Hydrogen storage unit service performance prediction method and system

Abstract

The invention provides a hydrogen storage unit service performance prediction method and system, and belongs to the technical field of hydrogen storage, and the method comprises the steps: determining an initial grid model through a three-dimensional geometric model of a hydrogen storage unit; constructing a structure objective function; optimizing the structure objective function to obtain an optimal grid model, constructing a simulation model of the optimal grid model, and performing a simulation test on the simulation model to obtain simulation data; calculating difference data between the simulation data and actual optimal test data, and calibrating the simulation data by using the difference data to obtain calibrated simulation data; performing correlation analysis on the calibration simulation data and the influence parameter of the hydrogen storage unit to obtain an influence weight of the influence parameter, determining the service life of the hydrogen storage unit based on the influence weight, and predicting the service performance of the hydrogen storage unit according to the service life of the hydrogen storage unit. The model closer to the actual working condition can be created, the accuracy of the simulation result is improved, and the design error is reduced.

Description

Technical Field

[0001] The present invention belongs to the technical field of hydrogen storage, and relates to the prediction and evaluation technologies of hydrogen storage units, specifically to a method and system for predicting the service performance of hydrogen storage units. Background Art

[0002] Hydrogen storage units are key components in the process of hydrogen energy storage and transportation, and are widely used in fields such as renewable energy storage, transportation (such as fuel cell vehicles), and industrial gas supply. With the continuous growth of global demand for clean energy, especially the increasing importance of hydrogen energy as a future energy carrier, efficient and safe hydrogen storage units have become one of the core technologies to promote the development of the hydrogen economy.

[0003] Traditional modeling methods often rely on simplified assumptions, ignoring the comprehensive effects of material properties (such as elastic modulus, Poisson's ratio, yield strength, etc.), structural properties (such as wall thickness distribution, shape factor), and boundary constraint conditions. For example, when performing finite element analysis, regular hexahedral elements are usually used. Although the calculation is simple, it is difficult to accurately capture complex geometric shapes and stress concentration areas.

[0004] Due to the mismatch between the model and the actual situation, the simulation results will deviate from the actual performance, resulting in the failure to identify potential problems in the design stage and increasing the cost of later modification. If the stress borne by the hydrogen storage unit during actual operation exceeds the design expectation, it may cause problems such as material fatigue and crack propagation, ultimately leading to container rupture and causing serious safety accidents. Summary of the Invention

[0005] Aiming at the technical problems described in the above background art, that is, the simulation results of hydrogen storage units in the prior art deviate from the actual performance and potential problems cannot be identified in the design stage, the present invention proposes a method and system for predicting the service performance of hydrogen storage units.

[0006] To solve the above technical problems, the present invention adopts the following technical solutions:

[0007] The method for predicting the service performance of the hydrogen storage unit of the present invention includes the following steps:

[0008] S1: Determine the initial grid model by using the three-dimensional geometric model of the hydrogen storage unit;

[0009] S2: Determine the number of grids and the distortion degree of all grids in the initial grid model, and construct a structural objective function by using the number of grids and the distortion degree of all grids in the initial grid model;

[0010] S3: Optimize the structural objective function to obtain the optimal grid model, construct a simulation model of the optimal grid model, and perform simulation tests on the simulation model to obtain simulation data;

[0011] S4: Calculate the difference data between the simulation data and the actual optimal test data, and use the difference data to calibrate the simulation data to obtain calibrated simulation data;

[0012] S5: Conduct a correlation analysis between the calibrated simulation data and the influencing parameters of the hydrogen storage unit to obtain the influence weights of the influencing parameters, determine the service life of the hydrogen storage unit based on the influence weights, and predict the service performance of the hydrogen storage unit according to the service life of the hydrogen storage unit.

[0013] Further defined, the step S1 specifically includes: obtaining the three-dimensional geometric model of the hydrogen storage unit, and performing finite element mesh division on the three-dimensional geometric model to obtain an initial mesh model.

[0014] Further defined, the performing finite element mesh division on the three-dimensional geometric model specifically includes: importing the three-dimensional geometric model into finite element analysis software, setting the initial mesh size value, and selecting unstructured hexahedron elements as the basic mesh elements for finite element mesh division.

[0015] Further defined, the step S2 specifically includes:

[0016] S2.1: Obtain the number of meshes in the initial mesh model and the distortion degree of each mesh, accumulate and calculate the distortion degree of each mesh to obtain the distortion degree of all meshes;

[0017] S2.2: Use the distortion degree of all meshes as the error rate of the initial mesh model, and construct a structural objective function using the relationship that the error rate of the initial mesh model is negatively correlated with the number of meshes.

[0018] Further defined, the step S3 is specifically:

[0019] S3.1: Optimize the structural objective function using a genetic algorithm to obtain the optimal solution of the structural objective function;

[0020] S3.2: Adjust the sizes of each mesh in the initial mesh model according to the optimal solution of the structural objective function to obtain an optimal mesh model.

[0021] Further defined, the step S4 specifically includes:

[0022] S4.1: Import the optimal mesh model into simulation software, set different working condition conditions in the simulation software, and perform simulation on the optimal mesh model under different working condition conditions to obtain simulation data;

[0023] S4.2: Obtain the actual optimal test data of the hydrogen storage unit under different working condition conditions;

[0024] S4.3: Calculate the difference data between the simulation data under different working conditions and the actual optimal test data, and use the difference data to calibrate the simulation data to obtain calibrated simulation data.

[0025] Further defined, the step S5 specifically includes:

[0026] S5.1: Use the calibrated simulation data to determine the optimal working condition among different working conditions;

[0027] S5.2: Calculate the correlation between the calibrated simulation data of the optimal working condition and the influence parameters of the hydrogen storage unit as the influence weight of the influence parameters;

[0028] S5.3: Collect the historical usage data of the hydrogen storage unit, and perform time series data prediction on the historical usage data to obtain the time series data prediction result;

[0029] S5.4: Determine the service life of the hydrogen storage unit according to the influence weight and the time series data prediction result, and predict the service performance of the hydrogen storage unit according to the service life of the hydrogen storage unit.

[0030] Further defined, the distortion degree T i of each grid is calculated by the formula:

[0031]

[0032] In the formula, i is the serial number of the grid; j is the angle serial number of each grid; θ i,j is the j-th angle value in the i-th grid; N is the total number of angles in the i-th grid, and j = 1, 2, 3,..., N.

[0033] Further defined, the structural objective function F is:

[0034] F = T′ * exp(-M)

[0035]

[0036] In the formula, T ′ is the average value of the distortion degrees of all grids; M is the number of grids; T i is the distortion degree of each grid; i is the serial number of the grid.

[0037] The hydrogen storage unit service performance prediction system of the present invention includes:

[0038] Model construction module: used to determine the initial grid model by using the three-dimensional geometric model of the hydrogen storage unit;

[0039] Structure objective function construction module: It is used to determine the number of meshes in the initial mesh model and the distortion degree of all meshes, and construct a structure objective function by using the number of meshes and the distortion degree of all meshes in the initial mesh model;

[0040] Simulation data acquisition module: It is used to optimize the structure objective function to obtain an optimal mesh model, construct a simulation model of the optimal mesh model, and perform simulation tests on the simulation model to obtain simulation data;

[0041] Calibration module: It is used to calculate the difference data between the simulation data and the actual optimal test data, and calibrate the simulation data by using the difference data to obtain calibrated simulation data;

[0042] And prediction module: It is used to perform correlation analysis on the calibrated simulation data and the influencing parameters of the hydrogen storage unit to obtain the influence weights of the influencing parameters, determine the service life of the hydrogen storage unit based on the influence weights, and predict the service performance of the hydrogen storage unit according to the service life of the hydrogen storage unit.

[0043] Compared with the prior art, the technical effect of the present invention is as follows:

[0044] 1. For the method for predicting the service performance of the hydrogen storage unit of the present invention, when constructing the three-dimensional geometric model of the hydrogen storage unit, by constructing a simulation model, various influencing factors are comprehensively considered, a model closer to the actual working conditions can be created, the accuracy of the simulation results can be improved, the design error can be reduced, potential problems can be identified in the design stage, and targeted design adjustments can be made to eliminate the error that the simulation results deviate from the actual performance, thereby improving the reliability and safety of the operation of the hydrogen storage unit.

[0045] 2. The present invention analyzes the historical test data of the product, predicts the long-term performance degradation trend of the hydrogen storage unit, obtains the remaining life, plans the maintenance strategy in advance, reduces the unplanned downtime, extends the equipment life, reduces the operation cost through a reasonable maintenance plan, and improves the overall economic benefits. Brief Description of the Drawings

[0046] Figure 1 It is a schematic diagram of the method for predicting the service performance of the hydrogen storage unit of the present invention;

[0047] Figure 2 It is a schematic diagram of the system for predicting the service performance of the hydrogen storage unit of the present invention. Detailed Embodiments

[0048] The technical solutions of the present invention will be further explained below in conjunction with the drawings and embodiments, but the present invention is not limited to the following described embodiments.

[0049] In the following description, numerous specific details are given to provide a more thorough understanding of the present invention. However, it will be apparent to those skilled in the art that the present invention may be practiced without one or more of these details. In other instances, some well-known technical features are not described to avoid obscuring the present invention.

[0050] Embodiment 1

[0051] This embodiment provides a method for predicting the service performance of a hydrogen storage unit as shown in Figure 1 and includes the following steps:

[0052] S1: Determine an initial grid model using the three-dimensional geometric model of the hydrogen storage unit;

[0053] S2: Determine the number of grids and the distortion degree of all grids in the initial grid model, and construct a structural objective function using the number of grids and the distortion degree of all grids in the initial grid model;

[0054] S3: Optimize the structural objective function to obtain an optimal grid model, construct a simulation model of the optimal grid model, and perform a simulation test on the simulation model to obtain simulation data;

[0055] S4: Calculate the difference data between the simulation data and the actual optimal test data, and calibrate the simulation data using the difference data to obtain calibrated simulation data;

[0056] S5: Conduct a correlation analysis between the calibrated simulation data and the influencing parameters of the hydrogen storage unit to obtain the influence weights of the influencing parameters, determine the service life of the hydrogen storage unit based on the influence weights, and predict the service performance of the hydrogen storage unit according to the service life of the hydrogen storage unit.

[0057] Among them, step S1 specifically includes: obtaining the three-dimensional geometric model of the hydrogen storage unit, and performing finite element mesh division on the three-dimensional geometric model to obtain an initial grid model.

[0058] Among them, obtaining the three-dimensional geometric model of the hydrogen storage unit specifically includes: using a high-precision 3D scanner to perform a full-circle scan of the hydrogen storage unit from multiple angles to ensure that all external surfaces of the hydrogen storage unit are covered and data loss caused by occlusion is avoided, obtaining the point cloud data of the hydrogen storage unit, and using a filtering algorithm to denoise the original point cloud data to remove invalid points and noise data, thereby obtaining the three-dimensional geometric model of the hydrogen storage unit. The filtering algorithm is point cloud mean filtering. Among them, the high-precision 3D scanner is, for example, a laser scanner or a structured light scanner. Performing finite element mesh division on the three-dimensional geometric model specifically includes: importing the three-dimensional geometric model into finite element analysis software, selecting unstructured hexahedral elements as the basic mesh elements for finite element mesh division, and setting the initial mesh size value in the finite element software to obtain the initial mesh model. Among them, the initial mesh size value is exemplarily set to 7 mm, and the finite element analysis software is ANSYS Workbench.

[0059] Step S2 specifically includes:

[0060] S2.1: Obtain the number of meshes in the initial mesh model and the distortion degree of each mesh, accumulate and calculate the distortion degree of each mesh to obtain the distortion degree of all meshes;

[0061] Specifically, obtain the unstructured hexahedral element corresponding to each mesh in the initial mesh model, and calculate the hexahedral element distortion degree based on the ideal inner angle value of 90 0 of the ideal element. For each hexahedral element, select N angles to calculate the distortion degree. Suppose N = 20 angles are selected for calculation, and define the distortion degree T i of the i-th mesh element as:

[0062]

[0063] i is the serial number of the mesh; j is the angle serial number of each mesh; θ i,j is the j-th angle value in the i-th mesh; N is the total number of angles in the i-th mesh, and j = 1, 2, 3,..., N. When the unit is completely ideal, all θ i,j = 90 0 °, and at this time T i = 0; if there is a deviation, then the larger T i is, the more severe the distortion of the unit.

[0064] S2.2: Use the distortion degree of all meshes as the error rate of the initial mesh model, and construct a structural objective function using the relationship that the error rate of the initial mesh model is negatively correlated with the number of meshes.

[0065] The greater the distortion of each grid in the initial grid model, the more severe the distortion of the corresponding hexahedron element of each grid compared to the ideal element. Furthermore, there is a large error between the initial grid model after meshing and the three-dimensional geometric model obtained from actual scanning. Therefore, in order to obtain a fine grid model after meshing the initial grid model, it is necessary to adjust the grid size value to make the grid model after meshing the three-dimensional geometric model more refined. The lower the grid size value, the finer the grid model after meshing, but the higher the computational cost in subsequent simulation calculations. Therefore, in order to obtain a hydrogen storage unit grid model with a balance between accuracy and computational cost, a structural objective function F is constructed:

[0066] F = T' * exp(-M)

[0067] M is the number of grids. The smaller the value of M, the lower the computational requirement for simulation calculations.

[0068] T' is the average distortion of all grids in the grid model when calculating the structural objective function value, that is: The larger the value of T', the greater the error between the grid model after meshing the three-dimensional geometric model and the three-dimensional geometric model. It is necessary to reduce the grid size value. Since the size of the three-dimensional geometric model is fixed, the value of M will increase; T ′ is the average distortion of all grids; T i is the distortion of each grid; i is the serial number of the grid.

[0069] In order to obtain a grid model with a balance between accuracy and computational cost, the structural objective function is constructed through the negative correlation between M and T'. The smaller the value of F, the more the grid model of the hydrogen storage unit meets the expected effect. Among them, the exp() function is an exponential function, and the negative sign is used to achieve negative correlation while realizing data normalization.

[0070] Step S3 is specifically as follows:

[0071] S3.1: Use the genetic algorithm to optimize the structural objective function to obtain the optimal solution of the structural objective function. Specifically, take the size of each grid and the number of grids M as iterative variables, and initialize the size of the grid and the number of grids in the initial grid model as the initial values of the iterative variables. Solve the minimum value (optimal value) of the structural objective function through the genetic algorithm. When using the genetic algorithm to solve the minimum value of the structural objective function, it is necessary to ensure that the number of grids after meshing the three-dimensional geometric model in each iteration process is equal to the total number of all grid elements. For the newly added grid elements in each iteration process, the size of the newly added grid elements is the same as the size of the grids in the initial grid model when meshing the newly added grid elements. Set the genetic algorithm to stop iterating after reaching the iteration stop condition. Among them, the iteration stop condition is that when the number of iterations is equal to 100 times, the genetic algorithm stops iterating.

[0072] S3.2: Adjust the sizes of the meshes in the initial mesh model according to the optimal solution of the structural objective function to obtain the optimal mesh model.

[0073] Step S4 specifically includes:

[0074] S4.1: Import the optimal mesh model into the simulation software, set different working conditions in the simulation software, and perform simulation on the optimal mesh model under different working conditions to obtain simulation data; specifically, set simulation conditions such as material and material density for the optimal mesh model, set working condition simulation settings such as boundary conditions, pressure values, temperature values, and humidity values in the simulation software, and set the simulation stop condition: stop the simulation when the hydrogen leakage value reaches the simulation stop condition, and obtain simulation indexes such as the calculated internal pressure value, local stress value, hydrogen leakage value, and simulation time through simulation calculation as the simulation data.

[0075] Exemplarily: Set an example of the simulation stop condition: the hydrogen leakage value is 0.05L, and an example of the simulation setting for the optimal mesh model conditions: the material is steel, and the material density is 7850 kg / m 3 , Young's modulus: 210 GPa, Poisson's ratio: 0.3, coefficient of thermal expansion: 1.2×10 -5 / K; Set an example of the working condition simulation: boundary condition: fixed constraint, working pressure range: 5 - 10 Pa, temperature range: 10°C - 80°C, humidity range: 30RH - 90RH, obtain the local maximum stress value of the optimal mesh model through static simulation calculation, obtain the maximum displacement of the optimal mesh model through dynamic simulation analysis, obtain the maximum internal pressure value of the optimal mesh model through thermal - structure coupling simulation, and obtain the hydrogen leakage value through fluid dynamics simulation.

[0076] S4.2: Obtain the actual optimal test data of the hydrogen storage unit under different working conditions, where the actual optimal test data is obtained by actually testing the hydrogen storage unit;

[0077] S4.3: Calculate the difference data between the simulation data and the actual optimal test data under different working conditions, and calibrate the simulation data using the difference data to obtain the calibrated simulation data.

[0078] Specifically, actually test the hydrogen storage unit, and stop the test when the hydrogen leakage value reaches the stop test condition. Due to limited actual test conditions and insufficient working condition test conditions, select to test the working conditions under multiple fixed conditions, obtain the measured data of the hydrogen storage unit product through the working conditions under multiple fixed conditions, and use sensors such as pressure sensors, position measuring rulers, stress measuring instruments, and hydrogen sensors to measure the test results, and obtain the local maximum stress value, maximum displacement, maximum internal pressure value, hydrogen leakage value, and test time as the product test data.

[0079] Exemplary: Working condition test environment under various fixed conditions: Working pressure range: 5Pa, 8Pa, 10Pa, temperature range: 10°C, 20°C, …, 80°C, humidity range: 30RH, 40RH, …, 90RH. Measured by a stress measuring instrument, the local maximum stress value of the hydrogen storage unit product is obtained. By a position measuring ruler, the maximum displacement of the hydrogen storage unit product is obtained. By a pressure sensor, the maximum internal pressure value of the hydrogen storage unit product is obtained. By a hydrogen sensor, the real-time hydrogen leakage amount value of the hydrogen storage unit product is obtained.

[0080] In actual tests, the test conditions are limited. For the sampling test results under different working conditions, in order to supplement the test data under the missing test conditions, the least squares method is used to interpolate the test data to obtain the supplemented working condition test data. The weighted average of the supplemented working condition test and simulation data at the same working condition and the same time node is used to obtain the calibrated simulation data. Among them, the weight values for the weighted average of the test and simulation data of the supplemented working condition are both 0.5.

[0081] Step S5 specifically includes:

[0082] S5.1: Determine the optimal working condition among different working condition conditions using the calibrated simulation data;

[0083] S5.2: Calculate the correlation between the calibrated simulation data of the optimal working condition and the influencing parameters of the hydrogen storage unit as the influencing weight of the influencing parameters;

[0084] S5.3: Collect the historical usage data of the hydrogen storage unit, perform time series data prediction on the historical usage data to obtain the time series data prediction result;

[0085] S5.4: Determine the service life of the hydrogen storage unit according to the influencing weight and the time series data prediction result, and predict the service performance of the hydrogen storage unit according to the service life of the hydrogen storage unit.

[0086] Specifically, obtain the usage time of the hydrogen storage unit under different working condition conditions in the calibrated simulation data, obtain the working condition condition corresponding to the longest usage time of the hydrogen storage unit as the optimal working condition, obtain each influencing parameter in the calibrated simulation data under the optimal working condition, and perform data normalization for each influencing parameter under its respective parameter to obtain the normalized data of each influencing parameter. Among them, the data normalization method uses the Z-score normalization method.

[0087] Since the hydrogen leakage amount value data can intuitively reflect the damaged degree of the energy storage unit, the Pearson correlation coefficient is used to calculate the normalized data of the remaining influencing parameters and the correlation between the normalized data of the hydrogen leakage amount value data respectively, and the correlation between the normalized data of the hydrogen leakage amount value data and the normalized data of the remaining influencing parameters is obtained as the influencing weight of the influencing parameters in the service life of the energy storage unit.

[0088] Obtain the historical usage data collected by each sensor during the use of the hydrogen storage unit, perform time series data prediction on the remaining influencing parameters in the historical usage data except for the hydrogen release amount value data, and obtain the prediction results of the remaining influencing parameters. Among them, the time series data prediction algorithm can adopt algorithms such as ARIMA algorithm, LSTM long short-term network, etc.

[0089] Multiply the prediction results of the remaining influencing parameters by their respective influencing parameter weights to obtain the weighted value data of the prediction results of the remaining influencing parameters. Add the weighted value data at the same moment in the prediction results of the remaining influencing parameters to obtain the predicted hydrogen release amount value data predicted by the remaining influencing parameters of the hydrogen storage unit. Obtain the time corresponding to the hydrogen release amount value standard in the predicted hydrogen release amount value data as the service life of the hydrogen storage unit.

[0090] In the present invention, the influencing parameter refers to the material characteristics (such as elastic modulus, Poisson's ratio, yield strength, etc.) and structural characteristics (such as wall thickness distribution, shape factor, etc.) of the hydrogen storage unit.

[0091] Embodiment 2

[0092] This embodiment provides a Figure 2 hydrogen storage unit service performance prediction system as shown, including:

[0093] Model construction module: used to determine the initial grid model using the three-dimensional geometric model of the hydrogen storage unit;

[0094] Structural objective function construction module: used to determine the number of grids and the distortion degree of all grids in the initial grid model, and construct a structural objective function using the number of grids and the distortion degree of all grids in the initial grid model;

[0095] Simulation data acquisition module: used to optimize the structural objective function to obtain the optimal grid model, construct a simulation model of the optimal grid model, and perform simulation tests on the simulation model to obtain simulation data;

[0096] Calibration module: used to calculate the difference data between the simulation data and the actual optimal test data, and calibrate the simulation data using the difference data to obtain calibrated simulation data;

[0097] And a prediction module: used to perform correlation analysis on the calibrated simulation data and the influencing parameters of the hydrogen storage unit to obtain the influencing weights of the influencing parameters, determine the service life of the hydrogen storage unit based on the influencing weights, and predict the service performance of the hydrogen storage unit according to the service life of the hydrogen storage unit.

[0098] The hydrogen storage unit service performance prediction system in this embodiment corresponds exactly to the hydrogen storage unit service performance prediction method in Embodiment 1. For the content not described in detail in the model construction module, structural objective function construction module, simulation data acquisition module, calibration module, and prediction module in this embodiment, please refer to the description in the above-mentioned Embodiment 1 part.

[0099] It should be noted that in this article, relational terms such as "one" and "two" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements includes not only those elements but also other elements not expressly listed, or also includes elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "comprising a..." does not exclude the presence of additional identical elements in the process, method, article or device comprising the element.

[0100] Although the embodiments of the present invention have been shown and described, those of ordinary skill in the art can understand that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for predicting the service performance of a hydrogen storage unit, characterized in that: The following steps are involved: S1: Determine the initial grid model using the three-dimensional geometric model of the hydrogen storage unit; S2: determining the number of meshes in the initial mesh model and the distortion of all meshes, and constructing a structural objective function using the number of meshes in the initial mesh model and the distortion of all meshes; S3: Optimizing the structural objective function to obtain an optimal grid model, constructing a simulation model of the optimal grid model, and performing simulation testing on the simulation model to obtain simulation data; S4: Calculate difference data between the simulation data and the actual optimal test data, and calibrate the simulation data using the difference data to obtain calibrated simulation data; S5: Perform correlation analysis on the calibration simulation data and the influencing parameters of the hydrogen storage unit to obtain the influence weights of the influencing parameters, determine the service life of the hydrogen storage unit based on the influence weights, and predict the service performance of the hydrogen storage unit according to the service life of the hydrogen storage unit.

2. The method for predicting the service performance of a hydrogen storage unit according to claim 1, characterized in that: The step S1 specifically includes: obtaining a three-dimensional geometric model of the hydrogen storage unit, performing finite element meshing on the three-dimensional geometric model, and obtaining an initial mesh model.

3. The method for predicting the service performance of a hydrogen storage unit according to claim 2, characterized in that: The finite element meshing of the three-dimensional geometric model specifically includes: importing the three-dimensional geometric model into finite element analysis software, setting an initial mesh size value, and selecting unstructured hexahedral units as basic mesh units for finite element meshing.

4. The method for predicting the service performance of a hydrogen storage unit according to claim 1, characterized in that: The step S2 specifically includes: S2.1: Obtain the number of meshes in the initial mesh model and the distortion of each mesh, and cumulatively calculate the distortion of each mesh to obtain the distortion of all meshes; S2.2: The distortion of all meshes is taken as the initial mesh model error rate, and the structural objective function is constructed using the negative correlation between the initial mesh model error rate and the number of meshes.

5. The method for predicting the service performance of a hydrogen storage unit according to claim 4, characterized in that: The step S3 is specifically as follows: S3.1: Use genetic algorithm to optimize the structural objective function and obtain the optimal solution of the structural objective function; S3.2: According to the optimal solution of the structural objective function, the size of each grid in the initial grid model is adjusted to obtain the optimal grid model.

6. The method for predicting the service performance of a hydrogen storage unit according to claim 1, characterized in that: The step S4 specifically includes: S4.1: importing the optimal grid model into simulation software, setting different working conditions in the simulation software, simulating the optimal grid model under different working conditions, and obtaining simulation data; S4.2: Obtain actual optimal test data of the hydrogen storage unit under different operating conditions; S4.3: Calculate the difference data between the simulation data under different working conditions and the actual optimal test data, and use the difference data to calibrate the simulation data to obtain calibrated simulation data.

7. The method for predicting the service performance of a hydrogen storage unit according to claim 1, characterized in that: The step S5 specifically includes: S5.1: Determine the optimal operating condition among different operating conditions using calibration simulation data; S5.2: Calculate the correlation between the calibration simulation data of the optimal working condition and the influencing parameters of the hydrogen storage unit as the influencing weight of the influencing parameters; S5.3: Collect historical usage data of the hydrogen storage unit, perform time series data prediction on the historical usage data, and obtain time series data prediction results; S5.4: Determine the service life of the hydrogen storage unit based on the impact weight and the prediction results of the time series data, and predict the service performance of the hydrogen storage unit based on the service life of the hydrogen storage unit.

8. The method for predicting the service performance of a hydrogen storage unit according to claim 4, characterized in that: The distortion T of each mesh i The calculation formula is: Where i is the grid number; j is the angle number of each grid; θ i,j is the jth angle value in the ith grid; N is the total number of angles in the ith grid, j = 1, 2, 3, …, N.

9. The method for predicting the service performance of a hydrogen storage unit according to claim 4, characterized in that: The structural objective function F is: F = T'*exp(-M) Where, T ′ is the average distortion of all grids; M is the number of grids; T i is the distortion of each mesh; i is the mesh number.

10. A hydrogen storage unit service performance prediction system, characterized in that: include: Model building module: used to determine the initial mesh model using the three-dimensional geometric model of the hydrogen storage unit; Structural objective function building module: used to determine the number of grids in the initial grid model and the distortion of all grids, and to build the structural objective function using the number of grids in the initial grid model and the distortion of all grids; Simulation data acquisition module: used to optimize the structural objective function, obtain the optimal grid model, build a simulation model of the optimal grid model, perform simulation testing on the simulation model, and obtain simulation data; Calibration module: used for calculating the difference data between the simulation data and the actual optimal test data, and calibrating the simulation data using the difference data to obtain calibrated simulation data; And a prediction module: used to perform correlation analysis between the calibration simulation data and the influencing parameters of the hydrogen storage unit, obtain the influence weights of the influencing parameters, determine the service life of the hydrogen storage unit based on the influence weights, and predict the service performance of the hydrogen storage unit according to the service life of the hydrogen storage unit.

Images