Hydrogen refueling station hydrogen explosion simulation system under obstacle condition

Abstract

The invention provides a hydrogen refueling station hydrogen explosion simulation system under obstacle conditions, and relates to the technical field of hydrogen energy safety application. The system comprises a data acquisition module used for collecting environmental conditions of the hydrogen refueling station and geometric features of obstacles and generating a random number set of obstacle position feature vectors; the calculation processing module is used for constructing a three-dimensional hydrogen refueling station model in a fluid dynamic model by using the data, simulating an explosion process and outputting a calculation result; the support vector regression prediction module is used for predicting explosion pressure under different obstacle layouts by taking the calculation result as input; and the training interaction module is used for providing a virtual environment interaction function so as to allow a user to adjust obstacle layout and / or explosion conditions, and response and update explosion prediction result information in real time. According to the system, the prediction precision is improved, the interactivity is enhanced, and the obstacle layout can be flexibly adjusted according to needs.

Description

Technical Field

[0001] The present application relates to the technical field of safe application of hydrogen energy, and in particular to a hydrogen explosion simulation system for a hydrogen refueling station under obstacle conditions. Background Art

[0002] As the global demand for clean energy grows, hydrogen, as an important representative of clean energy, has shown great potential in automobile manufacturing and the hydrogen energy industry. As a key link in the hydrogen energy industry chain, the safety management of hydrogen refueling stations has become an issue that needs to be urgently addressed.

[0003] Existing hydrogen explosion research technologies mainly include experimental methods and traditional numerical simulation methods. Among them, although the experimental method is direct, it is costly, dangerous, and difficult to reproduce diverse scenarios. Although the traditional numerical simulation method can simulate the basic process, it lacks the ability to dynamically analyze the relationship between the layout of diverse obstacles and the explosion results, and lacks real-time interaction and visualization capabilities, which limits its effectiveness in practical applications.

[0004] Therefore, it is particularly important to develop a hydrogen explosion simulation system that can flexibly adjust obstacle layout, improve prediction accuracy and enhance interactivity. Summary of the invention

[0005] The purpose of the present application is to provide a hydrogen explosion simulation system for a hydrogen refueling station under obstacle conditions, which can improve prediction accuracy and enhance interactivity, and can flexibly adjust the obstacle layout as needed.

[0006] This application is implemented as follows: In a first aspect, the present application provides a hydrogen explosion simulation system for a hydrogen refueling station under obstacle conditions, including: a data acquisition module, used to obtain the environmental conditions of the hydrogen refueling station and the geometric characteristics of the obstacles, and obtain a random number set of obstacle position feature vectors; a calculation processing module, used to use the random number set of obstacle position feature vectors in a computational fluid dynamics model to establish a three-dimensional model of the hydrogen refueling station, simulate the explosion process, and output a three-dimensional calculation result data set; a support vector regression prediction module, used to use a support vector regression prediction model, with the three-dimensional calculation result data set as input data, to predict the pressure generated by the hydrogen explosion under different obstacle layouts, and obtain explosion prediction result information; a training interaction module, used to provide interactive operation functions in a virtual environment, so as to allow the user to adjust the obstacle layout and / or explosion conditions, and respond and update the explosion prediction result information in real time.

[0007] In a second aspect, the present application provides a method for simulating hydrogen explosions at a hydrogen refueling station under obstacle conditions, which includes the following steps: obtaining the environmental conditions of the hydrogen refueling station and the geometric characteristics of the obstacles to obtain a random number set of obstacle position feature vectors; using the random number set of obstacle position feature vectors in a computational fluid dynamics model to establish a three-dimensional model of the hydrogen refueling station, simulate the explosion process, and output a three-dimensional calculation result data set; using a support vector regression prediction model, using the three-dimensional calculation result data set as input data, predicting the pressure generated by the hydrogen explosion under different obstacle layouts, and obtaining explosion prediction result information; based on the interactive operation function in the preset virtual environment, responding to the user's adjustment of the obstacle layout and / or explosion conditions, updating the explosion prediction result information.

[0008] In a third aspect, the present application provides an electronic device, comprising a memory for storing one or more programs; a processor; when the one or more programs are executed by the processor, the method described in the second aspect is implemented.

[0009] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method described in the second aspect above.

[0010] Compared with the prior art, the present invention has at least the following advantages or beneficial effects: This application proposes a hydrogen explosion simulation system for hydrogen refueling stations under obstacle conditions. Through the close cooperation of the data acquisition module and the calculation and processing module, the system can accurately reflect the actual environment and obstacle layout of the hydrogen refueling station, thereby realizing accurate simulation of the hydrogen explosion process. This provides users with more reliable data support and helps to optimize the safety design and emergency response strategy of hydrogen refueling stations. The support vector regression prediction module can establish an accurate prediction model based on hydrogen explosion data under different obstacle layouts. This provides users with explosion pressure prediction results under different conditions, which helps to conduct risk assessment and formulate preventive measures. The training interaction module provides interactive operation functions in a virtual environment, allowing users to perform simulation drills and parameter adjustments. This reduces the cost and risk of traditional training and drills, while improving users' actual operating capabilities and emergency response speed. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0012] Figure 1 This is a structural block diagram of an embodiment of a hydrogen explosion simulation system for a hydrogen filling station under obstacle conditions of the present application; Figure 2 This is a top view of an explosion storage tank at a hydrogen refueling station in one embodiment of the present application; Figure 3 A top view of the monitoring point position and a schematic diagram of the air inlet and outlet in one embodiment of the present application; Figure 4 This is a schematic diagram of the relative positions of monitoring points and obstacles in one embodiment of the present application; Figure 5 This is a schematic diagram of grid independence verification in an embodiment of the present application; Figure 6 A structural block diagram of an electronic device provided in an embodiment of the present application.

[0013] Icon: 201, processor; 202, memory; 203, communication interface. DETAILED DESCRIPTION

[0014] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application described and shown in the drawings here can be arranged and designed in various different configurations.

[0015] Therefore, the following detailed description of the embodiments of the present application provided in the accompanying drawings is not intended to limit the scope of the present application for which protection is sought, but merely represents selected embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without creative work are within the scope of protection of the present application.

[0016] In conjunction with the accompanying drawings, some embodiments of the present application are described in detail below. In the absence of conflict, the following embodiments and features in the embodiments can be combined with each other.

[0017] Example 1 With the surge in global demand for clean energy, hydrogen has become increasingly important as a clean energy source, especially in the automobile manufacturing and hydrogen energy industries. As the key to the hydrogen energy industry chain, the safety management of hydrogen refueling stations is crucial. However, existing hydrogen explosion research technologies have obvious shortcomings. Experimental methods are costly, dangerous, and difficult to reproduce diverse scenarios. Traditional numerical simulation methods lack dynamic analysis, real-time interaction, and visualization capabilities.

[0018] In this regard, an embodiment of the present application provides a hydrogen explosion simulation system for a hydrogen refueling station under obstacle conditions, which not only improves the prediction accuracy but also greatly enhances the interactivity.

[0019] See also Figure 1 The hydrogen explosion simulation system of a hydrogen refueling station under obstacle conditions includes: a data acquisition module, which is used to obtain the environmental conditions of the hydrogen refueling station and the geometric characteristics of the obstacles, and obtain a random number set of obstacle position feature vectors; a calculation processing module, which is used to use the random number set of obstacle position feature vectors in a computational fluid dynamics model to establish a three-dimensional model of the hydrogen refueling station, simulate the explosion process, and output a three-dimensional calculation result data set; a support vector regression prediction module, which is used to use a support vector regression prediction model and the three-dimensional calculation result data set as input data to predict the pressure generated by the hydrogen explosion under different obstacle layouts and obtain explosion prediction result information; a training interaction module, which is used to provide an interactive operation function in a virtual environment to allow the user to adjust the obstacle layout and / or explosion conditions, and respond and update the explosion prediction result information in real time.

[0020] In the above embodiment, the data acquisition module is responsible for collecting the environmental conditions of the hydrogen refueling station (such as ambient temperature, atmospheric pressure and wind speed, etc.) and the geometric characteristics of obstacles (such as the distance of the obstacle from the explosion source, the width of the obstacle, the height of the obstacle and the number of obstacles, etc.). These data are obtained through sensors or manual input, and converted into a random number set of obstacle position feature vectors. This ensures that the system can accurately reflect the physical environment in the real scene, and enables the simulation system to be applicable to hydrogen explosion simulations under different environments and obstacle layouts, enhancing the versatility and flexibility of the system.

[0021] The computational processing module receives the random number set of obstacle position feature vectors output by the data acquisition module, and builds a three-dimensional model of the hydrogen refueling station in the computational fluid dynamics (CFD) model. Then, the module simulates the explosion process and outputs a three-dimensional calculation result data set. These data sets contain key information such as hydrogen concentration distribution, temperature distribution, and pressure distribution. Through the CFD model, it can accurately simulate the explosion process and provide accurate data support for subsequent support vector regression prediction. It also improves the computational efficiency and accuracy of the simulation system, enabling the system to generate high-quality simulation results in a short time.

[0022] The support vector regression prediction module uses the support vector regression (SVR) prediction model and the three-dimensional calculation result data set output by the calculation processing module as input data to predict the pressure generated by hydrogen explosion under different obstacle layouts. By training the SVR model, the module establishes a mapping relationship between obstacle layout and explosion pressure, and can output corresponding explosion prediction result information based on the obstacle layout parameters input by the user. This provides accurate prediction of explosion pressure under different obstacle layouts, providing strong support for safety risk assessment and emergency response of hydrogen refueling stations. It can improve the accuracy and reliability of prediction results and reduce the uncertainty of safety risk assessment and emergency response.

[0023] The training interaction module provides interactive operation functions in a virtual environment, allowing users to adjust obstacle layout and / or explosion conditions, and respond to and update explosion prediction result information in real time. Users can use this module to perform simulation exercises, adjust parameters, and other operations to better understand and respond to the risks of hydrogen explosions. As a result, hydrogen explosions at hydrogen refueling stations can be simulated and drilled in a virtual environment, reducing the cost and risk of training and drills. In addition, users can adjust obstacle layout and explosion conditions as needed to more comprehensively assess explosion risks.

[0024] In summary, through the close collaboration of the data acquisition module and the calculation and processing module, the system can accurately reflect the actual environment and obstacle layout of the hydrogen refueling station, thereby realizing the accurate simulation of the hydrogen explosion process. This provides users with more reliable data support, which helps to optimize the safety design and emergency response strategy of the hydrogen refueling station. The support vector regression prediction module can establish an accurate prediction model based on the hydrogen explosion data under different obstacle layouts. This provides users with explosion pressure prediction results under different conditions, which helps to conduct risk assessment and formulate preventive measures. The training interaction module provides interactive operation functions in a virtual environment, allowing users to perform simulation drills and parameter adjustments. This reduces the cost and risk of traditional training and drills, while improving users' actual operation capabilities and emergency response speed.

[0025] That is, the above embodiment not only improves the accuracy and practicality of the simulation, but also enhances the visualization effect and intuitiveness, providing users with more comprehensive and reliable data support. At the same time, through functions such as predictive analysis and risk assessment, this technical solution also helps to improve users' understanding and response capabilities of hydrogen explosion risks and reduce the cost and risk of training and drills.

[0026] Based on the above scheme, in some implementations of the present application, the support vector regression prediction module uses a Gaussian radial basis kernel function for nonlinear fitting. The expression of the Gaussian radial basis kernel function is: ,in and Respectively represent two samples in the sample space, represents the Euclidean distance between two samples, It is a hyperparameter of the Gaussian kernel function, which controls the width of the kernel function.

[0027] It should be noted that in the support vector regression prediction module, the Gaussian radial basis kernel function is used to map the input data from the original space to a high-dimensional feature space, in which the originally linearly inseparable data may become linearly separable or easier to linearly fit. By performing linear regression in the high-dimensional feature space, the support vector regression prediction module can fit the nonlinear relationship in the original space.

[0028] Since the relationship between explosion pressure and obstacle layout in hydrogen refueling station hydrogen explosion simulation is often a complex nonlinear relationship, the Gaussian kernel function can provide more accurate predictions. In addition, by performing linear regression in high-dimensional feature space, the support vector regression prediction module can avoid the problem of overfitting and show good generalization performance on unknown data. This helps to accurately predict hydrogen explosion pressure under different obstacle layouts.

[0029] Based on the aforementioned scheme, in some implementations of the present application, the function coefficient of the support vector regression prediction model is set to 0.1, the interval band width is set to 0.1, and the penalty factor of samples outside the isolation band is set to 100.

[0030] It should be noted that the above function coefficients refer to parameters related to the kernel function. By setting them to 0.1, the risk of overfitting can be reduced. The interval band width (the width of the ε-insensitive loss function) is a key parameter in the support vector regression prediction model, which defines the model's tolerance for prediction errors. When the prediction error is within ±ε, the loss function is 0, that is, the support vector regression prediction model does not penalize these errors. In the above implementation, setting the interval band width to 0.1 means that the model does not impose a penalty on samples with a prediction error less than 0.1, which helps the support vector regression prediction model reduce sensitivity to data noise while maintaining prediction accuracy. In addition, in the above implementation, by setting the penalty factor for samples outside the isolation band to 100, the prediction error of samples outside the isolation band can be minimized.

[0031] That is, by carefully setting the function coefficients of the support vector regression prediction model, the width of the interval band, and the penalty factor of the samples outside the isolation band in the above implementation method, a balance point is found between overfitting and underfitting, so that the model is neither too complex nor too simple. While maintaining a certain generalization ability, it can better fit the key features in the training data, thereby improving the prediction accuracy.

[0032] Based on the aforementioned scheme, in some implementations of the present application, the computing and processing module, when establishing a three-dimensional model of a hydrogen refueling station, includes: setting a computing domain, monitoring points, and flow channels, and meshing the three-dimensional model to simulate the pressure distribution after a hydrogen explosion, and obtain a hydrogen storage tank explosion shock wave cloud map under obstacles at different locations, as well as the explosion pressure value of each monitoring point.

[0033] It should be noted that the computational domain is the physical space range considered during the simulation process. It must be large enough to include the hydrogen station and its surrounding areas to ensure that the propagation of the explosion shock wave is not limited by the boundaries of the computational domain. The setting of the computational domain should be determined based on the layout of the hydrogen station, the distribution of obstacles, and the expected explosion impact range. The monitoring points are specific locations selected within the computational domain to record the changes in key parameters such as hydrogen concentration, temperature, and pressure during the simulation process. By setting monitoring points at these locations, detailed information at different locations during the explosion process can be obtained to provide data support for subsequent analysis and prediction. The number and location of the monitoring points should be determined based on the actual situation of the hydrogen station and the simulation requirements. The flow channel is the diffusion path of hydrogen after leakage, and it is also the main channel for the propagation of the explosion shock wave. In the calculation processing module, the shape, size, and location of these channels need to be accurately described to ensure the accuracy of the simulation results. The setting of the flow channel should be based on information such as the pipeline layout, valve position, and obstacle distribution of the hydrogen station.

[0034] Meshing is the process of dividing the computational domain into a series of small, interconnected units (i.e., meshes). In the computational processing module, professional meshing tools or algorithms can be used to generate high-quality meshes based on the three-dimensional model of the hydrogen refueling station and simulation requirements. The meshing should ensure sufficient resolution in key areas (such as near obstacles, flow channels, etc.) to capture details of the explosion process.

[0035] After the three-dimensional model of the hydrogen refueling station is established and meshed, the computational processing module will use the computational fluid dynamics (CFD) model for simulation. During the simulation, the leakage, diffusion and explosion process of hydrogen will be considered, and the shock wave cloud map of the hydrogen storage tank explosion under obstacles at different locations and the explosion pressure value of each monitoring point will be calculated. These results will be used to evaluate the safety of the hydrogen refueling station and provide a scientific basis for the formulation of emergency response strategies.

[0036] Thus, by executing the above steps, the calculation processing module can establish an accurate three-dimensional model of the hydrogen refueling station and simulate the pressure distribution after the hydrogen explosion. This can not only provide a detailed explosion shock wave cloud map and the explosion pressure value of each monitoring point, but also provide strong support for the safety design and emergency response strategy of the hydrogen refueling station.

[0037] Based on the above scheme, in some implementations of the present application, the computing and processing module uses an adaptive grid refinement algorithm when establishing a three-dimensional model of a hydrogen refueling station, and dynamically adjusts the grid density according to the complexity of hydrogen leakage and diffusion.

[0038] In the above implementation, the accuracy and efficiency of the simulation can be improved by introducing an adaptive mesh refinement algorithm. Among them, by introducing an adaptive mesh refinement algorithm, the density of the mesh can be automatically adjusted by analyzing the changes in the physical field (such as hydrogen concentration, temperature, pressure, etc.) during the simulation process. In areas where the physical field changes dramatically (such as near the hydrogen leak source, around obstacles, etc.), the algorithm will generate a denser mesh to capture more detailed information; while in areas where the physical field changes gently, a sparser mesh is used to reduce the amount of calculation.

[0039] Specifically, the introduction of the adaptive mesh refinement algorithm can play the following roles: (1) Dynamically adjust the grid density: Based on the simulation results of hydrogen leakage and diffusion, the algorithm can adjust the grid density in real time. The grid will be automatically refined near the hydrogen leakage source, around obstacles, and in areas where the explosion shock wave may propagate, in order to capture subtle changes in hydrogen concentration and the propagation process of the explosion shock wave.

[0040] (2) Improving simulation accuracy: By dynamically adjusting the grid density, the algorithm can ensure sufficient resolution in key areas, thereby more accurately simulating the process of hydrogen leakage and diffusion, as well as the propagation characteristics of the explosion shock wave. This helps to improve the accuracy and reliability of the simulation results.

[0041] (3) Optimizing computational efficiency: In areas where the physical field changes slowly, the algorithm will generate a sparser grid to reduce the amount of computation. This helps to improve computational efficiency and shorten simulation time while ensuring simulation accuracy.

[0042] (4) Automatic processing of complex situations: The adaptive grid refinement algorithm can automatically process complex situations in the hydrogen leakage and diffusion process, such as multiphase flow, turbulence, chemical reactions, etc. This makes the algorithm more adaptable and robust when simulating complex physical phenomena such as hydrogen explosions at hydrogen refueling stations.

[0043] Based on the aforementioned scheme, in some implementations of the present application, the initial conditions and boundary conditions set by the computing and processing module when establishing a three-dimensional model of the hydrogen refueling station include environmental parameters around the explosion location and the settings of the hydrogen inlet, air inlet, and outlet.

[0044] In the above implementation, by carefully setting the initial conditions and boundary conditions, the calculation processing module can more accurately simulate the process of hydrogen explosion at the hydrogen refueling station and obtain more reliable simulation results. This not only helps to evaluate the safety of hydrogen refueling stations, but also provides a more scientific basis for formulating emergency response strategies.

[0045] Specifically, the initial condition refers to the distribution of various physical fields (such as hydrogen concentration, temperature, pressure, etc.) in the calculation domain at the beginning of the simulation. When the calculation processing module establishes a three-dimensional model of the hydrogen refueling station, the initial conditions that need to be set mainly include: (1) Hydrogen concentration distribution: According to the actual layout of the hydrogen refueling station and the location of the hydrogen storage tank, set the concentration distribution of hydrogen in the storage tank and its surrounding area at the initial moment. Factors such as the rate of hydrogen leakage, the size and shape of the leakage port need to be considered. (2) Temperature distribution: Set the temperature distribution in the calculation domain at the initial moment. In the hydrogen refueling station environment, the temperature may be affected by many factors, such as ambient temperature, solar radiation, equipment heating, etc. Therefore, it is necessary to set a reasonable temperature distribution according to the actual situation. (3) Pressure distribution: Set the pressure distribution in the calculation domain at the initial moment. When the hydrogen refueling station is operating normally, the pressure is usually kept within a certain range. However, during hydrogen leakage and explosion, the pressure may change significantly. Therefore, it is necessary to set a reasonable initial pressure distribution to reflect the actual situation of the hydrogen refueling station.

[0046] Boundary conditions refer to the changing rules or constraints of physical fields (such as hydrogen concentration, temperature, pressure, etc.) on the boundaries of the computational domain. When the computational processing module establishes a three-dimensional model of a hydrogen refueling station, the boundary conditions that need to be set mainly include: (1) Environmental parameters around the explosion location: This includes the distribution of obstacles, topography, building structure, etc. around the explosion location. These parameters have an important impact on the propagation and attenuation of the explosion shock wave. Therefore, it is necessary to set reasonable environmental parameters based on the actual situation to reflect the actual situation of the explosion location. (2) Hydrogen inlet: Set the inlet position and leakage rate of hydrogen leakage. This needs to be determined based on the actual layout of the hydrogen refueling station and the leakage characteristics of the hydrogen storage tank. During the simulation, hydrogen will enter the computational domain through these inlets and diffuse into the surrounding environment. (3) Air inlet: Set the inlet position and flow rate of air entering the computational domain. Air is an indispensable factor in the diffusion and explosion process of hydrogen. By setting a reasonable air inlet position and flow rate, the mixing process of air and hydrogen and the propagation process of the explosion shock wave in the air can be simulated. (4) Exit: Set the outlet position and conditions for gas outflow in the computational domain. During the simulation, as hydrogen diffuses and the explosion shock wave propagates, some gas will flow out of the computational domain. By setting reasonable outlet locations and conditions, the gas outflow process can be simulated and its impact on the physical field in the computational domain can be considered.

[0047] Based on the aforementioned scheme, in some implementations of the present application, the system also includes a result analysis module for performing in-depth analysis of the three-dimensional calculation result data set and explosion prediction result information, and providing statistical and visual display of key parameters including explosion impact range and maximum explosion pressure.

[0048] In the above implementation, by introducing the result analysis module, the following effects can be achieved: (1) Improving decision-making efficiency: By providing intuitive and accurate information on the explosion impact range and maximum explosion pressure, the module helps decision makers respond faster and develop more effective emergency response strategies. (2) Optimizing safety design: The data and visual display provided by the module provide an important reference for the safety design of hydrogen refueling stations. Designers can adjust the layout of hydrogen refueling stations, obstacle settings, etc. based on this information to reduce the risk of explosion. (3) Enhancing public trust: By displaying detailed results and visual analysis of explosion simulations, the public's trust and understanding of the safety of hydrogen refueling stations can be enhanced. This will help promote the sustainable development of the hydrogen energy industry.

[0049] Exemplarily, when conducting in-depth analysis, the following steps may be included: (1) Data processing: The result analysis module first receives the three-dimensional calculation result data set and explosion prediction result information from the calculation processing module. These data usually include physical field information such as hydrogen concentration distribution, temperature distribution, pressure distribution, and the propagation characteristics of the explosion shock wave. (2) Key parameter extraction: The module extracts key parameters such as the explosion impact range and maximum explosion pressure from the huge data set through algorithms and technical means. These parameters are crucial for evaluating the safety of hydrogen refueling stations, formulating emergency response strategies, and optimizing safety designs. (3) Statistical analysis: The module performs statistical analysis on the extracted key parameters, including the calculation of statistical quantities such as mean, standard deviation, maximum, and minimum. This helps to understand the overall characteristics and trends of the explosion process.

[0050] For example, the following methods may be included when performing visualization: (1) Display of the explosion impact range: The result analysis module uses visualization technology to display the explosion impact range in an intuitive graphical manner. This usually includes a two-dimensional plane map and a three-dimensional stereo map, which can clearly show the propagation path of the explosion shock wave, the affected area, and the impact of obstacles on the explosion process. (2) Display of maximum explosion pressure: The module will also display the location and value of the maximum explosion pressure. This is usually presented in the form of color coding or contour maps, so that users can intuitively see which areas are hit by the greatest explosion pressure.

[0051] Based on the aforementioned scheme, in some implementations of the present application, the system further includes a model accuracy testing module for testing the support vector regression prediction model using a test set segmented from an experimental data set to verify the accuracy and reliability of the model.

[0052] In the above implementation, by introducing the model accuracy test module, the problems existing in the model can be discovered in time and optimized, thereby improving the prediction accuracy and reliability of the model. In addition, the model that has been rigorously tested can win the trust of users, thereby improving the overall credibility of the hydrogen explosion simulation system of the hydrogen refueling station.

[0053] Based on the aforementioned scheme, in some implementations of the present application, the system also includes an optimization algorithm module for automatically searching and recommending the optimal obstacle layout scheme based on the explosion prediction result information to minimize the hazards that may be caused by the hydrogen explosion.

[0054] It should be noted that by introducing the optimization algorithm module, it is possible to analyze key information in the explosion prediction results, such as the explosion impact range, maximum explosion pressure, etc., and then evaluate the impact of different obstacle layouts on explosion hazards. Based on these analyses, the module can automatically search and recommend the optimal obstacle layout plan to reduce the risk and harm of explosion accidents.

[0055] In order to make the technical personnel in this field understand the present application more intuitively, a specific example will be used here to illustrate. In this example, the high-pressure hydrogen storage bottle of the 35MPa hydrogen refueling station commonly used in China is taken as the research object, and the data processing process of the hydrogen explosion simulation system of the hydrogen refueling station under obstacle conditions is explained (the training and verification process of the support vector regression prediction model is also given).

[0056] Step 1: Establish a random number set of obstacle position feature vectors. Input the environmental conditions of the hydrogen refueling station and the geometric characteristics of the obstacles in the data acquisition module. The geometric characteristics of the obstacles include: the distance from the explosion source (i.e., the high-pressure hydrogen storage bottle), the width of the obstacle, the width of the obstacle, and the number of obstacles. The length of the obstacle from the high-pressure hydrogen storage bottle is recorded as feature vector 1, the width of the obstacle is recorded as feature vector 2, the height of the obstacle is recorded as feature vector 3, and the number of obstacles is recorded as feature vector 4. See Table 1 for details.

[0057] EXCEL was used to generate random arrays of four sets of eigenvectors. The distance between the obstacle and the tank ranged from 3m to 12m (including 3m and 12m); the sum of the obstacle widths ranged from 1m to 30m (including 1m and 30m); the height was calculated from the blocking ratio and the width, and the obstacle thickness was set to 0.5m without considering the stability of the building; the obstacle interval was 2 meters.

[0058] Table 1 Obstacle location data

[0059] Step 2: Use Fluent software to build a three-dimensional model of the hydrogen refueling station. In the calculation processing module, according to the random number set of obstacle position feature vectors obtained in step 1, use CFD software Fluent to build Figure 2 The model shown. The volume of the high-pressure hydrogen storage tank in the hydrogen refueling station is 890L, the pressure is 35MPa~43.8MPa, and the three-dimensional calculation domain range is 50m×50m×50m. The overall explosion of the storage tank with the most serious consequences is selected as the research condition. Figure 2 This is a top view of the tank location. The tank is 20m away from the lower boundary and 30m away from the upper boundary. The tank is centered on the left and right. The total cross-sectional area of ​​the flow channel in the computational domain is 2500m2. A 5×5 matrix monitoring point is set. The monitoring point is 3m away from the tank, the monitoring point spacing is 3m, and the monitoring point is 18m away from the tank at most. One side of the computational domain is set as the air inlet ( Figure 3 The other side is the air outlet ( Figure 3 The monitoring points are distributed as shown in the figure below: Figure 3 , Figure 4 As shown (the cross circle symbol in the figure is the monitoring point).

[0060] The mesh software was used to mesh the three-dimensional model. Taking the 21st group of obstacle data in Table 1 as an example, the influence of three mesh sizes of 1.5m, 2m, and 2.5m on the solution results was verified. The number of units was 272394, 185398, and 127492, respectively. Under the three mesh conditions, the maximum difference in pressure values ​​was 0.2%. The curve of pressure change with time steps after hydrogen explosion was obtained, as shown in Figure 2. Figure 5 In order to ensure the accuracy of the calculation results and improve the computer's operating speed, a grid size of 2m is selected for simulation.

[0061] Step 3: Obtain the three-dimensional calculation result data set as the input data in the support vector regression prediction module. For the three blocking ratios and 24 obstacles in different positions in Table 1, step 20 is selected as the stable simulation result based on simulation experience. The shock wave cloud map of the hydrogen storage tank explosion in step 20 obtained by simulation calculation.

[0062] Fluent simulation outputs a total of 24 sets of simulation data, each set of data includes the explosion pressure values ​​of 25 monitoring points for 20 steps, and the pressure values ​​of the 25 monitoring points corresponding to the 4 eigenvectors are used as input data in the simulation support vector regression (SVR) prediction module. Taking the first and second sets of eigenvector sets as examples, the monitoring point data obtained are shown in Table 2.

[0063] Table 2 Pressure data of the first and second groups of characteristic vector monitoring points

[0064] Step 4: Train the support vector regression prediction model. The above 24 sets of feature vectors and the corresponding monitoring points are input into the SVR model as data sets. The model is trained using the training set, and the hyperparameters are adjusted to optimize the model performance. The performance of the model is evaluated using the validation set to ensure the accuracy and reliability of the prediction model. The distance from the obstacle to the tank, the width of the obstacle, the height of the obstacle, and the number of obstacles are set as feature vectors feature1, feature2, feature3, and feature4, respectively. The explosion pressure measured at different feature vector values ​​is set as the target.

[0065] Before model training, the data was preprocessed. The read_csv function of the Pandas library was called to read the experimental data in the experimental data file. The distance from the obstacle to the tank, the width of the obstacle, the height of the obstacle, and the number of obstacles were read in as feature vectors, and the pressure value was read in as target. The shuffle function of the Pandas library was called to shuffle the experimental data to prevent overfitting during training and improve the generalization ability of the model. The train_test_split function was used to split the experimental data set into a training data set and a test data set. In order to retain sufficient data for model training, the training data set is large, accounting for 80% of the data set, and the test data set accounts for 20% of the data set.

[0066] Furthermore, the Gaussian radial basis kernel function is selected for nonlinear fitting. The Gaussian radial basis kernel function is a commonly used kernel function, and its expression is: (1) Among them, and Respectively represent two samples in the sample space, represents the Euclidean distance between two samples, It is a hyperparameter of the Gaussian kernel function, which controls the width of the kernel function.

[0067] The SVR function in the sklearn library is called for training. In terms of parameter setting, the function coefficient γ is set to 0.1, the interval band width ε is set to 0.1, and the error term, that is, the penalty factor C of the samples outside the isolation band, is set to 100.

[0068] Step 5: Model accuracy test: In the process of model evaluation, the model is first tested using a test set separated from the experimental data set. It should be emphasized that the test data set used in this application does not participate in model training, and the test results are objective.

[0069] The prediction accuracy of the SVR model is calculated to be 0.859649. During the test, 10% of the experimental data was used as test data. The test results showed that the prediction accuracy of the model was greater than 85%, which met the prediction accuracy requirements.

[0070] The blocking ratio is selected as 15%, and two groups of obstacle positions are set to predict and verify the prediction level of SVR. The positions of the obstacles are shown in Table 3.

[0071] Table 3 Obstacle locations

[0072] The above two groups of obstacles are modeled and calculated, and the corresponding pressures of 25 monitoring points are shown in Tables 4 and 5. The predicted pressure of the first group of obstacles is 101356.7802887Pa, and the accuracy of the predicted value compared with the maximum value is 0.999863055; the accuracy of the predicted value compared with the mean is 0.999760242. The predicted pressure of the second group of obstacles is 101359.02528595Pa, and the accuracy of the predicted value compared with the maximum value is 0.999997777; the accuracy of the predicted value compared with the mean is 0.999691595.

[0073] The predicted results are highly consistent with the simulation data, verifying the accuracy and reliability of the model.

[0074] Table 4 Pressure at each monitoring point of obstacle 1

[0075] Table 5 Pressure at each monitoring point of obstacle 2

[0076] Step 6: Users can adjust obstacle layout or explosion conditions through the interactive operation function in the virtual environment provided by the system. The system will update the explosion prediction result information output by the support vector regression prediction module in real time and provide feedback to the user on the predicted pressure value of each monitoring point after adjustment. This real-time feedback mechanism helps users to more quickly evaluate the explosion consequences of different obstacle layouts and make more accurate decisions.

[0077] Example 2 The present application provides a method for simulating hydrogen explosion at a hydrogen filling station under obstacle conditions, which comprises the following steps: The environmental conditions of the hydrogen refueling station and the geometric characteristics of the obstacles are obtained to obtain a random number set of obstacle position feature vectors; a three-dimensional model of the hydrogen refueling station is established using the random number set of obstacle position feature vectors in a computational fluid dynamics model, and the explosion process is simulated to output a three-dimensional calculation result data set; a support vector regression prediction model is used to use the three-dimensional calculation result data set as input data to predict the pressure generated by the hydrogen explosion under different obstacle layouts to obtain explosion prediction result information; based on the interactive operation function in the preset virtual environment, the explosion prediction result information is updated in response to the user's adjustment of the obstacle layout and / or explosion conditions.

[0078] For the specific implementation process of the above method, please refer to the hydrogen explosion simulation system for a hydrogen filling station under obstacle conditions provided in Example 1, which will not be repeated here.

[0079] Example 3 See also Figure 6 , the embodiment of the present application provides an electronic device, which includes at least one processor 201 and at least one memory 202; wherein the processor 201 and the memory 202 are directly connected to each other, or communicate with each other through a communication interface 203, or are electrically connected through one or more communication buses or signal lines to achieve data transmission or interaction; the memory 202 stores program instructions that can be executed by the processor 201, and the processor 201 calls the program instructions to execute a method for simulating hydrogen explosion in a hydrogen filling station under obstacle conditions. For example, the following is realized: The environmental conditions of the hydrogen refueling station and the geometric characteristics of the obstacles are obtained to obtain a random number set of obstacle position feature vectors; a three-dimensional model of the hydrogen refueling station is established using the random number set of obstacle position feature vectors in a computational fluid dynamics model, and the explosion process is simulated to output a three-dimensional calculation result data set; a support vector regression prediction model is used to use the three-dimensional calculation result data set as input data to predict the pressure generated by the hydrogen explosion under different obstacle layouts to obtain explosion prediction result information; based on the interactive operation function in the preset virtual environment, the explosion prediction result information is updated in response to the user's adjustment of the obstacle layout and / or explosion conditions.

[0080] Among them, the memory 202 can be, but is not limited to, a random access memory (RAM), a read only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable read-only memory (EEPROM), etc.

[0081] The processor 201 may be an integrated circuit chip with signal processing capability. The processor 201 may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0082] Understandably, Figure 6 The structure shown is for illustration only. The electronic device may also include Figure 6 More or fewer components as shown, or with Figure 6 Different configurations shown. Figure 6 Each component shown in the figure can be implemented by hardware, software or a combination thereof.

[0083] Example 4 The present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor 201, implements a method for simulating hydrogen explosion at a hydrogen filling station under obstacle conditions. For example, the following is implemented: The environmental conditions of the hydrogen refueling station and the geometric characteristics of the obstacles are obtained to obtain a random number set of obstacle position feature vectors; a three-dimensional model of the hydrogen refueling station is established using the random number set of obstacle position feature vectors in a computational fluid dynamics model, and the explosion process is simulated to output a three-dimensional calculation result data set; a support vector regression prediction model is used to use the three-dimensional calculation result data set as input data to predict the pressure generated by the hydrogen explosion under different obstacle layouts to obtain explosion prediction result information; based on the interactive operation function in the preset virtual environment, the explosion prediction result information is updated in response to the user's adjustment of the obstacle layout and / or explosion conditions.

[0084] If the above functions are implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, and other media that can store program codes.

[0085] It will be apparent to those skilled in the art that the present application is not limited to the details of the exemplary embodiments described above, and that the present application can be implemented in other specific forms without departing from the spirit or essential features of the present application. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the present application is defined by the appended claims rather than the above description, and it is intended that all changes falling within the meaning and scope of the equivalent elements of the claims be included in the present application. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. A hydrogen explosion simulation system for a hydrogen filling station under obstacle conditions, characterized in that: include: The data acquisition module is used to obtain the environmental conditions of the hydrogen refueling station and the geometric characteristics of the obstacles, and obtain a random number set of obstacle position feature vectors; A calculation processing module, used to establish a three-dimensional model of the hydrogen refueling station using the random number set of the obstacle position feature vector in a computational fluid dynamics model, simulate the explosion process, and output a three-dimensional calculation result data set; A support vector regression prediction module is used to use a support vector regression prediction model and the three-dimensional calculation result data set as input data to predict the pressure generated by hydrogen explosion under different obstacle layouts and obtain explosion prediction result information; The training interaction module is used to provide interactive operation functions in a virtual environment to allow users to adjust obstacle layout and / or explosion conditions, respond in real time and update explosion prediction result information.

2. The system according to claim 1, characterized in that The geometric characteristics of the obstacles include the distance of the obstacles from the explosion source, the width of the obstacles, the height of the obstacles and the number of obstacles; the environmental conditions of the hydrogen refueling station include ambient temperature, atmospheric pressure and wind speed.

3. The system according to claim 1, characterized in that The support vector regression prediction module uses a Gaussian radial basis kernel function for nonlinear fitting.

4. The system according to claim 1 or 3, characterized in that: The function coefficient of the support vector regression prediction model is set to 0.1, the width of the isolation band is set to 0.1, and the penalty factor of samples outside the isolation band is set to 100.

5. The system according to claim 1, characterized in that When establishing a three-dimensional model of a hydrogen refueling station, the calculation processing module includes: setting a calculation domain, monitoring points and flow channels, and meshing the three-dimensional model to simulate the pressure distribution after a hydrogen explosion, and obtaining a hydrogen storage tank explosion shock wave cloud map under obstacles at different locations, as well as the explosion pressure value of each monitoring point.

6. The system according to claim 1 or 5, characterized in that: When establishing the three-dimensional model of the hydrogen refueling station, the calculation processing module adopts an adaptive grid refinement algorithm to dynamically adjust the grid density according to the complexity of hydrogen leakage and diffusion.

7. The system according to claim 1, characterized in that The initial conditions and boundary conditions set by the calculation processing module when establishing the three-dimensional model of the hydrogen refueling station include environmental parameters around the explosion location and the settings of the hydrogen inlet, air inlet, and outlet.

8. The system according to claim 1, characterized in that The system also includes a result analysis module for performing in-depth analysis on the three-dimensional calculation result data set and explosion prediction result information, and providing statistical and visual display of key parameters including explosion impact range and maximum explosion pressure.

9. The system according to claim 1, characterized in that The system also includes a model accuracy testing module, which is used to test the support vector regression prediction model using a test set segmented from the experimental data set to verify the accuracy and reliability of the model.

10. The system according to claim 1, characterized in that The system also includes an optimization algorithm module for automatically searching and recommending an optimal obstacle layout plan based on explosion prediction result information to minimize the hazards that may be caused by hydrogen explosions.

Images