An aircraft fuel tank inerting process simulation method based on gas-liquid-solid smooth particles

Abstract

The application discloses a kind of based on gas-liquid solid smooth particle aircraft fuel tank inerting process simulation method, comprising the following steps: discrete loading to external fuel tank model and boundary condition configuration;Initialization spatial distribution model, define in-out buffer layer;Dynamic creation / removal buffer layer particle, construct adjacent particle linked list structure;Calculate the coupling effect of multiple phases and update particle physical properties and position;Characteristic extraction is carried out to particle information, generate current time data file;Calculate next time data, until simulation termination;Fuel state and inerting dynamic process are visualized to show each time data;Analysis and evaluation are carried out to fuel state and inerting performance.The application has the advantages of strong universality, high calculation efficiency, reliable simulation accuracy, and can meet the needs of aircraft fuel tank inerting process simulation under complex fuel tank structure and working condition.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of numerical analysis of aircraft fuel, and particularly relates to a simulation method for an aircraft fuel tank inerting process based on gas-liquid-solid smooth particles. BACKGROUND

[0002] The aircraft fuel tank inerting technology is a technology for reducing the volume fraction of space oxygen to within a safe range by filling inert gas into the gas phase space of the fuel tank, so as to reduce the probability of tank fire and explosion caused by the comprehensive thermal management method of fuel, and to keep the aircraft in a stable state. With the increase of the flight speed and performance of the fighter, the aerodynamic heating and electronic equipment thermal load are also increasing, which puts forward higher requirements for the design of the inerting system.

[0003] At present, researchers at home and abroad mainly use the finite volume method based on the Euler method to carry out computational fluid dynamics numerical simulation in the initial stage of the inerting system design. For example, the commonly used commercial software ANSYS Fluent, CFX, etc. The algorithm integrates and solves the Euler control equation of the fluid in the unit control body, constructs a complex mathematical model by using physical information such as gas pressure, compartment pressure difference, communication relationship, gas inlet flow and exhaust hole position, and iteratively calculates the nitrogen inerting process.

[0004] The traditional inerting process simulation method is not applicable when facing complex boundary conditions and multiphase coupling relationship. The numerical model based on the finite volume method depends on the grid division method and mass, and lacks consideration of the oil tank compartment communication mode, the layout of the oil tank accessories and different oil filling ratios. The mathematical model needs to be reset for different oil tank structures, and it is difficult to balance the calculation efficiency and accuracy at the same time. In addition, due to the discontinuity and randomness of the fluid distribution inside the oil tank, the control volume unit cannot reflect the large deformation of the fluid surface and the multiphase coupling effect of gas-liquid-solid, which causes a large error in result verification and visualization analysis. SUMMARY

[0005] In order to overcome the shortcomings of the above simulation model, the application provides a simulation method for an aircraft fuel tank inerting process based on gas-liquid-solid smooth particles. Based on the smooth particle dynamics theory without grid, a gas-liquid-solid multiphase coupling calculation framework is proposed, a dynamic change model of the aircraft fuel tank inerting process is established, the numerical calculation accuracy and efficiency are improved, and the whole process visualization simulation is realized.

[0006] To achieve the above purpose, the application adopts the following technical solutions:

[0007] A simulation method for an aircraft fuel tank inerting process based on gas-liquid-solid smooth particles, comprising the following steps:

[0008] Step 1) Discrete loading of the external oil tank model and configuration of the boundary conditions;

[0009] Step 2) initialize the spatial distribution model, define the buffer layer in and out;

[0010] Step 3) dynamically create / remove buffer layer particles, build adjacent particle linked list structure;

[0011] Step 4) calculate the multi-phase coupling effect and update the physical properties and position of the particles;

[0012] Step 5) feature extraction is performed on the particle information to generate the current time data file;

[0013] Step 6) repeat steps 3) to 5), calculate the next time data until the simulation is terminated;

[0014] Step 7) visualize the fuel state and inerting dynamic process at each time;

[0015] Step 8) analyze and evaluate the fuel state and inerting performance.

[0016] Further improvement of the application is that the step 1) is the method for discrete loading of external tank model and boundary condition configuration:

[0017] Step 1.1), pre-processing the standard tank object generated by CATIA or CAD three-dimensional design program, splitting and extracting triangular facet geometric features, and simplifying the tank model information to balance precision and loading efficiency;

[0018] Step 1.2), using the contour shape and topological structure of the external tank object, matching the discrete node position in the Cartesian grid with a set resolution, and dividing the fluid feasibility area;

[0019] Step 1.3), changing the distribution position of the tank object by using translation, scaling or rotation geometric transformation of the discrete node;

[0020] Step 1.4), configuring the model boundary conditions, defining the simulation space range and assigning key attributes of solid phase particles, including reference density, artificial viscosity and Young's modulus, and creating solid phase boundary particles.

[0021] Further improvement of the application is that the step 2) is the method for initializing the spatial distribution model and defining the buffer layer in and out:

[0022] Step 2.1), using XML extensible markup language to store gas-liquid phase fluid properties and parameters for system operation;

[0023] Step 2.2), setting the initial oil filling ratio and oil surface angle, defining the fuel distribution space, and creating different types of gas-liquid phase static particles;

[0024] Step 2.3), defining the open boundary buffer layer at the oil tank internal oil pipe orifice and the inert gas inlet and outlet position, binding the gas-liquid phase fluid properties, and converting the fuel consumption and inert gas flow rate working indicators of the inerting system into the flow rate data of the open boundary buffer layer;

[0025] Step 2.4), outputting the initial state information of the fluid, and completing the initialization of the multiphase coupling model.

[0026] Further improvement of the present application is that the step 3) is the method for dynamically creating / removing the buffer layer particles and constructing the adjacent particle linked list structure, which is as follows:

[0027] Step 3.1), calculating the number of particles entering and leaving each buffer layer at the current time according to the configuration information and flow rate data of the open boundary buffer layer;

[0028] Step 3.2), creating new particles at the buffer layer inlet area and assigning corresponding fluid properties, and removing the outflow particles at the buffer layer outlet area;

[0029] Step 3.3), dividing the simulation area into a large number of grids by the smoothing kernel length and sequentially numbering, and establishing the linked list structure according to the grid position of the fluid particle.

[0030] Further improvement of the present application is that the step 4) is the method for calculating the multiphase coupling effect and updating the particle physical properties and position, which is as follows:

[0031] Step 4.1), locating the adjacent grid of the particle by the adjacent particle linked list, and searching for all particle serial numbers within the influence domain of the particle in the adjacent grid;

[0032] Step 4.2), executing the operation by the CUDA parallel computing architecture, and solving the multiphase coupling state equation and control equation by the smoothing kernel function;

[0033] Step 4.3), matching the maximum velocity and acceleration of the particles in the fluid feasible region with the adaptive time step, and balancing the numerical accuracy and iteration efficiency;

[0034] Step 4.4), preventing the non-natural cavity of the particle from being generated by correcting the fluid surface displacement, and maintaining the numerical stability;

[0035] Step 4.5), iteratively updating the physical properties and position of the particle by the Verlet time step integration, and the physical properties include the density, pressure, acceleration and velocity;

[0036] Step 4.6), judging whether the particle position is within the range of the fluid feasible region, and removing the particles that escape from the calculation.

[0037] A further improvement of this invention is that, in step 5), the method for extracting features from particle data and generating the current time-time data file is as follows:

[0038] Step 5.1) Retrieve the dataset of the computation results from the CUDA device and classify different particle types based on feature values;

[0039] Step 5.2): Use OpenMP to design a multi-threaded model to extract fluid particle properties and position information, and create necessary components.

[0040] Step 5.3) Use a binary format container to generate a data file for the current moment and save it to an external storage device for quick reading without loss of precision.

[0041] A further improvement of this invention is that, in step 7), the method for visualizing the fuel state and inertization dynamic process at various time points is as follows:

[0042] Step 7.1): Convert the data files at each time point into VTK format files for particle visualization analysis and store them in a folder in the form of a time series.

[0043] Step 7.2) Use the spatial distribution of fluid particles to fit and generate a three-dimensional data isosurface to realize the drawing of the free surface of the fluid, and store it as a VTK format file in the form of a time series.

[0044] Step 7.3) Using the time series correspondence, import the VTK file of fluid particles or free surface into the 3D interactive scene and load the particle number, velocity, acceleration, density and pressure particle data of the corresponding frame.

[0045] Step 7.4) Adjust the rendering perspective by scaling, moving and rotating the graphics, use the Cycles rendering engine to edit and configure the shaders, update the rendering objects according to the progress of the simulation steps and output the animation at the set frame rate to obtain image and video files that demonstrate the simulation process.

[0046] A further improvement of the present invention is that, in step 8), the method for analyzing and evaluating the fuel condition and inerting performance is as follows:

[0047] Step 8.1) Calculate the fuel tank volume and remaining fuel amount at each time point using fuel particle data, generate multi-compartment fuel consumption curves for the inerting process, and analyze fuel flow performance.

[0048] Step 8.2) Calculate the center of gravity position of the multi-compartment fuel tank at each time point, generate the center of gravity distribution change curve during the inerting process, and evaluate the impact of fuel sloshing on the aircraft handling performance during the inerting process.

[0049] Step 8.3) Calculate the oxygen content in the tank at that moment using inert gas particle data, generate the average oxygen volume fraction curve of the inerting process, and determine the flow direction and volume flow distribution between compartments by particle motion.

[0050] Step 8.4): Based on the safe oxygen volume fraction range standard, evaluate the spatial distribution of oxygen volume fraction in the fuel tank and the variation law of average oxygen volume fraction over time, identify the oxygen mass accumulation area in the fuel tank, and optimize the design of the inerting system to make it suitable for complex fuel tank structures and gas flow patterns, thereby improving the fire prevention and explosion suppression capability of the fuel tank.

[0051] Compared with the prior art, the present invention has at least the following beneficial technical effects:

[0052] This invention provides a simulation method for the inerting process of aircraft fuel tanks based on smooth particles in a gas-liquid-solid system. Based on the smooth particle dynamics theory of meshless numerical computation, the simulation region within the fuel tank is divided into a massive number of interacting particle combinations, avoiding the meshing operations of traditional Euler numerical computation methods and improving the accuracy of solid boundary and fluid surface deformation handling. Simultaneously, by establishing a neighboring particle chain list structure to search the influence domain of each particle, and using the CUDA parallel computing architecture to solve the state equations and control equations, efficient and accurate calculation of inter-particle interactions is achieved.

[0053] The computational framework for gas-liquid-solid multiphase coupling proposed in this invention incorporates the background pressure and cohesive force terms of the gas into a single fluid state equation, solving the typical inert gas-fuel coupling problem in the field of fuel tank explosion suppression system research. It achieves inert gas inflow / outflow and fuel height variation through an open boundary buffer layer method; prevents unnatural voids in particles and maintains numerical stability through phase interface dynamics calculations and free surface displacement correction; and provides a reliable basis for evaluating the performance of the inerting system through simulation visualization methods.

[0054] This invention has the advantages of strong versatility, high computational efficiency, and reliable simulation accuracy. It can meet the simulation needs of aircraft fuel tank inerting process under complex fuel tank structure and operating conditions. Its method also has guiding significance for multiphase coupling simulation problems in the field of computational fluid dynamics. Attached Figure Description

[0055] Figure 1 This is a schematic diagram of the process of the present invention.

[0056] Figure 2 This is a schematic diagram illustrating the dynamic inerting process of the present invention.

[0057] Figure 3 This is a graph showing the average oxygen volume fraction in the fuel tank according to the present invention. Detailed Implementation

[0058] The present invention will be further described below with reference to the accompanying drawings and embodiments. It should be noted that the described embodiments are merely preferred embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0059] like Figure 1 As shown, the present invention provides a simulation method for the inerting process of aircraft fuel tanks based on gas-liquid-solid smooth particles, comprising the following steps:

[0060] 1. Discretize and load the external fuel tank model and configure boundary conditions.

[0061] 1.1) Preprocess the externally loaded standard STL fuel tank model, decompose and extract the geometric features of the triangular facets, and simplify the information content of the fuel tank model;

[0062] 1.2) Using the contour shape and topology of the aircraft fuel tank object, the discrete node positions are matched in a Cartesian grid with a set resolution (node ​​spacing of 0.01m) to divide the fluid feasibility region.

[0063] 1.3) The discrete nodes are transformed using geometric transformations such as translation, scaling, or rotation to change the distribution position of the tank object;

[0064] 1.4) Configure the model boundary conditions, define the simulation space as 5.0m × 5.0m × 5.0m, and assign the key constant parameter for solid particles—reference density—to 2820 kg / m³. 3 The oil tank boundary particles are created using properties such as artificial viscosity of 0.01 and Young's modulus of 200 GPa.

[0065] 2. Initialize the spatial distribution model and define the ingress and egress buffer layers.

[0066] 2.1) The gas-liquid phase fluid properties and system execution parameters are stored in XML Extensible Markup Language. The gas-liquid phase fluid properties are shown in Table 1. The system execution parameters are as follows: smooth kernel length coefficient is 0.9, CFL convergence condition judgment number is 0.2, maximum simulation time is 20 min, intermediate process file output interval time is 0.03 s, and oxygen volume fraction target is 12%.

[0067] Table 1. Gas-Liquid Phase Fluid Properties

[0068]

[0069]

[0070] 2.2), Set the initial fuel filling ratio to 50% and the pitch angle to 6°, solve for the fuel level height, divide the fuel and gas spaces, and create different types of air-fuel static particles;

[0071] 2.3), Define an open boundary buffer layer at the positions of the internal fuel pipe outlet and the inert gas inlet and outlet in the fuel tank, bind the gas-liquid phase fluid properties, as shown in Table 2. Convert the working indicators such as the fuel consumption and inert gas flow rate of the inerting system into the flow velocity data of the open boundary buffer layer;

[0072] Table 2 Open boundary buffer layer table

[0073]

[0074] 2.4) Output the spatial distribution model and fluid initial state information (system configuration parameters, open boundary buffer layer information, the number of fluid particles in each phase), and the initialization of the multiphase coupling model is completed.

[0075] 3. Dynamically create / remove buffer layer particles and construct an adjacent particle linked list structure.

[0076] 3.1), Calculate the number of inlet / outlet particles in each buffer layer at the current moment according to the configuration information and flow velocity data of the open boundary buffer layer;

[0077] 3.2), Create new particles in the buffer layer inlet area and assign corresponding fluid properties, and remove the outflow particles in the buffer layer outlet area;

[0078] 3.3), Divide the simulation area into a large number of grids by the smoothing kernel length and number them in sequence, and establish a linked list structure according to the grid positions where the fluid particles are located;

[0079] 4. Calculate the multiphase coupling effect and update the physical properties and positions of the particles.

[0080] 4.1), Use the adjacent particle linked list to locate the adjacent grids of the grid where the particle is located, and search for the serial numbers of all particles within the influence domain of the particle in the adjacent grids;

[0081] 4.2), Perform operations on all particles respectively through the CUDA parallel computing architecture, set the BlockSize to 256, and solve the multiphase coupling state equation and control equation using the smoothing kernel function;

[0082] 4.3), Match the maximum velocity and acceleration of the particles in the fluid feasibility region with the adaptive time step (satisfying t < h / Vmax) to balance the numerical accuracy and iteration efficiency;

[0083] 4.4), Prevent the calculation from generating unnatural voids in the particles by correcting the fluid surface displacement and maintain numerical stability;

[0084] 4.5) The physical properties (density, pressure, acceleration and velocity) and position of the particles are iteratively updated using Verlet time step integration;

[0085] 4.6) Determine whether the particle position is within the fluid feasibility zone and remove any particles that have escaped from the calculation.

[0086] 5) Extract features from particle information to generate a data file for the current moment.

[0087] 5.1) The dataset of the calculation results is retrieved from the NVIDIA GeForce RTX 3090 device and the nitrogen particles, air particles and fuel particles are divided by feature value.

[0088] 5.2) Using OpenMP, a multi-threaded model is designed to extract fluid particle properties and position information, and density, position, velocity, acceleration, and pressure components are created.

[0089] 5.3) Use a binary format container to generate a data file for the current moment and save it to a disk device for fast reading without loss of precision.

[0090] 6) At the current moment, a termination judgment is made. If the iteration time reaches the maximum simulation time, proceed to step 7); otherwise, repeat steps 3) to 5) to perform data calculation for the next moment.

[0091] 7) Visualize the fuel status and inertization dynamic process at each time step. Figure 2 ).

[0092] 7.1) Convert the data files at each time point into VTK format files for particle visualization analysis and store them in a folder as a time series.

[0093] 7.2) The three-dimensional data isosurface is generated by fitting the spatial distribution of fluid particles in the intermediate process file to realize the drawing of the free surface of the fluid, and is also stored as a VTK format file in the form of time series.

[0094] 7.3) Using the time series correspondence, import the air particle file, nitrogen particle file and fuel free surface file into the 3D interactive scene, and load the particle data such as particle number, velocity, acceleration, density and pressure of the corresponding frame;

[0095] 7.4) Adjust the rendering perspective through graphical operations such as scaling, moving, and rotating, use the Cycles rendering engine to edit and configure shaders, update the rendering objects according to the progress of the simulation steps, and output the animation at a frame rate of 24fps to obtain image and video files that demonstrate the simulation process.

[0096] 8) Analyze and evaluate the fuel condition and inerting performance.

[0097] 8.1) Calculate the fuel tank volume and remaining fuel amount at each time point using fuel particle data, generate multi-compartment fuel consumption curves for the inerting process, and analyze fuel flow performance.

[0098] 8.2) Calculate the center of gravity position of the multi-compartment fuel tank at each time point, generate the center of gravity distribution change curve during the inerting process, and evaluate the impact of fuel sloshing on the aircraft handling performance during the inerting process.

[0099] 8.3) Calculate the oxygen content in the tank at that moment using inert gas particle data, generate the average oxygen volume fraction curve of the inerting process, and determine the flow direction and volume flow distribution between compartments by particle motion.

[0100] 8.4) Based on the safe oxygen volume fraction range standard, evaluate the spatial distribution of oxygen volume fraction in the fuel tank and the variation of the average oxygen volume fraction over time. Figure 3 The goal is to identify areas where oxygen accumulates within the fuel tank and optimize the inerting system to adapt it to complex tank structures and gas flow patterns, thereby improving the fuel tank's fire and explosion suppression capabilities.

[0101] The above description of the embodiments is only for the purpose of helping to understand the core idea of ​​the present invention; at the same time, for those skilled in the art, there will be changes in the specific implementation and application scope based on the idea of ​​the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A simulation method for the inerting process of aircraft fuel tanks based on gas-liquid-solid smooth particles, characterized in that, Includes the following steps: Step 1) Discretize and load the external fuel tank model and configure boundary conditions, including: Step 1.1) Preprocess the standard fuel tank object generated by CATIA or CAD 3D design program, split and extract the geometric features of triangular facets, and simplify the fuel tank model information to balance accuracy and loading efficiency. Step 1.2) Using the outline shape and topology of the external tank object, the discrete node positions are matched in a Cartesian grid with a set resolution to divide the fluid feasibility region. Step 1.3) Change the distribution position of the fuel tank object by using translation, scaling or rotation geometric transformations on the discrete nodes; Step 1.4) Configure model boundary conditions, define the simulation space range and assign key properties to solid particles, including reference density, artificial viscosity and Young's modulus, and create solid boundary particles; Step 2) Initialize the spatial distribution model and define the inlet and outlet buffer layers, including: Step 2.1) Utilize XML (Extensible Markup Language) to store gas-liquid phase fluid properties and parameters for system operations; Step 2.2) Set the initial fuel fill ratio and fuel level angle, define the fuel distribution space, and create different types of gas-liquid phase fluid static particles; Step 2.3) Define an open boundary buffer layer at the oil inlet and inlet / outlet of the inert gas inside the fuel tank, bind the gas-liquid phase fluid properties, and convert the fuel consumption and inert gas flow rate of the inert system into the flow rate data of the open boundary buffer layer. Step 3) Dynamically create / remove buffer layer particles and construct a neighboring particle list structure, including: Step 3.1) Calculate the number of particles entering / exiting each buffer layer at the current moment based on the configuration information and flow rate data of the open boundary buffer layer; Step 3.2): Create new particles in the inlet region of the buffer layer and assign them corresponding fluid properties; remove outflowing particles in the outlet region of the buffer layer. Step 3.3) Divide the simulation region into a large number of grids according to the smoothing kernel length and number them sequentially. Establish a linked list structure based on the grid position of the fluid particles. Step 4) Calculate multiphase coupling effects and update particle physical properties and positions, including: Step 4.1) Use the neighboring particle list to locate the adjacent grids of the grid where the particle is located, and search for all particle numbers within the influence domain of the particle in the adjacent grids; Step 4.2) Execute the computation using the CUDA parallel computing architecture and solve the multiphase coupled state equations and control equations using a smooth kernel function; Step 4.3) Match the maximum particle velocity and acceleration within the fluid feasibility region with an adaptive time step to balance numerical accuracy and iteration efficiency; Step 4.4) prevents the calculation from generating unnatural voids in particles and maintains numerical stability by correcting for fluid surface displacement; Step 4.5) Iteratively update the physical properties and position of the particle using Verlet time step integration. The physical properties include density, pressure, acceleration, and velocity. Step 4.6) Determine whether the particle position is within the fluid feasibility region and remove the calculated escaped particles; Step 5) Extract features from the particle information to generate the current time-time data file; Step 6) Repeat steps 3) through 5) to calculate the data for the next time step, until the simulation terminates; Step 7) Visualize the data at each time point to show the fuel status and inertization dynamic process; Step 8) Analyze and evaluate the fuel condition and inerting performance.

2. The simulation method for aircraft fuel tank inerting process based on gas-liquid-solid smooth particles according to claim 1, characterized in that, Step 5), which involves feature extraction from particle data to generate the current time-time data file, is as follows: Step 5.1) Retrieve the dataset of the computation results from the CUDA device and classify different particle types based on the feature values; Step 5.2) Utilize OpenMP to design a multi-threaded model to extract fluid particle properties and position information, and create necessary components; Step 5.3) Use a binary format container to generate a data file for the current moment and save it to an external storage device for fast reading without loss of precision.

3. The simulation method for aircraft fuel tank inerting process based on gas-liquid-solid smooth particles according to claim 2, characterized in that, Step 7) involves visualizing the fuel status and inertization dynamics at various time points as follows: Step 7.1) Convert the data files at each time point into VTK format files for particle visualization analysis and store them in a folder in the form of a time series; Step 7.2) Use the spatial distribution of fluid particles to fit and generate a three-dimensional data isosurface to realize the drawing of the free surface of the fluid, and store it as a VTK format file in the form of a time series. Step 7.3) Using the time series correspondence, import the VTK file of fluid particles or free surface into the 3D interactive scene and load the particle number, velocity, acceleration, density and pressure particle data of the corresponding frame; Step 7.4) Adjust the rendering perspective by scaling, moving and rotating the graphics, edit and configure the shaders, update the rendering objects according to the progress of the simulation steps and output the animation at the set frame rate to obtain image and video files that show the simulation process.

4. The simulation method for aircraft fuel tank inerting process based on gas-liquid-solid smooth particles according to claim 3, characterized in that, Step 7.4) Use the Cycles rendering engine to edit and configure the shaders.

5. The simulation method for aircraft fuel tank inerting process based on gas-liquid-solid smooth particles according to claim 3, characterized in that, The method for analyzing and evaluating the fuel condition and inerting performance in step 8) is as follows: Step 8.1) Calculate the fuel tank volume and remaining fuel amount at each time point using fuel particle data, generate multi-compartment fuel consumption curves for the inerting process, and analyze fuel flow performance. Step 8.2) Calculate the center of gravity position of the multi-compartment fuel tank at each time point, generate the center of gravity distribution change curve during the inerting process, and evaluate the impact of fuel sloshing on the aircraft handling performance during the inerting process. Step 8.3) Calculate the oxygen content in the tank at that moment using inert gas particle data, generate the average oxygen volume fraction curve of the inerting process, and determine the flow direction and volume flow distribution between compartments by particle motion. Step 8.4) Based on the safe oxygen volume fraction range standard, evaluate the spatial distribution of oxygen volume fraction in the fuel tank and the variation law of average oxygen volume fraction over time, identify the oxygen mass accumulation area in the fuel tank, and optimize the design of the inerting system to make it suitable for complex fuel tank structures and gas flow patterns, thereby improving the fire prevention and explosion suppression capability of the fuel tank.

Images