A numerical simulation method for the explosion and release characteristics of multi-component gases at the full-field scale

Abstract

This invention discloses a numerical simulation method for the explosion and release characteristics of multi-component gases at the full-field scale, comprising the following steps: Step S1, constructing a proportional physical model according to the shape and size of the required physical model, and performing grid structured generation, fluid domain division, and initial setting of boundary conditions; Step S2, calculating and setting the fluid domain boundaries and initial conditions, performing grid independence verification, and finalizing the grid scheme; Step S3, importing the UDF boundary condition control program, enabling the energy equation, P1 radiation model, k-ε model, and component transport model, and importing the UC San Diego reaction mechanism; Step S4, controlling the boundary conditions of the explosion pressure relief port, refining the boundary condition settings, and setting the mixed components of reactants and products; Step S5, performing data analysis using software after the simulation; This method has advantages such as comprehensive data analysis, accurate numerical results, not limited to a single gas, wide applicability, and flexibility.

Description

Technical Field

[0001] This invention belongs to the field of gas explosion simulation technology, specifically relating to a numerical simulation method for the characteristics of multi-element gas explosion and release at the full-field scale. Background Technology

[0002] During the production, storage, transportation, and use of combustible gases, explosions can occur due to human factors or other unpredictable causes, resulting in serious casualties and property damage. Explosion venting is one of the effective measures to reduce explosion risks. To explore the dynamic coupling pressure evolution process during explosion venting, various studies related to gas explosions have been conducted both domestically and internationally. However, these studies are based on setting up simulation platforms rather than numerical simulation methods, leading to high costs and low numerical accuracy related to the instantaneous venting characteristics of gas explosions. Even when numerical simulations are conducted for gas explosions, a complete system is lacking, and the initial parameter settings for simulations are not perfect, making it currently impossible to conduct complete and highly accurate numerical simulations of gas explosion venting.

[0003] Moreover, most numerical simulation studies on explosions focus on the characteristics of explosions in confined spaces, while most numerical simulation studies on the entire dynamic process of explosion release are based on a single gas and have overly simplistic reaction mechanisms. The key parameters in the simulation process of maximum controlled explosion pressure are unclear, making it impossible to more accurately grasp the simulation process of multi-gas explosion release. Summary of the Invention

[0004] This invention aims to provide a numerical simulation method for the discharge characteristics of multi-element gas explosions at the full-field scale. It features high accuracy and low cost in numerical simulation of multi-element gas explosion discharge, solving the problems of current methods that can only simulate the discharge of single gas explosions, and that the numerical simulation of gas explosion discharge is characterized by low accuracy and high cost.

[0005] Therefore, the technical solution adopted in this invention is: a numerical simulation method for the explosion and release characteristics of multi-element gases at the full-field scale, comprising the following steps:

[0006] Step S1: Based on the shape and size of the required physical model, construct a scaled physical model and perform mesh structure generation, fluid domain generation, and initial setting of boundaries and initial conditions;

[0007] Step S2: Perform fluid domain boundary and initial condition calculation settings, perform mesh independence verification, select several measurement points as mesh independence verification parameters, and increase mesh density starting from the second group of fluid domains. When the relative change rate between the calculation results under the new mesh refinement scheme and the calculation results under the previous mesh scheme is less than 5%, the mesh independence is considered to be verified, and the mesh generation scheme at this time is selected.

[0008] Step S3: Import the UDF boundary condition control program, and use the program to set the static opening pressure of the pressure relief port, open the transient environment, open the energy equation, P1 radiation model, k-ε model and component transport model, and import the UC San Diego reaction mechanism.

[0009] Step S4: Control the boundary conditions of the explosion relief port, improve the boundary condition settings, set monitoring points in the computational domain, set the mixture of reactant and product components, set the ignition location, radius, energy and duration, initialize the initial conditions of the fluid domain, set the simulation time and step size according to the requirements, and then start the simulation.

[0010] Step S5: After the simulation, use software to perform data analysis; capture the transient process of the numerical simulation, observe the propagation of flame temperature and / or explosion overpressure, extract temperature cloud maps at equal time intervals, process the flame images using a programming language, and extract the precise changes in flame propagation speed; process the data and observe the changes in explosion overpressure parameters over time.

[0011] As a preferred embodiment of the above scheme, in step S1, a proportional physical model is constructed using Gambit. This model consists of a sphere with a radius of 160mm to 621mm and a cuboid containing the sphere with dimensions of 3m to 4m in length and 1m to 1.5m in width. The sphere has a circular pressure relief port with a radius of 20mm to 60mm. The sphere uses a triangular structured mesh, while the cuboid uses a quadrilateral unstructured mesh. Pressure relief ports with different locations and sizes can be set according to different venting conditions, resulting in a reasonable design size range, wide applicability, and strong simulation flexibility. The entire fluid domain is divided using Gambit software, with the interior of the sphere designated as the inner field fluid domain and the exterior of the sphere designated as the outer field fluid domain. The boundary conditions are initially set using Gambit software, with all outer field boundaries set as free field boundaries, and the inner field boundaries, except for the pressure relief port which is set as a mesh, set as Walls.

[0012] Further preferably, in step S2, the established physical model is imported into Fluent for calculation settings. After importation, the scale is adjusted to ensure that the size of the physical model is consistent with the actual size, and the mesh quality is checked. One and four measuring points are selected inside and outside the sphere, respectively, as mesh independence test parameters. The measuring points are located inside and outside the sphere, which has high accuracy. By comparing the maximum explosion overpressure of each measuring point, the relative change rate of the calculation results of the new set of mesh schemes with the calculation results of the previous set of mesh schemes is obtained, verifying the mesh independence, thus finding the optimal solution for the number of explosion meshes, which facilitates the stability of subsequent simulations.

[0013] More preferably, in step S3, the transient environment in Fluent is opened, the energy equation and P1 radiation model are enabled, the RNG k-ε turbulence model is selected and the component transport model is opened simultaneously, a mixture is selected and its composition is modified, and then the UC San Diego reaction mechanism file is imported. The UC San Diego reaction mechanism file can be modified according to different reacting gases, which is highly flexible and can select gas mixtures of different components as needed, making it suitable for multi-component gas environments; at the same time, a finite-rate chemical reaction model is selected for calculation, which is a reasonable calculation method.

[0014] More preferably, in step S4, monitoring points are set at any location within the calculation domain to monitor the changes in instantaneous temperature, relative and absolute pressure, and molar concentration parameters of each component. The data is then grouped according to the different monitoring data, and the save location and filename of the out data file are filled in, along with the automatic saving of the data and case data files, to facilitate subsequent data processing.

[0015] More preferably, in step S4, Spark ignition is used to set the ignition position, radius, energy, and duration. Initializing the fluid domain involves setting the initial temperature and pressure parameters of the gas components within the fluid domain, thereby performing a comprehensive numerical simulation of the system and improving the realism of the simulation.

[0016] More preferably, in step S4, under the control of the UDF program, when the pressure at the pressure relief port reaches the static opening pressure, the calculation will stop, and then the boundary condition at the pressure relief port will be changed to Internal. At the same time, in the component transport model, the finite rate chemical reaction model will be changed to the finite rate / vortex dissipation model to continue the solution, so as to solve both inside and outside the sphere and obtain the explosion release values ​​inside and outside the sphere in a comprehensive and complete manner.

[0017] More preferably, in step S5, after the simulation is completed, CFD-Post, Origin, and Matlab software are used to perform data analysis on the automatically saved data, case, and out files of monitoring point records. Using CFD-Post, Origin, and Matlab code to process the calculated data allows for a more accurate analysis of the coupling relationship between the flow field, flame, and overpressure during the explosion release process, as well as various explosion release parameters. In CFD-Post, the transient process of the numerical simulation is captured to observe the propagation of flame temperature and / or explosion overpressure. Temperature cloud maps at equal time intervals are extracted, placed in the same folder, and named sequentially. Flame images are processed using Matlab programming language. The out files are then imported into Origin software to process the data calculated in Fluent, observing the changes in explosion overpressure parameters over time. This comprehensive data analysis allows for multi-faceted analysis with high accuracy.

[0018] The beneficial effects of this invention are:

[0019] (1) The UC San Diego reaction mechanism is adopted and the solution is based on the finite rate model and the finite rate / vortex dissipation model. It can comprehensively and systematically simulate the explosion and release situation inside and outside the sphere. The data analysis is carried out by software, which is comprehensive and accurate. It has great practical significance for people to formulate relevant laws and regulations, industry standards, better prevent and control the explosion of the mixed gas, assist in accident investigation and spatiotemporal inversion.

[0020] (2) Compared with setting up real experimental devices, numerical simulation is significantly cheaper, easier to operate, more flexible to use, and allows for arbitrary adjustment of gas mixture components, thereby accurately understanding the characteristics of multi-element gas explosion and release, rather than being limited to a single gas. It has a wide range of applications and high flexibility.

[0021] (3) Importing the UDF boundary condition control program is a program designed specifically for setting the static opening pressure of the pressure relief port in numerical simulation of the explosion and release characteristics of multi-component gases. The UC San Diego reaction mechanism is an integration of current gas reaction mechanisms. In order to improve the accuracy of the entire simulation and simplify unnecessary processes, the combination of energy equation, P1 radiation model, k-ε model and component transport model can fully simulate the gas explosion situation in reality.

[0022] (4) Finally, software is used for data analysis to obtain the transient process of numerical simulation, the propagation of flame temperature and / or explosion overpressure, temperature cloud maps at equal time intervals, precise changes in flame propagation speed and changes in explosion overpressure parameters over time. This allows for a comprehensive and systematic understanding of multi-element gas explosions, multi-angle and multi-directional understanding of explosion parameters, and complete and comprehensive data analysis with high accuracy.

[0023] In summary, it has advantages such as data analysis, comprehensive analysis, accurate numerical values, not limited to a single gas, wide range of applications, and flexibility. Attached Figure Description

[0024] Figure 1 This is a schematic diagram of a numerical simulation method for the characteristics of multi-element gas explosion and release at the full-field scale.

[0025] Figure 2 This is a schematic diagram of the geometric model.

[0026] Figure 3 This diagram illustrates the mesh generation and independence test of the geometric model (a. Mesh generation; b. Mesh independence test).

[0027] Figure 4 The graph shows the instantaneous overpressure curve at the measuring point.

[0028] Figure 5 This is a graph showing the instantaneous temperature at the measuring point.

[0029] Figure 6 This is a diagram showing the dynamic evolution of the flame shape.

[0030] Figure 7 This is a schematic diagram of the axial evolution of the propagation velocity of the explosion shock wave.

[0031] Figure 8 This is a schematic diagram of the axial evolution of flame propagation speed. Detailed Implementation

[0032] The present invention will be further described below with reference to the embodiments and accompanying drawings:

[0033] Combination Figure 1 — Figure 8 As shown, a numerical simulation method for the explosion and release characteristics of multi-element gases at the full-field scale is presented, and the specific implementation steps are as follows:

[0034] Step S1: Based on the shape and size of the required physical model, construct a scaled physical model and perform mesh structured generation, fluid domain generation, and initial setting of boundaries and initial conditions.

[0035] In step S1, a proportional physical model is constructed using Gambit. This model consists of a sphere with a radius of 160mm to 621mm and a cuboid containing the sphere with a length of 3m to 4m and a width of 1m to 1.5m.

[0036] The sphere has a circular pressure relief port, preferably with a radius of 20mm to 60mm.

[0037] The sphere uses a triangular structured mesh, while the cuboid uses a quadrilateral unstructured mesh.

[0038] The entire fluid domain was divided using Gambit software, with the inside of the sphere designated as the inner field fluid domain and the outside of the sphere designated as the outer field fluid domain.

[0039] The initial boundary conditions were set using Gambit software. All external field boundaries were set as free field boundaries, and the internal field boundaries, except for the pressure relief port which was set as a mesh, were set as Wall.

[0040] The files were then exported as Mesh files, and the four mesh schemes were named A, B, C, and D respectively.

[0041] Step S2: Perform fluid domain boundary condition calculation and setting, perform mesh independence verification, select several measurement points as mesh independence verification parameters, and increase mesh density starting from the second group of fluid domains. When the relative change rate between the calculation results under the new mesh refinement scheme and the calculation results under the previous mesh scheme is less than 5%, the mesh independence is considered to be verified, and the mesh generation scheme at this time is selected.

[0042] One and four measuring points are selected inside and outside the sphere, respectively, as parameters for grid independence verification. The relative rate of change between the calculation results of the new grid scheme and the calculation results of the previous grid scheme is obtained by comparing the maximum explosion overpressure of each measuring point, thus verifying grid independence.

[0043] The smaller the pressure relief port, the more complex the disturbance formed at the pressure relief port, and the higher the requirements for mesh quality. Therefore, this paper only selects the geometric model with the smallest pressure relief port for mesh independence verification.

[0044] To ensure the accuracy of the calculations, four meshing schemes were designed, and the specific parameters are shown in the table below. The table displays the maximum explosion overpressure ("P"_"max"^("P"_"n")) at each measuring point outside the pressure relief port under different meshing schemes.

[0045]

[0046] As the number of meshes in Mesh 1-4 increases, the rate of change of the maximum explosion overpressure ("P"_"max"^("P"_"n")) at each measuring point decreases. Considering both computational speed and accuracy, Mesh 3 is selected as the computational mesh. As shown in the mesh generation diagram, a total of 1,164,060 meshes are generated, with a maximum EquiSize Skew value of 0.07. Furthermore, a transition mesh with a length of 0.7m and a width of 0.5m is set at the connection between the inside and outside of the 20L sphere to improve the accuracy of the numerical simulation using physical methods and meet the computational requirements.

[0047] In step S2, the established physical model is imported into Fluent for calculation settings, which means importing four mesh schemes: A, B, C, and D. After importing, the scale is adjusted to ensure that the size of the physical model is consistent with the actual size, and the mesh quality is checked.

[0048] Step S3: Import the UDF boundary condition control program, and use the program to set the static opening pressure of the pressure relief port, open the transient environment, open the energy equation, P1 radiation model, k-ε model and component transport model, and import the UC San Diego reaction mechanism.

[0049] In step S3, open the transient environment in Fluent, then enable the energy equation and P1 radiation model. Select the RNGk-ε turbulence model and simultaneously enable the component transport model. Select a mixture and change its composition, while changing Density to ideal-gas.

[0050] Then, the UC San Diego reaction mechanism file was imported; and a finite-rate chemical reaction model was selected for calculation.

[0051] Step S4: Control the boundary conditions of the explosion relief port, improve the boundary and initial condition settings, set monitoring points in the computational domain, set the reactant product mixture components, set the ignition position, radius, energy and duration, initialize the initial conditions of the fluid domain, set the simulation time and step size according to the requirements, and then start the simulation.

[0052] In step S4, select User-Define in the Fluent panel and click Compiled UDFs to compile the UDF. Select the UDF path and import it on the Add page. Enter "Static start pressure" in User-Defined Scalars, execute the UDF, and finally make supplementary settings for the pressure relief port boundary conditions on the boundary condition settings page.

[0053] Set monitoring points at any location within the computational domain to monitor changes in instantaneous temperature, relative and absolute pressure, and molar concentration of each component. Group the monitored data according to their different characteristics, and specify the save location and filename for the OUT data file and the automatic saving of the DATA and CASE data files.

[0054] In step S4, Spark ignition is used to set the ignition position, radius, energy, and duration. Initializing the fluid domain involves setting the initial temperature and initial pressure parameters of the gas components within the fluid domain.

[0055] In step S4, under the control of the UDF program, the calculation will stop when the pressure at the pressure relief port reaches the static opening pressure. Then, the boundary conditions at the pressure relief port will be changed to Internal, and the finite rate chemical reaction model in the component transport model will be changed to a finite rate / vortex dissipation model to continue the solution.

[0056] Step S5: After the simulation, use software to perform data analysis; capture the transient process of the numerical simulation, observe the propagation of flame temperature and / or explosion overpressure, extract temperature cloud maps at equal time intervals, process the flame images using a programming language, and extract the precise changes in flame propagation speed; process the data and observe the changes in explosion overpressure parameters over time.

[0057] In step S5, after the simulation is completed, CFD-Post, Origin, and Matlab software are used to analyze the automatically saved data, case, and out files of monitoring points. In CFD-Post, the transient process of the numerical simulation is captured to observe the propagation of flame temperature and / or explosion overpressure. Temperature cloud maps with equal time intervals are extracted, placed in the same folder and named in order, and flame images are processed using the Matlab programming language.

[0058] The Matlab programming language can extract precise changes in flame propagation speed; import the .out file into Origin software, process the data calculated in Fluent, and observe the changes in explosion overpressure parameters over time.

Claims

1. A numerical simulation method for the explosion and release characteristics of multi-element gases at a full-field scale, characterized in that, Includes the following steps: Step S1: Based on the shape and size of the required physical model, construct a scaled physical model and perform mesh structure generation, fluid domain generation, and boundary and initial condition settings. In step S1, a proportional physical model is constructed using Gambit. This model consists of a sphere with a radius of 160mm to 621mm and a cuboid containing the sphere, with dimensions of 3m to 4m in length and 1m to 1.5m in width. The sphere has a circular pressure relief port with a radius of 20mm to 60mm. Gambit software is used for mesh generation. A triangular structured mesh is used inside the sphere, and a quadrilateral unstructured mesh is used in the cuboid. Gambit software is used to divide the entire fluid domain, with the inside of the sphere designated as the inner field fluid domain and the outside of the sphere designated as the outer field fluid domain. Gambit software is used to initially set the boundary conditions, setting all outer field boundaries as free field boundaries, and setting all inner field boundaries except for the pressure relief port, which is set as a mesh, as Walls. Step S2: Perform fluid domain boundary and initial condition calculation settings, perform mesh independence verification, select several measurement points as mesh independence verification parameters, and increase mesh density starting from the second group of fluid domains. When the relative change rate between the calculation results under the new mesh refinement scheme and the calculation results under the previous mesh scheme is less than 5%, the mesh independence is considered to be verified, and the mesh generation scheme at this time is selected. In step S2, the established physical model is imported into Fluent for calculation settings. After importing, the scale is adjusted to ensure that the size of the physical model is consistent with the actual size, and the mesh quality is checked. One and four measuring points are selected inside and outside the sphere, respectively, as parameters for grid independence verification. The relative rate of change between the calculation results of the new grid scheme and the calculation results of the previous grid scheme is obtained by comparing the maximum explosion overpressure of each measuring point, thus verifying grid independence. Step S3: Import the UDF boundary condition control program, and use the program to set the static opening pressure of the pressure relief port, open the transient environment, open the energy equation, P1 radiation model, k-ε model and component transport model, and import the UC San Diego reaction mechanism. Step S4: Control the boundary conditions of the explosion relief port, improve the boundary condition settings, set monitoring points in the computational domain, set the mixture of reactant and product components, set the ignition location, radius, energy and duration, initialize the initial conditions of the fluid domain, set the simulation time and step size according to the requirements, and then start the simulation. In step S4, monitoring points are set at any location within the calculation domain to monitor the changes in instantaneous temperature, relative pressure, absolute pressure, and molar concentration parameters of each component. The data is then grouped according to the different monitoring data. The save location and filename of the out data file are filled in, and the automatic saving of the data and case data files is set. In step S4, under the control of the UDF program, when the pressure at the pressure relief port reaches the static opening pressure, the calculation will stop, and then the boundary conditions at the pressure relief port will be changed to Internal. At the same time, in the component transport model, the finite rate chemical reaction model will be changed to the finite rate / vortex dissipation model to continue the solution. Step S5: After the simulation, use software to perform data analysis; capture the transient process of the numerical simulation, observe the propagation of flame temperature and / or explosion overpressure, extract temperature cloud maps at equal time intervals, process the flame images using a programming language, and extract the precise changes in flame propagation speed; process the data and observe the changes in explosion overpressure parameters over time.

2. The numerical simulation method for the explosion and release characteristics of multi-element gas at the full-field scale according to claim 1, characterized in that: In step S3, the transient environment in Fluent is opened, the energy equation and P1 radiation model are enabled, the RNG k-ε turbulence model is selected and the component transport model is enabled, a mixture is selected and its composition is modified, and then the UC San Diego reaction mechanism file is imported; at the same time, the finite rate chemical reaction model is selected for calculation.

3. The numerical simulation method for the explosion and release characteristics of multi-element gas at the full-field scale according to claim 1, characterized in that: In step S4, Spark ignition is used to set the ignition position, radius, energy, and duration. Initializing the fluid domain involves setting the initial temperature and initial pressure parameters of the gas components within the fluid domain.

4. The numerical simulation method for the explosion and release characteristics of multi-element gas at the full-field scale according to claim 1, characterized in that: In step S5, after the simulation is completed, CFD-Post, Origin, and Matlab software are used to analyze the automatically saved data, case, and out files of monitoring point records. In CFD-Post, the transient process of the numerical simulation is captured to observe the propagation of flame temperature and / or explosion overpressure. Temperature cloud maps with equal time intervals are extracted, placed in the same folder and named in order. Flame images are processed using the Matlab programming language. The out files are imported into Origin software to process the data calculated in Fluent and observe the changes of explosion overpressure parameters over time.

Images