sealing method of claim 1, wherein, ​ 4. The 4D printed shape memory alloy cavern hydrogen storage liner predictive sealing method of claim 1, wherein, ​ Based on the cave inner wall curved surface radius, the cylinder wall thickness, the cylinder wall outer diameter and the negative Poisson's ratio structure parameters, the curved surface and the cylinder wall outer diameter difference value at each corresponding measuring point is measured, the measuring is read by a contact gauge and converted into a numerical value, recorded in the coordinate scale table, and the measuring point number is marked, and the measuring difference coordinate set is generated; Based on the measuring difference coordinate set, the difference value of each measuring point is compared with the corresponding radius of the alloy elastic limit stress, and the measuring points exceeding the allowed radius are removed, the measuring point number is used as the primary key during comparison, and the measuring point curvature radius value is bound to the same record unit, and the matching curvature radius combination data is established.

5. The 4D printed shape memory alloy cavern hydrogen storage liner predictive sealing method of claim 1, wherein, The specific steps for generating the phase-changeable adaptive combination set are: Based on the matching curvature radius combination data, the curvature radius is matched with the ring direction strain value and the temperature cycle interval monitored by the fiber grating, when matching, the Kalman filter is used to predict and correct the strain value and the temperature data and remove abnormal fluctuations, the time index is set for each group of curvature radius, the corrected strain value and the temperature interval are bound, the mapping is stored in the record unit, and the curvature mapping data table is generated; Based on the curvature mapping data table, the temperature value of each node in the time sequence and the recorded curvature radius number are detected, the critical temperature points of phase change start and end are locked by comparing the temperature value change range, and the critical value and the curvature radius number are bound, and the phase change critical temperature set is generated; Based on the phase change critical temperature set, the combinations with critical temperature exceeding the set range are removed, the data units meeting the conditions are retained by screening the curvature radius number and the corresponding strain range, and the curvature change amplitude is sorted from high to low, and the phase-changeable adaptive combination set is generated.

6. The 4D printed shape memory alloy cavern hydrogen storage liner predictive sealing method of claim 5, wherein, When matching the curvature radius with the ring direction strain value and the temperature cycle interval monitored by the fiber grating, the collected strain value and temperature data are arranged in time sequence to form an observation sequence, the initial state is set as the state estimation value by the measurement value at the first time point, the prediction value is calculated according to the previous state estimation value and the state transition relationship at each subsequent time point, the residual error is obtained by subtracting the prediction value from the current measurement value, the update gain is obtained by weight calculation of the residual error and the preset noise variance, the update gain is multiplied by the residual error and added to the prediction value to form a new state estimation value, and the state estimation value is replaced by the original observation data as the corrected strain value and temperature data.

7. The 4D printed shape memory alloy cavern hydrogen storage liner predictive sealing method of claim 1, wherein, The specific steps for generating the position number list to be adjusted are: Based on the phase-changeable adaptive combination set, the position number of the curvature radius in the combination is marked, the position number and the coordinate relationship are read and mapped to the node index table, and the mapping data table is stored synchronously, and the node number coordinate set is generated; Based on the node number coordinate set, the node coordinates are matched with the sealing contact surface node pressure value and the PHB coating heat conduction region coordinates, when matching, the node deformation variable value is calculated synchronously, and the numbers with positive deformation variable and target difference value of the sealing gap are arranged in sequence, and the position number list to be adjusted is obtained.

8. The 4D printed shape memory alloy cavern hydrogen storage liner predictive sealing method of claim 1, wherein, The specific steps for generating the segmented loading pressure distribution table are: Based on the position number to be adjusted, the curvature radius and deformation of the numbered position are read, the corresponding deformation value is calculated by matching the stored data, and the two are bound with the position coordinates to form a set of data records to generate a loading preparation data set; Based on the loading preparation data set, incremental radial pressure is applied in sequence according to the number, and the pressure value is increased step by step through the loading device while recording the friction heat. During the recording process, the pressure curve and friction heat signal are decomposed into low-frequency trend and high-frequency fluctuation components by wavelet decomposition, and the decomposed data are stored one by one corresponding to the position number and associated with the PHB coating bacterial activation zone to generate a loading pressure heat table; Based on the loading pressure heat table, the position curvature change and the target gap difference are monitored, the curvature radius is measured after each loading and the change ratio is calculated, and when the ratio is equal to the target value, the loading is stopped, the number and corresponding pressure data are stored, and a segmented loading pressure distribution table is established.

9. The 4D printed shape memory alloy cavern hydrogen storage liner predictive sealing method of claim 8, wherein, The wavelet decomposition, when recording the pressure curve and friction heat signal, constructs a data vector according to the sampling time sequence of the continuously collected signal, selects a preset signal length as a decomposition window, performs a decomposition operation on the signal in the window to separate multiple scale coefficients, recombines the low-frequency coefficients obtained by decomposition to form a low-frequency trend signal, recombines the high-frequency coefficients to form a high-frequency fluctuation signal, and establishes a corresponding index relationship between the low-frequency trend signal and the high-frequency fluctuation signal and the original position number, and stores it as a decomposed pressure and friction heat value set.

10. The 4D printed shape memory alloy cavern hydrogen storage liner predictive sealing method of claim 1, wherein, The specific steps for generating the sealing completion morphology parameter set are: Based on the segmented loading pressure distribution table, a set pressure is applied to the corresponding node position according to the number, the loading start and end time and position number are recorded during the application process, and a corresponding record table of time and number is generated to generate a loading position record table; Based on the loading position record table, the curvature radius of the loaded node position is measured, the measurement result is compared with the combined data of the matching curvature radius one by one, and the comparison data are bound with the hot spot expansion pressure adjustment record field to obtain the sealing completion morphology parameter set.

Images