Take-off point rear side operation safety interval calculation method and system under engine jet flow influence

Abstract

The invention discloses a takeoff point rear side operation safety interval calculation method and system under the influence of engine jet flow, and the method combines laser wind finding radar measurement experiment, data assimilation, computational fluid dynamics (CFD) numerical simulation and other methods, and measures the engine jet flow data of a takeoff aircraft at an airport runway end through a laser wind finding radar; constructing a double-engine nozzle model by using a computational fluid mechanics method, correcting a numerical simulation model by using a data assimilation method in combination with high-credibility laser radar wind measurement experimental data, and calculating an influence distance of typical aircraft engine jet flow; a model of the airplane under the influence of engine jet flow is established by using a CFD method, the model and the model are combined to calculate the minimum safety interval of rear side crossing, and reliable data reference is provided for a multi-runway airport to use a take-off point rear side runway crossing mode. The accurate safety interval between the front and rear airports is defined, and the overall operation efficiency of airports in China is remarkably improved.

Description

Technical Field

[0001] The present invention relates to the field of civil aviation safety technology, in particular to a method and system for calculating a safety interval for rear-side operation of a take-off point under the influence of an engine jet. Background Art

[0002] At present, the runway crossing methods of domestic multi-runway airports are mainly the front side crossing and the circumnavigation method of the take-off point, which have certain runway incursion risks and low economic benefits. The runway crossing method of the rear side of the take-off point can not only avoid safety issues such as runway incursion, but also improve the efficiency of field operations. The key to the implementation of the runway crossing method of the rear side of the take-off point is the safe interval between the departing aircraft and the crossing aircraft. The jet impact distance of the leading aircraft and the crosswind resistance of the trailing aircraft are the decisive factors for determining the minimum safe interval between the leading and trailing aircraft in the rear side operation method of the take-off point. For the jet impact distance of the leading aircraft, the computational fluid dynamics (CFD) method is usually used for simulation, the engine structure grid is drawn using ICEM and other software, and the Fluent software is used for calculation to obtain the influence range of the jet of the studied engine. For the crosswind resistance of the trailing aircraft, the numerical simulation results of the effect of the jet of the leading aircraft on the trailing aircraft at a certain distance are combined with the factors affecting the taxiing stability of the trailing aircraft and the anti-side force model to make a judgment, and the taxiing stability and crosswind resistance of the trailing aircraft when crossing the runway are analyzed.

[0003] In previous studies, due to limitations in computing resources, equipment, and other aspects, the acquisition of large-scale, high-precision aircraft jet impact distance data was accompanied by high costs and difficulties. In order to solve the problem of high-precision takeoff aircraft jet impact distance data being difficult to obtain and high cost, computational fluid dynamics has been well developed with the advancement of computer technology. Compared with experimental fluid dynamics, it has the advantages of low cost and easy implementation. CFD simulation requires a lot of computing resources, especially for high-precision and complex simulation scenes, which require powerful computer hardware and long computing time. In order to make the calculation feasible, CFD simulation usually requires simplifications and assumptions about the physical process. These simplifications may cause the results to deviate from the actual situation, especially in the case of complex flows (such as turbulence) or multiphase flows. The accuracy of the simulation results is highly dependent on the quality of the grid. A grid that is too coarse may result in insufficient accuracy, while a grid that is too fine will increase the computing cost and time.

[0004] At the same time, when predecessors established the front aircraft model, they only selected the single-engine nozzle model for research, and there would be certain errors in the relevant calculation results; since model establishment and numerical simulation take a lot of time and cannot be comprehensive, only two groups of front and rear aircraft combinations were studied for the taxiing stability of the rear aircraft. Summary of the invention

[0005] In view of this, the purpose of the present invention is to provide a method for calculating the safe distance of the aircraft engine jet to the rear side of the take-off point under the influence of the engine jet. The method constructs a twin-engine nozzle model and uses a data assimilation method to calculate the safe distance of the aircraft engine jet to the rear side of the take-off point.

[0006] In order to achieve the above object, the present invention provides the following technical solutions:

[0007] The method for calculating the safety interval of the rear side of the take-off point under the influence of the engine jet provided by the present invention comprises the following steps:

[0008] S1: Obtain the aircraft's engine jet data;

[0009] S2: Use computational fluid dynamics to build a twin-engine nozzle model; use data assimilation methods and engine jet data to perform numerical simulations of the twin-engine nozzle model;

[0010] S3: Use CFD method to establish a model of the aircraft subjected to crosswind under the influence of engine jet;

[0011] S4: Calculate the minimum safe interval for rear crossing.

[0012] Furthermore, the engine jet data in step S1 is obtained by measuring the take-off aircraft at the end of the airport runway using a laser wind radar.

[0013] Further, the twin-engine nozzle model in step S2 is constructed according to the following steps:

[0014] S21: determine the engine model and its parameters, and construct a three-dimensional geometric model of the engine;

[0015] S22: determining a fluid calculation domain, wherein the fluid calculation domain includes the inner and outer ducts of the nozzle and the external flow calculation domain;

[0016] S23: Determine the engine surface grid and the internal and external duct grid layout;

[0017] S24: Calculate the boundary conditions of the inner and outer ducts of the engine at takeoff thrust; obtain the total temperature and total pressure of the inner and outer ducts of the engine at full thrust, reduced thrust, and reduced thrust working states;

[0018] S25: Set the parameters of the internal and external duct flow monitoring points, initialization and iteration steps. After all settings are completed, perform the solution calculation.

[0019] Further, the processing by using data assimilation in step S2 adopts any one of the following two schemes:

[0020] Option 1 assimilates the full-field turbulent eddy viscosity data predicted by the turbulence model; or

[0021] Scheme 2 assimilates the empirical parameters of the existing turbulence model;

[0022] For the above scheme 1, the full-field turbulent eddy viscosity data obtained by assimilation and update are entered into the CFD solution iteration by using the frozen one-way coupling method;

[0023] For the above-mentioned scheme 2, the empirical model parameters obtained by assimilation are substituted back into the turbulence model in a two-way coupling manner to participate in the subsequent CFD solution iteration;

[0024] The turbulence field data obtained through assimilation will be used as samples for studying the laws of turbulence fields and turbulence models.

[0025] Furthermore, the specific steps of coupling data assimilation with CFD solution in the data assimilation processing in step S2 are as follows:

[0026] T = k assimilation step analysis field;

[0027] T = k assimilation step state quantity X;

[0028] CFD coupling calculation (one-way and two-way coupling);

[0029] T = k + 1 assimilation step prediction field.

[0030] Furthermore, the data assimilation process in step S2 is performed according to the following steps:

[0031] S211: Setting assimilation parameters, including task name and type; setting of assimilated observation variables; setting of assimilation algorithm; setting of assimilation steps, etc.;

[0032] S212: Combine the computational grid file, experimental data file / high-precision numerical simulation results; interpolate and link the experimental data to the computational grid space; and output the interpolation results;

[0033] S221: Generate a collection sample by combining the disturbance parameter type, parameter disturbance method, and the number of collection samples.

[0034] S222: Then the external solver calculates and initializes the convergent flow field;

[0035] S231: Read the sample flow field;

[0036] S232: construct data assimilation state vector;

[0037] S233: construct observation vectors according to computational grid files, experimental data files / high-precision numerical simulation results, and perform data assimilation algorithm update steps; update flow field variables; external solver calculations;

[0038] S234: Calculate the converged flow field;

[0039] S241: flow field visualization output;

[0040] S242: Sample statistical information analysis;

[0041] S243: Combined with the error convergence condition, the maximum number of assimilation rounds t * >tmax, determine whether the end condition is met, if not, return to read the sample flow field;

[0042] S244: If yes, the task ends.

[0043] Further, the step S3 uses the CFD method to establish a model of the aircraft subjected to the crosswind under the influence of the engine jet, and is performed according to the following steps:

[0044] S31: Draw the aerodynamic shape of the aircraft under study based on the aircraft shape data in the aircraft performance manual;

[0045] S32: Establish aircraft unstructured grid;

[0046] S33: Calculate the model and post-process the results using Tecplot software;

[0047] Furthermore, the step S4 calculates the minimum safety interval for rear-side crossing, and the specific steps are as follows:

[0048] S41: First, through numerical simulation, the nozzle velocity distribution, including the entire velocity field, is obtained, and then the crosswind boundary conditions of a certain aircraft model crossing the runway are obtained;

[0049] S42: Simulate the crosswind boundary with different incoming crosswind wind speeds, perform numerical simulation on the crosswind incoming flow of typical aircraft models, calculate the approximate resistance and moment coefficients, and calculate the critical wind speed after the rear tire reaches the maximum static friction through the force coefficient calculation;

[0050] S43: Based on the critical wind speed and the wake speed distribution, the safety distance is obtained by finding the position corresponding to the critical wind speed in the wake flow field.

[0051] The present invention provides a system for calculating the safety interval of the rear side of the take-off point under the influence of the engine jet, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor. The method is implemented when the processor executes the program.

[0052] The beneficial effects of the present invention are:

[0053] The present invention provides a method for calculating the safe interval of the rear side of the take-off point under the influence of the engine jet. The method combines laser wind radar measurement experiments, data assimilation, computational fluid dynamics (CFD) numerical simulation and other methods. The laser wind radar is used to measure the engine jet data of the take-off aircraft at the end of the airport runway; the computational fluid dynamics method is used to construct a twin-engine nozzle model, and the data assimilation method is used in combination with high-reliability laser radar wind measurement experimental data to correct the numerical simulation model and calculate the influence distance of the typical aircraft engine jet; the CFD method is used to establish a model of the aircraft under the influence of the engine jet affected by the crosswind, and the two are combined to calculate the minimum safe interval of the rear side crossing, providing a reliable data reference for multi-runway airports using the take-off point rear side crossing runway method. The precise safety interval between the front and rear aircraft is clarified, which significantly improves the overall operation efficiency of my country's airports.

[0054] The method provided by the present invention determines the engine jet influence analysis method and the influence distance of typical aircraft models in the twin-engine mode; the method is based on the computational fluid dynamics (CFD) numerical simulation method, constructs an engine jet model in the twin-engine mode, uses the CFD method to analyze the engine jet influence distance, studies the engine jet flow characteristics under different thrust states, determines the influence distance of typical aircraft models, and provides reliable data reference for the safety interval between the front and rear aircraft in the runway crossing mode after the take-off point at a multi-runway airport.

[0055] The method provided by the present invention collects the jet data of aircraft engines in actual operation based on the laser wind radar experiment; the method carries out the laser wind radar experiment, and intends to correct and improve the CFD model according to the real experimental data, so as to more accurately simulate the engine jet characteristics and flow field distribution at the time of aircraft takeoff.

[0056] The method provided by the present invention is based on the correction of the turbulence model through data assimilation. In order to further improve the accuracy of numerical simulation, the SST k-ω turbulence model is recalibrated using the ensemble Kalman filter algorithm based on the experimental data measured by lidar, and a comparative analysis is conducted on the degree of improvement in the prediction accuracy of the turbulence model.

[0057] The method provided by the present invention determines the crosswind resistance capability of a typical aircraft model based on mechanical analysis and CFD numerical simulation methods; by performing force analysis on the aircraft under crosswind conditions and performing numerical simulation on the crosswind flow of a typical aircraft model, aerodynamic forces, moment coefficients, lateral force coefficients, etc. are calculated, and the crosswind resistance critical speed of a typical aircraft model under different friction coefficients and weights is determined.

[0058] The method provided by the present invention determines the safe interval for crossing the runway at the rear side of the take-off point under a typical aircraft type combination; through the numerical simulation results of the front aircraft engine jet combined with the analysis of the rear aircraft's crosswind resistance capability, the safe interval distance for crossing the runway at the rear side of the take-off point under a typical aircraft type combination is given.

[0059] Other advantages, objectives and features of the present invention will be described in the following description to some extent, and to some extent, will be obvious to those skilled in the art based on the following examination and study, or can be taught from the practice of the present invention. The objectives and other advantages of the present invention can be realized and obtained through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] In order to make the purpose, technical solution and beneficial effects of the present invention clearer, the present invention provides the following drawings for explanation.

[0061] Figure 1 The flowchart of the method for calculating the safe interval for operation behind the take-off point under the influence of the engine jet is shown in FIG.

[0062] Figure 2 Schematic diagram of radar measurement in PPI scanning mode.

[0063] Figure 3 is the outflow calculation domain.

[0064] Figure 4 The engine surface grid and the inner and outer duct grid layout.

[0065] Figure 5 Schematic diagram of the engine boundary layer grid and the flow field encryption area grid.

[0066] Figure 6 This is the Gas Turb 14 calculation software interface.

[0067] Figure 7 This is the velocity attenuation curve at the axis of one side of the nozzle in the twin-engine mode.

[0068] Figure 8 This is a contour map of the jet velocity.

[0069] Fig. 9 Schematic diagram of the coupled iteration of data assimilation and CFD solution.

[0070] Fig.10 Figure 1 is a flow chart of the data assimilation procedure.

[0071] Fig.11 Velocity distributions calculated for the Standard Model and the initial sample set.

[0072] Fig.12 This is the speed comparison at the engine axis after assimilation.

[0073] Fig.13 Error comparison before and after assimilation.

[0074] Fig.14 The overall calculation domain (left: top view, right: side view).

[0075] Fig.15 This is the process of analyzing the stability of the subsequent machine.

[0076] Fig.16 Flowchart of collecting real data for experiments.

[0077] Fig.17 Create a flow chart for the twin-engine nozzle model.

[0078] Fig.18 Flowchart for calculating boundary conditions using software. DETAILED DESCRIPTION

[0079] The present invention is further described below in conjunction with the accompanying drawings and specific embodiments so that those skilled in the art can better understand the present invention and implement it, but the embodiments are not intended to limit the present invention.

[0080] Example 1

[0081] like Figure 1 As shown, Figure 1 The flowchart of the method for calculating the safe interval of the rear side of the take-off point under the influence of the engine jet is as follows. The method for calculating the safe interval of the rear side of the take-off point under the influence of the engine jet provided in this embodiment includes the following steps:

[0082] S1: Acquire the engine jet data of the aircraft. In this embodiment, the engine jet data of the take-off aircraft is measured at the end of the airport runway by using a laser wind measurement radar;

[0083] S2: Use CFD method to build a twin-engine nozzle model;

[0084] After establishing the twin-engine nozzle model, this embodiment uses the data assimilation method combined with high-reliability laser radar wind measurement experimental data to correct the numerical simulation model (twin-engine nozzle model) and calculate the influence distance of the typical aircraft engine jet;

[0085] S3: Use CFD method to establish the aircraft side wind model under the influence of engine jet;

[0086] S4: The two are combined to calculate the minimum safe interval for rear crossing, providing a reliable data reference for multi-runway airports using the rear crossing runway method of the take-off point;

[0087] In this embodiment, the influence distance of the typical aircraft engine jet is calculated based on the twin-engine nozzle model constructed by the computational fluid dynamics method in step S2; and the aircraft crosswind resistance (critical wind speed value) is calculated based on the rear crosswind model of the aircraft under the influence of the engine jet established by the CFD method in step S3, so as to calculate the minimum safety interval for rear crossing;

[0088] The safety interval calculation method provided in this embodiment combines laser wind radar measurement experiments, data assimilation, computational fluid dynamics (CFD) numerical simulation and other methods; it clarifies the precise safety interval between the front and rear aircraft, provides reliable data for crossing the runway behind the take-off point, and can significantly improve the overall operation efficiency of the airport.

[0089] In this embodiment, the engine jet data is obtained through a ground experiment on the engine jet using a laser wind measuring radar. The laser wind measuring radar is an instrument that uses a laser beam to measure wind speed and direction. A laser beam is emitted by a laser to measure the size and direction of the Doppler frequency shift, and the speed and direction of the particles are calculated, thereby obtaining the engine jet flow field information.

[0090] like Fig.16 As shown, Fig.16 The specific process of collecting real data through experiments in this embodiment is as follows:

[0091] Engine jet measurement experiment;

[0092] Select the location of the experimental airport runway and determine the measurement plan;

[0093] Laser wind radar on-site commissioning;

[0094] Measure the jet data of takeoff aircraft at a fixed azimuth angle;

[0095] Combined with the QAR data of the take-off aircraft for post-processing;

[0096] Obtain the relationship between takeoff aircraft distance and jet speed.

[0097] Specifically, the PPI scanning mode of the laser radar is used for collection. The radar detection distance is 75m-4000m. Considering the dissipation distance of the jet, the data within 75m-1440m is processed. The radar measures the radial velocity, and the configured scanning speed is 1 degree / s. The data result is saved in the format of: the velocity data corresponding to a certain measurement azimuth and a certain radius, such as Figure 2 As shown in the dots, Figure 2 Schematic diagram of radar measurement in PPI scanning mode.

[0098] The radial velocity measured by the radar is back-projected to obtain the jet velocity along the runway backwards, and the velocity distribution in polar coordinates with the radar as the center is converted to the distribution in rectangular coordinates with the radar as the center. The X-axis is a line parallel to the runway starting from the radar origin, and the Y-axis is a line perpendicular to the runway. The specific position of the aircraft on the runway is determined by combining the relative position of the runway and the radar, thereby obtaining the jet velocity data behind the take-off aircraft.

[0099] Combined with the aircraft's latitude and longitude information in the aircraft's QAR data, the aircraft's position on the runway is determined in a rectangular coordinate system with the radar as the origin. The aircraft's position and the jet data measured by the radar are matched to the time recorded in the QAR data and the original data of the lidar. As the aircraft taxis forward, the distance between the aircraft and the measurement point continues to increase. The jet velocity captured at this position point continues to decrease as the aircraft moves away, and the corresponding relationship between the jet velocity and the distance from the aircraft can be analyzed.

[0100] The real engine jet data collected through the experiment provides valuable and highly reliable experimental data for the numerical simulation of the jet impact distance of typical aircraft engines.

[0101] This embodiment conducts a numerical simulation study on the influence distance of a typical aircraft engine jet: a computational fluid dynamics (CFD) numerical simulation method is used to construct a twin-engine nozzle model to calculate the influence distance of a typical aircraft engine jet.

[0102] The specific steps of establishing the twin-engine nozzle model are as follows: a 1:1 single-engine nozzle model is constructed using CATIA software according to the aircraft model manual and the engine manual; then, based on the actual aircraft engine hoisting position distance, the twin-engine structure model is drawn using the symmetrical rotation method in CATIA software to restore the twin-engine relationship in the actual aircraft state at a 1:1 ratio.

[0103] The structural relationship of the twin-engine nozzle in this embodiment: The twin-engine nozzle model is used to calculate the jet impact distance of a typical aircraft model, and the fluid mechanics equations are solved by numerical simulation to more accurately simulate the flow details of the twin-engine nozzle model, including turbulent motion and dynamic pressure distribution. This provides data support for the use of data assimilation methods to correct the twin-engine nozzle model, thereby accurately determining the flow field range of the engine jet when the aircraft is taking off.

[0104] In the process of building the simulation model, this method adopts a 1:1 restoration ratio of the nozzle, and takes into account the jet mixing effect and ground effect of the twin-engine nozzle, which is close to the actual situation of the jet operation. According to the selected model and its engine size, the CATIA software is used to build a refined engine three-dimensional geometric model.

[0105] like Fig.17 As shown, Fig.17A flow chart is established for the twin-engine nozzle model. The process of establishing the twin-engine nozzle model is as follows:

[0106] Twin-engine nozzle model construction;

[0107] Collect aircraft model manual and engine manual data;

[0108] Use CATIA to build a 1:1 model of a single-engine nozzle;

[0109] Use CATIA to stretch, rotate, chamfer, etc. to build a symmetrical double-engine structure;

[0110] Check dimensions and assembly relationships, and optimize details;

[0111] Get the twin-engine nozzle model.

[0112] The twin-engine nozzle flow calculation in this embodiment is performed according to the following steps:

[0113] S21: Determine the engine model and its parameters, including the detailed size and shape of the engine nozzle and the details of the inner wall of the nozzle. After building a 1:1 model of a single-engine nozzle, use the CATIA software function to determine the longitudinal center plane of the aircraft, symmetrically copy the single-engine model, and obtain a twin-engine nozzle model.

[0114] S22: determining a fluid calculation domain, wherein the fluid calculation domain includes the inner and outer ducts of the nozzle and the external flow calculation domain;

[0115] S23: Determine the engine grid and the layout of the inner and outer duct grids;

[0116] S24: Calculate the boundary conditions of the inner and outer ducts of the engine at takeoff thrust using Gas Turb 14 software; obtain the total temperature and total pressure of the inner and outer ducts of the engine at full thrust (100% N1), reduced thrust (85% N1), and reduced thrust (70% N1) working conditions;

[0117] S25: Set the parameters of the internal and external duct flow monitoring points, initialization (Initialization) and iteration steps. After all settings are completed, perform solution calculation.

[0118] The research object of this method is the engine jet, considering the internal flow of the engine nozzle, and truly restoring the aerodynamic laws of the engine. The fluid calculation domain includes the internal and external ducts of the nozzle and the external flow calculation domain, and the construction of the entire calculation domain is also completed in CATIA software.

[0119] The construction of the computational domain needs to be fully combined with the physical object to be studied. The research object is the jet of a takeoff aircraft. Considering the ground effect, a rectangular outer flow domain is selected, and the lower surface is the ground boundary. The computational domain is as follows: Figure 3 As shown, Figure 3 is the outflow calculation domain.

[0120] The specific dimensions of the far field are as follows: the range of the outer flow domain in the X direction extends from 1.6m to 800m to ensure a sufficiently long space for jet calculation, and the flow field ranges in the Z and Y directions are 0m to 30m and -45m to 45m respectively. Ensure that the impact of the jet on the flow field of different cross sections along the flow direction can be included in the calculation range. The engine grid is drawn using ICEM, and partitions are created using the Blocking tool. The partitions define the approximate area of ​​the grid division and the placement of the grid on the geometric model. The grid is then divided by setting parameters such as the number of grids and the growth rate, and finally the quality of the grid is checked.

[0121] In the drawing process, the surface mesh is drawn first, and the surface mesh adopts a hybrid mesh, such as Figure 4 As shown, Figure 4 The engine surface grid and the inner and outer duct grid layout. The boundary layer grids are drawn on the inner and outer duct walls of the engine respectively, and the height of the first grid layer is 4×10 -6 m, and after generating the boundary layer mesh, the unstructured mesh is used to fill the remaining space. A cylindrical densification area is used after the engine outlet to obtain a more accurate axis velocity. The densification area extends to 600m after the nozzle, such as Figure 5 As shown, Figure 5 Schematic diagram of the engine boundary layer grid and the flow field encryption area grid.

[0122] Gas Turb is a professional software for simulating and optimizing the performance of gas turbines. The software is mainly used in the engineering field, especially in the fields of aerospace, energy, etc., to design, analyze and improve the performance of gas turbines. Gas Turb can simulate various working conditions and operating states of gas turbines, including combustion process, air flow, thermodynamic cycle, etc. This method is based on the Gas Turb software to calculate the boundary conditions of the internal and external ducts of the engine under takeoff thrust. The latest version of the software operation interface is as follows: Figure 6 As shown, Figure 6 This is the Gas Turb 14 calculation software interface, where: Figure 6 The interface diagram can be replaced by a flowchart of the software calculating the boundary conditions, such as Fig.18 As shown, Fig.18 Flowchart for calculating boundary conditions using software.

[0123] The process of using software to calculate boundary conditions provided in this embodiment includes the following steps:

[0124] Obtaining boundary conditions for engine jet calculation;

[0125] Input the known parameters from the engine manual into the Gas Turb software to build a numerical model of the turbofan engine;

[0126] Obtaining state parameters of various engine components during the cyclic working process;

[0127] Generate an engine baseline diagram to obtain parameters such as temperature and pressure of the engine's internal and external ducts;

[0128] Export boundary conditions.

[0129] This embodiment provides calculations using Gas Turb software, and the calculated boundary conditions are derived to obtain the total temperature and total pressure of the engine inner and outer ducts under full thrust (100% N1), reduced thrust (85% N1), and reduced thrust (70% N1) working states;

[0130] The software for numerical calculation of flow field is Fluent commercial software. In the General setting, the density-based solver is selected to solve the Reynolds-averaged Navier-Stokes (NS) equation. The turbulence model adopts the SSTk-ω model, the finite volume method is adopted, the flux discretization format is the ASUM format, the time term adopts the implicit format, the calculated fluid medium is set to air, the density calculation method adopts the ideal gas model, the specific heat ratio is set to a constant of 1006.43 J / (kgK), the thermal conductivity is set to a constant of 0.0242 W / (m*K), and the viscosity coefficient is calculated using the Sutherland formula. In order to ensure the stability and convergence of the calculation, the CFL number is taken as 1.

[0131] In the boundary condition setting, the lower surface of the rectangular far field is set as the wall boundary condition, and the far field outlet is set as pressure-outlet; the inlet of the inner and outer ducts of the nozzle is set as pressure-inlet, and the total temperature (TotalTemperature) and total pressure (Gauge Total Pressure) parameters of the inlet are set, and the remaining wall surfaces are set as wall boundary conditions.

[0132] Finally, set the parameters of the internal and external duct flow monitoring points, initialization (Initialization) and iteration steps. After all settings are completed, the software will perform solution calculations.

[0133] Post-processing of results is an important part of CFD numerical simulation. The post-processing software selected in this method is Tecplot flow field display software. By importing the Fluent flow field solution results, the flow field velocity, streamlines, cloud maps, etc. are displayed. Considering the influence distance and influence range of the engine jet,

[0134] This method mainly focuses on the spatial velocity attenuation variation law in the direction of the two engine axes under the condition of twin engines, as well as the spatial velocity attenuation variation law in the direction of the axis where the aircraft symmetry plane is located, to obtain the area with the maximum jet velocity of the engine of interest;

[0135] On the other hand, in order to study the stability of the rear machine, the velocity magnitude at the jet distance of 300 meters and 500 meters, the velocity attenuation law on the section perpendicular to the axis, and the velocity attenuation law on the section are also analyzed.

[0136] The following results are exported for analytical verification of numerical simulation.

[0137] Attenuation curve. In the twin-engine mode, extract the velocity data on a certain nozzle axis to draw the engine jet velocity attenuation curve. Through the jet axis velocity attenuation curve, you can intuitively see the change trend of the jet velocity, such as Figure 7 As shown, Figure 7 This is the velocity attenuation curve at the axis of one side of the nozzle in the twin-engine mode.

[0138] Contour cloud map. By drawing contour cloud maps, we can observe the spatial distribution of different physical quantities, such as pressure and velocity. The post-processing jet contour cloud map is shown in Figure 8. Figure 8 This is a contour map of the jet velocity.

[0139] The engine nozzle flow field data assimilation process based on experimental data provided in this embodiment is specifically as follows:

[0140] Based on the ground experiment of engine jet of laser wind radar and the numerical simulation of the influence distance of typical aircraft engine jet, the data assimilation method is used to modify the numerical simulation model with experimental data, and further analyze the assimilation effect. Therefore, two assimilation schemes are considered:

[0141] Scheme 1 assimilates the full-field turbulent eddy viscosity data predicted by the turbulence model; Scheme 2 assimilates the model empirical parameters of the existing turbulence model; For the above scheme 1, the full-field turbulent eddy viscosity data obtained by assimilation and update adopts the method of frozen unidirectional coupling to enter the subsequent CFD solution iteration; For the above scheme 2, the model empirical parameters obtained by assimilation are substituted back into the turbulence model in a bidirectional coupling manner to participate in the subsequent CFD solution iteration;

[0142] The full-field turbulence data obtained through assimilation, which is more consistent with the experimental prediction, can be used as a high-reliability sample for the study of turbulence field laws and turbulence models through theoretical analysis or data-driven methods;

[0143] In this embodiment, the engine turbulence model is established in step S23 as follows: the structural grid is selected to mesh the nozzle model, and the twin-engine nozzle geometry model (including the computational domain) completed by CATIA modeling in S21-S22 is imported into the Pointwise software. Secondly, the mesh control area tool is used to create partitions, which define the approximate area of ​​the mesh division and how to place the mesh on the geometric model. These partitions can be set by defining boundaries, vertices, lines, and surfaces on the geometric model. The mesh is then divided by setting parameters such as the number of meshes and the growth rate, and finally the quality of the mesh is checked.

[0144] The engine turbulence model provided in this embodiment is used for the turbulence field data assimilation of the subsequent twin-engine nozzle geometry model.

[0145] The data assimilation of this embodiment is a method. The twin-engine nozzle is a model, and the jet measured by the laser sidewind radar is data. Therefore, the twin-engine nozzle model is studied by using the data assimilation method and the jet data measured by the laser sidewind radar.

[0146] Now let’s introduce how to define the key states and steps in turbulence field data assimilation:

[0147] The flow field solved by CFD is taken as the system to be studied, and the system state is defined as:

[0148] X=[x1,x2,…,x N ] (5-1)

[0149] In the formula, x i is the vector composed of the state variables of the set member i, and N is the number of members in the set sample; for the above two assimilation schemes, the state vectors are defined as follows:

[0150] For option one:

[0151] x i =[μ i,1 ,μ i,2 ,…,μ i,n ,ξ i,1 ,ξ i,2 ,…,ξ i,M ] T (5-2)

[0152] Where n is the total number of cells in the CFD calculation grid, M represents the number of observation points, and μ i,j represents the value of the turbulent eddy viscosity at the cell center of the grid numbered j in the flow field of the i-th set member; i,k It represents the observable variable in the flow field of the i-th set member, and the subscript k indicates that its location corresponds to the k-th observation point;

[0153] For option 2, there are:

[0154] x i =[θ i,1 ,θ i,2 ,…,θ i,m ,ξ i,1 ,ξ i,2 ,…,ξ i,M ] T (5-3)

[0155] Where m is the number of turbulence model parameters involved in assimilation update, M represents the number of observation points, and θ i,j represents the value of the jth turbulence model parameter corresponding to the i-th set member; ξ i,k It represents the observable variable in the flow field of the i-th set member, and the subscript k indicates that its location corresponds to the k-th observation point.

[0156] In the above two definitions, i,k The represented variables can be pressure, velocity, friction coefficient, etc., which can be directly measured by experiments, or other quantities that can be indirectly measured by experiments. At the same time, high-precision numerical simulations, such as flow field data provided by direct numerical simulation (DNS) or large eddy simulation (LES), can also be regarded as high-precision experimental data and participate in the data assimilation process.

[0157] The covariance of the system state is defined as the covariance of the state matrix X, and the calculation process is:

[0158]

[0159] Among them, 1 N×N Represents an N×N dimensional all-1 matrix;

[0160] System observation is defined as:

[0161] Y=[y1,y2,…,y N ] (5-7)

[0162] y i =[ω i,1 ,ω i,2 ,…,ω i,M ] T (5-8)

[0163] Among them, ω i,k represents the observation result of the kth observation point of the flow field of the i-th set member;

[0164] For the covariance of the set observations, the observation vector can be regarded as a random variable in the same way as the variance of the system state, and the covariance of the observation value can be obtained through the set observations.

[0165] However, since the observation error distribution of experimental data is usually unknown, and there is usually only one set of observation results given by fluid mechanics experiments, the experimental data is approximated as the true observation value, and it is assumed that the observation error distribution between observation points is independent and normally distributed. Based on this, the observation true value of each point can be perturbed and sampled separately to generate a series of observation vectors as the observations of each set member. Directly calculating the covariance of the set of observation vectors under this distribution assumption usually leads to an excessively large condition number of the observation covariance matrix, which affects the stability of the Kalman filter gain solution inverse matrix process and easily causes assimilation update failure.

[0166] Therefore, the theoretical value of the observation error coordination difference matrix based on the above observation error distribution assumption is adopted as shown in Formula 5-9.

[0167]

[0168] Compared with directly taking the true value of the observation as the observation of each set member, the set of observation vectors generated by this method can provide additional randomness for the set members during the update, and at the same time appropriately increase the variance between the set members to avoid the variance between the members converging too small, which makes the assimilation update ineffective. At the same time, the observation error covariance matrix involved in the Kalman gain calculation is a diagonal matrix, which can significantly improve the reliability of the inverse matrix process of calculating the Kalman gain without the help of pseudo-inverse analysis.

[0169] The goal of this method is to assimilate the steady-state RANS flow field, so the state transfer equation of the system should be a steady-state RANS equation group. The process from the updated flow field CFD iterative solution to the next convergent flow field state is regarded as a state transfer step. At the same time, it is assumed that no additional uncertainty is introduced in the state transfer process (no systematic error).

[0170] The state transition process between adjacent assimilation steps t+1 and t can be expressed as shown in Equation 5-10.

[0171] X t+1 =GX t (5-10)

[0172] Where G is the equivalent operator of the process in which the CFD solver solves the updated states of each set member through the steady-state RANS equations until convergence.

[0173] The above expression can also be considered as the prediction process of obtaining the prior distribution of the next step set by forward integration of the updated set.

[0174] It should be noted that in the prediction step;

[0175] The updated turbulent eddy viscosity field in Scheme 1 is added to the calculation by one-way coupling freezing;

[0176] In the second solution, the SA parameter update result is added to the calculation through bidirectional coupling:

[0177] like Fig. 9 As shown, Fig. 9 This is a schematic diagram of the coupled iteration of data assimilation and CFD solution. The specific steps are as follows:

[0178] T = k assimilation step analysis field; where T is a discrete time step and k is a specific time step index. The assimilation step analysis field refers to the estimated result of the current physical field (temperature, pressure) obtained after data assimilation at the kth time step.

[0179] T = k assimilation step state quantity X; where the assimilation step state quantity X refers to the state of the entire system represented by vector X, obtained after data assimilation at the kth time step.

[0180] CFD coupling calculation (one-way and two-way coupling); CFD coupling calculation refers to the integration of the SST k-ω model with other models or physical processes such as SA to simulate the complex phenomena of engine jet.

[0181] T=k+1 assimilation step prediction field; where the assimilation step prediction field refers to the state of the next time step (i.e., k+1 time step) predicted by numerical simulation using the T=k assimilation step analysis field (or state quantity X) as the initial condition. This prediction field is the input of the next data assimilation step and is used to merge with the new experimental data to generate the analysis field of the next time step.

[0182] In scheme 1 and scheme 2, the method of generating the differential initialization flow field is achieved by perturbing the parameters of the SA model, as follows:

[0183] Before starting data assimilation, it is necessary to generate a set state vector with differences as a prior distribution, and it is necessary to design an initialization perturbation strategy based on the different characteristics of the state variables. Perturbation variables are mainly divided into two categories: single-valued system parameters and field distribution variables. Single-valued system parameters include flow field boundary conditions such as Reynolds number and related parameters of turbulence models such as SA model. Field distribution variables include the distribution of flow field variables such as pressure field, velocity field, density field, Reynolds stress field and turbulent eddy viscosity field. The perturbation of single-valued system parameters usually only requires designing the corresponding perturbation interval and perturbation distribution according to the reference value of the parameter and inputting them into the system. The perturbation of field distribution variables can be divided into direct and indirect methods. The direct method can superimpose random sampling of a given distribution on the reference value of each field distribution variable that needs to be disturbed. For the Reynolds stress tensor field, the Reynolds stress tensor can be decomposed and represented in the barycentric triangle, and the perturbation of the Reynolds stress tensor can be achieved by translating the feature points to the vertices or edges of the triangle. Although the initial samples can be generated by directly adding random sampling to the flow field solved by RANS, the continuity and physical properties of the flow field samples generated by this method are destroyed, and it will be difficult to achieve convergence in subsequent calculations. Indirect perturbation is to perturb the field distribution variables by perturbing the quantities related to the field distribution variables. For example, when generating a collection of turbulent eddy viscosity field samples, different turbulent eddy viscosity field samples can be indirectly generated by perturbing the turbulence model parameters or the flow field boundary conditions.

[0184] First, the disturbance amplitude range is selected for the SA model parameters;

[0185] Then, Latin Hypercube Sampling (LHS) is used to obtain the corresponding SA parameter distribution, and the perturbed SA parameters are substituted into the RANS solver, and the solution is initialized until convergence to obtain a differential eddy viscosity field;

[0186] The flow field variables of the samples generated in this way can well meet the requirements of continuity and physics due to the constraints of the SA model equation, which is convenient for the convergence of subsequent calculations;

[0187] Based on the tested data assimilation algorithm, write a computer program to realize the whole process of turbulence field data assimilation according to the specific functional requirements of turbulence field data assimilation;

[0188] As shown in the figure, Fig.10 This is a flowchart of the data assimilation program. The program is mainly written in Python and includes a series of parts such as the pre-processing module, initialization module, data assimilation core program module, and post-processing module. The program structure and operation process are shown in Figure 1. Fig.10 As shown, the various modules in the turbulence field data assimilation are as follows:

[0189] The pre-processing module is responsible for reading in key control parameters of data assimilation, such as the number of assimilation steps, the number of samples generated, the initialization sample perturbation strategy and the perturbation amplitude, determining the case type, identifying the type of variables used as observation constraints, and linking and corresponding the observation data with the CFD simulation calculation grid space through interpolation.

[0190] The initialization module is responsible for realizing the generation of set samples, creating corresponding working paths for the sample set, perturbing the corresponding SA model parameters according to the given strategy, and batch calling the external CFD solver to solve the initial flow field samples until convergence. The data assimilation core program module is mainly responsible for executing the core algorithm of data assimilation. The algorithm currently available is mainly the generalized Kalman filter algorithm based on the set. By reading the flow field variables of the set samples and combining the physical quantities corresponding to the experimental observations, the state vector matrix and the corresponding observation vector matrix for the Kalman filter assimilation algorithm are established, and the two are mapped through the observation matrix. At the same time, the corresponding error covariance matrix is ​​set according to the given observation error variance level. The generalized Kalman filter algorithm is called to update the state vector matrix to obtain the analysis field of the current step. The analysis values ​​of the set members are provided to the external CFD solver for a new round of prediction step calculation until the assimilation termination condition is reached.

[0191] The data assimilation core program module is mainly responsible for executing the core algorithm of data assimilation. The currently available algorithm is mainly a generalized Kalman filter algorithm based on a set. By reading the flow field variables of the set samples and combining the physical quantities corresponding to the experimental observations, a state vector matrix and a corresponding observation vector matrix for the Kalman filter assimilation algorithm are established, and the two are mapped through the observation matrix. At the same time, the corresponding error covariance matrix is ​​set according to the given observation error variance level. The generalized Kalman filter algorithm is called to update the state vector matrix to obtain the analysis field of the current step. The analysis values ​​of the set members are provided to the external CFD solver for a new round of prediction step calculations until the assimilation termination condition is reached.

[0192] The post-processing module is mainly used to realize the visualization output of the collective sample flow field and the analysis and output of the statistical information such as the variance, mean, and error of the state variables of the collective samples. At the same time, the post-processing module determines whether the current step number reaches the maximum assimilation step number and whether the sample prediction value and the observation error are within the expected range, which serves as the basis for whether the assimilation continues to iterate.

[0193] The process of data assimilation of the turbulence field is carried out according to the following steps:

[0194] S211: Setting assimilation parameters, including task name and type; setting of assimilated observation variables; setting of assimilation algorithm; setting of assimilation steps, etc.;

[0195] S212: Combine the computational grid file, experimental data file / high-precision numerical simulation results; interpolate and link the experimental data to the computational grid space; and output the interpolation results;

[0196] S221: Generate a collection sample by combining the disturbance parameter type, parameter disturbance method, and the number of collection samples.

[0197] S222: Then the external solver calculates and initializes the convergent flow field;

[0198] S231: Read the sample flow field;

[0199] S232: construct data assimilation state vector;

[0200] S233: construct observation vectors according to computational grid files, experimental data files / high-precision numerical simulation results, and perform data assimilation algorithm update steps; update flow field variables; external solver calculations;

[0201] S234: Calculate the converged flow field;

[0202] S241: flow field visualization output;

[0203] S242: Sample statistical information analysis;

[0204] S243: Combined with the error convergence condition, the maximum number of assimilation rounds t * >tmax, determine whether the end condition is met, if not, return to read the sample flow field;

[0205] S244: If yes, the task ends.

[0206] The grid used for calculation and assimilation in this embodiment is the grid provided by NASA's official website. The Kalman filter algorithm is used for assimilation. The data assimilation target is the velocity distribution of the jet centerline measured by the PIV experiment. There are two assimilation methods: (1) assimilation of SA model parameters and (2) assimilation of turbulent eddy viscosity coefficient distribution.

[0207] Based on the above research and verification of the data assimilation method, the ensemble Kalman filter algorithm is combined with the real takeoff aircraft engine jet velocity data collected by the lidar measurement experiment for data assimilation. In order to ensure that the assimilation set can capture enough flow field change information, the SST k-ω model parameter change range determined by the perturbation analysis method is set within 300% of the original value. The SST k-ω model assimilation parameter values ​​are shown in Table 1.

[0208] Table 1 Range of SST k-ω model assimilation parameters

[0209]

[0210] The calibration process first generates 20 samples of turbulence model constants through Latin hypercube sampling, and then inputs these 20 sets of sample parameter sets into the standard SST k-ω for simulation model calculation, and uses Tecplot software to post-process the calculation results to obtain the initial sample velocity prediction set. Fig.11 As shown, Fig.11 Velocity distribution calculated by the standard model and the initial sample set, where X represents the distance from a certain point on the rear side of the engine to the nozzle, X velocity represents the jet velocity on the engine axis, the blue curve represents the predicted velocity for each set of parameters in the set, and the red curve represents the predicted velocity for the standard turbulence model parameters.

[0211] The experimental data and the results of the assimilation set are substituted into the EnKF data assimilation process to perform ensemble Kalman filter data assimilation. The SST k-ω turbulence model modified by the ensemble Kalman filter algorithm is recalculated, the changes in the prediction accuracy of the turbulence model before and after assimilation are compared and analyzed, and the jet impact distances of the two models are recalculated.

[0212] The speed comparison at the engine axis on one side of a certain model is selected, and the axis speed curve of the left engine before and after assimilation and the experimental data analysis are extracted, such as Fig.12 As shown, Fig.12 The speed comparison at the engine axis after assimilation. The red curve represents the calculation result using the standard SST k-ω model, the blue curve represents the calculation result based on the SST k-ω model enhanced by EnKF, and the black triangle is the experimental measurement data (exp).

[0213] The relative error is used to evaluate the improvement of the optimized turbulence model. The relative error between the axis-horizontal jet velocity predicted by the SST k-ω model before and after assimilation and the experimental data is defined as:

[0214]

[0215] Here, U exp is the experimental measurement speed of the measurement point, U * Predict velocities for the model at the locations of the experimental data.

[0216] like Fig.13 As shown, Fig.13 To compare the errors before and after assimilation, the SST k-ω model with optimized constants after assimilation has a smaller relative error as the speed decays. Combined with the experimental data measured by lidar, the SST k-ω turbulence model was recalibrated using the ensemble Kalman filter algorithm, and the engine jet influence range of typical aircraft models was calculated.

[0217] The safety analysis of the aircraft affected by the front engine jet provided in this embodiment is as follows:

[0218] The CFD method is used to analyze and study the taxiing stability of the rear aircraft when crossing the runway from the rear side of the take-off point. The safety distance refers to the minimum distance between the front and rear aircraft required to ensure the safe crossing of the rear aircraft. Factors that affect the safety distance of the rear side crossing of the take-off aircraft include the geometric size, center of gravity, position of the front main wheel of the rear aircraft and the engine jet velocity distribution of the front aircraft. Therefore, the engine jet velocity distribution is calculated by combining experiments and calculations, the aerodynamic force and torque on the rear aircraft are calculated by CFD, and then the critical wind speed is calculated by the critical wind speed expression based on the maximum static friction. Finally, the safety distance is calculated by combining the engine jet velocity distribution and the critical wind speed.

[0219] In order to analyze the impact of the front engine jet on the rear aircraft, a CFD model is established in which the front engine jet directly acts on the rear aircraft at a certain distance to analyze the crosswind resistance of the rear-side crossing aircraft. The specific steps are as follows:

[0220] The CFD model in this embodiment refers to the model in the previous step S2, that is, two CFD models are established, one is the twin-engine jet model of the front aircraft, and the other is the aircraft crosswind force model of the rear aircraft.

[0221] S31: First, based on the aircraft shape data in the aircraft performance manual, the aerodynamic shape of the aircraft under study is drawn using the SoildWorks software;

[0222] S32: Secondly, the Pointwise software is used to establish the aircraft unstructured grid, and the quality of grid calculation is improved by smoothing the wall and density box;

[0223] Compared with the structured grid, the three-dimensional unstructured grid provided in this embodiment has greater flexibility and adaptability, can efficiently perform local grid refinement, and is convenient for grid deformation and movement, and can be used to process areas with complex and irregular aircraft shapes.

[0224] The specific process of establishing an unstructured grid is as follows: First, the aerodynamic shape of the aircraft under study drawn by SoildWorks software is imported into Pointwise software to determine the global size, local size, boundary layer parameters, etc. of the overall grid of the aircraft. Secondly, combining the advantages of different unit types, the aircraft surface grid is generated by a mixed mode of triangle / tetrahedron and quadrilateral / hexahedron, and the surface grid is extended to the entire calculation domain to generate a volume grid. At the same time, a boundary layer grid is generated, and the key areas are locally encrypted to capture the details of the flow field. Then, a grid quality check is performed, including unit size, unit shape, orthogonality, and aspect ratio. After that, the grid is optimized and corrected. If the grid quality does not meet the standard, the grid needs to be re-divided and the grid parameters adjusted. The grid is smoothed to reduce unit deformation and improve grid quality. Finally, the aircraft unstructured grid is exported for subsequent calculations.

[0225] S33: Finally, the model was calculated using Fluent software, and the results were post-processed using Tecplot software;

[0226] The calculation model provided in this embodiment is performed in the following manner:

[0227] First, import the aircraft unstructured grid file in Fluent software, check the grid information, including the number of cells, number of nodes, etc., and use the "Check" function to check the grid quality, such as the minimum cell volume, orthogonality, aspect ratio, etc. Ensure that the grid quality meets the calculation requirements. Secondly, set the basic parameters of the physical model, select the solver type as Pressure-Based, select Steady, SST k-ω turbulence model calculation, and define the material properties of the fluid and solid. Then, set all boundary conditions, including inlet, outlet, wall, symmetry, etc. Next, select SIMPLE pressure-velocity coupling algorithm and second-order upwind space discretization format. Adjust the appropriate relaxation factor to control the convergence speed. And set the convergence standard and monitor to monitor the changes of specific physical quantities, including pressure, velocity, force, etc. Finally, initialize and solve, select "Standard Initialization", set the number of iterations to 50,000, and start solving.

[0228] After the calculation is completed, post-processing and result analysis are carried out, and contour maps and velocity vector maps are drawn in Tecplot software to display the distribution of variables such as pressure, velocity, temperature, turbulence, etc. A report is also generated to output numerical results such as force, drag, and moment coefficients. This is used to analyze the aircraft's resistance to crosswind forces.

[0229] The computational domain is Fig.14 As shown, Fig.14It is the overall calculation domain (left: top view, right: side view); the x direction is the incoming flow direction, the y direction is the aircraft axis, the z direction is the direction perpendicular to the ground, and the length of the calculation domain is 800*800*300m.

[0230] The fluid motion control equations require boundary conditions, and the specific boundary conditions have a decisive effect on the calculation results. The boundary conditions of the calculation domain are shown in Table 2.

[0231] Table 2 Boundary condition settings of rear machine

[0232]

[0233] In this embodiment, the process of analyzing the stability of the rear machine is as follows: Fig.15 As shown, the calculation of the minimum safety interval for rear-side crossing in step S4 includes the following steps:

[0234] S41: First, through numerical simulation, the nozzle wake velocity distribution, including the entire velocity field, is obtained, and then the crosswind boundary conditions of a certain aircraft type crossing the runway are obtained;

[0235] S42: On the other hand, by simulating this crosswind boundary with different incoming crosswind speeds, numerical simulation of the crosswind incoming flow of typical aircraft models is carried out, and the approximate drag and moment coefficients are calculated. The critical wind speed after the rear tire reaches the maximum static friction is obtained through force coefficient calculation, which is used to analyze the aircraft's ability to resist crosswind forces.

[0236] S43: Finally, based on the critical wind speed and wake speed distribution, the safety distance is obtained by finding the position corresponding to the critical wind speed in the wake flow field.

[0237] Fig.15 It is the process of analyzing the stability of the rear machine;

[0238] Based on the previous analysis of the impact range of the aircraft engine jet and the stability of the rear aircraft against crosswinds, the minimum safety distance under different combinations of front and rear aircraft models can be obtained.

[0239] The above-described embodiments are only preferred embodiments for fully illustrating the present invention, and the protection scope of the present invention is not limited thereto. Equivalent substitutions or changes made by those skilled in the art based on the present invention are within the protection scope of the present invention. The protection scope of the present invention shall be subject to the claims.

Claims

1. A method for calculating the safety interval behind the take-off point under the influence of the engine jet, characterized in that: The following steps are involved: S1: Obtain the aircraft's engine jet data; S2: Use computational fluid dynamics to build a twin-engine nozzle model; use data assimilation methods and engine jet data to perform numerical simulations of the twin-engine nozzle model; S3: Use CFD method to establish a model of the aircraft subjected to crosswind under the influence of engine jet; S4: Calculate the minimum safe interval for rear crossing.

2. The method for calculating the safe interval for operation behind the take-off point under the influence of the engine jet according to claim 1, characterized in that: The engine jet data in step S1 is obtained by measuring the take-off aircraft at the end of the airport runway using a laser wind radar.

3. The method for calculating the safe interval for operation behind the take-off point under the influence of the engine jet as claimed in claim 1, characterized in that: The twin-engine nozzle model in step S2 is constructed according to the following steps: S21: determine the engine model and its parameters, and construct a three-dimensional geometric model of the engine; S22: determining a fluid calculation domain, wherein the fluid calculation domain includes the inner and outer ducts of the nozzle and the external flow calculation domain; S23: Determine the engine surface grid and the internal and external duct grid layout; S24: Calculate the boundary conditions of the inner and outer ducts of the engine at takeoff thrust; obtain the total temperature and total pressure of the inner and outer ducts of the engine at full thrust, reduced thrust, and reduced thrust working states; S25: Set the parameters of the internal and external duct flow monitoring points, initialization and iteration steps. After all settings are completed, perform the solution calculation.

4. The method for calculating the safe interval for operation behind the take-off point under the influence of the engine jet as claimed in claim 1, characterized in that: In step S2, the processing using data assimilation adopts any of the following two schemes: Option 1 assimilates the full-field turbulent eddy viscosity data predicted by the turbulence model; or Scheme 2 assimilates the empirical parameters of the existing turbulence model; For the above scheme 1, the full-field turbulent eddy viscosity data obtained by assimilation and update are entered into the CFD solution iteration by using the frozen one-way coupling method; For the above-mentioned scheme 2, the empirical model parameters obtained by assimilation are substituted back into the turbulence model in a two-way coupling manner to participate in the subsequent CFD solution iteration; The turbulence field data obtained through assimilation will be used as samples for studying the laws of turbulence fields and turbulence models.

5. The method for calculating the safe interval for operation behind the take-off point under the influence of the engine jet as claimed in claim 4, characterized in that: The specific steps of coupling data assimilation with CFD solution in the data assimilation processing in step S2 are as follows: T = k assimilation step analysis field; T = k assimilation step state quantity X; CFD coupling calculation, including one-way and two-way coupling calculation; T = k + 1 assimilation step prediction field.

6. The method for calculating the safe interval for operation behind the take-off point under the influence of the engine jet as claimed in claim 1, characterized in that: The data assimilation process in step S2 is performed according to the following steps: S211: Setting assimilation parameters, the parameters including task name and type; Assimilation observation variable setting; assimilation algorithm setting; assimilation step number setting, etc.; S212: Combine the computational grid file, experimental data file / high-precision numerical simulation results; interpolate and link the experimental data to the computational grid space; and output the interpolation results; S221: Generate a set sample by combining the disturbance parameter type, the parameter disturbance method, and the number of set samples; S222: Then the external solver calculates and initializes the convergent flow field; S231: Read the sample flow field; S232: construct data assimilation state vector; S233: construct observation vectors according to computational grid files, experimental data files / high-precision numerical simulation results, and perform data assimilation algorithm update steps; update flow field variables; external solver calculations; S234: Calculate the converged flow field; S241: flow field visualization output; S242: Sample statistical information analysis; S243: Combined with the error convergence condition, the maximum number of assimilation rounds t * >tmax, determine whether the end condition is met, if not, return to read the sample flow field; S244: If yes, the task ends.

7. The method for calculating the safe interval for operation behind the take-off point under the influence of the engine jet as claimed in claim 1, characterized in that: The step S3 uses the CFD method to establish a model of the aircraft under the influence of the engine jet and the crosswind, and is performed according to the following steps: S31: Draw the aerodynamic shape of the aircraft under study based on the aircraft shape data in the aircraft performance manual; S32: Establish aircraft unstructured grid; S33: Calculate the mesh model and use Tecplot software to post-process the results.

8. The method for calculating the safe interval for operation behind the take-off point under the influence of the engine jet as claimed in claim 1, characterized in that: The step S4 is to calculate the minimum safety interval for rear-side crossing, and the specific steps are as follows: S41: First, through numerical simulation, the velocity distribution of the nozzle wake is obtained, including the entire velocity field, and then the boundary conditions of a certain aircraft model affected by the crosswind are obtained; S42: Simulate the crosswind boundary with different incoming crosswind wind speeds, perform numerical simulation on the crosswind incoming flow of typical aircraft models, calculate the approximate resistance (force coefficient) and moment coefficient, and calculate the critical wind speed after the rear tire reaches the maximum static friction through the force coefficient; S43: Based on the crosswind critical wind speed and the engine jet velocity distribution, the safety distance is obtained by finding the corresponding position of the critical wind speed in the wake flow field.

9. A system for calculating the safety interval of the take-off point rear side under the influence of the engine jet, comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the method according to any one of claims 1 to 8 is implemented.

Images