WO2016169249A1 - 一种高速行驶交通工具仿真模型前处理模板及处理方法 - Google Patents

一种高速行驶交通工具仿真模型前处理模板及处理方法 Download PDF

Info

Publication number
WO2016169249A1
WO2016169249A1 PCT/CN2015/094563 CN2015094563W WO2016169249A1 WO 2016169249 A1 WO2016169249 A1 WO 2016169249A1 CN 2015094563 W CN2015094563 W CN 2015094563W WO 2016169249 A1 WO2016169249 A1 WO 2016169249A1
Authority
WO
WIPO (PCT)
Prior art keywords
simulation model
processing
analysis
model
grid
Prior art date
Application number
PCT/CN2015/094563
Other languages
English (en)
French (fr)
Inventor
尚克明
李文化
田爱琴
杜俊涛
宋波
蔡军爽
杜健
王学亮
Original Assignee
中车青岛四方机车车辆股份有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 中车青岛四方机车车辆股份有限公司 filed Critical 中车青岛四方机车车辆股份有限公司
Publication of WO2016169249A1 publication Critical patent/WO2016169249A1/zh

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00Computer-aided design [CAD]

Definitions

  • the invention belongs to the technical field of vehicle quantity model simulation processing, and particularly relates to a pre-processing template and a processing method for a high-speed traveling vehicle simulation model.
  • Modern railway vehicle design is a complex system engineering, usually composed of multiple professional subsystems.
  • the complexity of products and projects is increasing.
  • the excellent design scheme must be integrated and coordinated after each professional subsystem. Can be produced, so collaborative work has become an urgent need of enterprises, but also the main goal of enterprise information.
  • the fluid dynamics simulation model is more and more applied to train design, and the fluid mechanics based simulation model is mainly to establish the mathematical model of fluid mechanics, which not only includes the discretization and solution method of mathematical equations, but also Including the establishment of the calculation grid and the processing of the boundary conditions. Therefore, the pre-processing template of the high-speed vehicle simulation model is mainly to complete the establishment of the computational domain grid, the processing of the boundary conditions, etc., and due to the professional complexity of the discipline simulation, various The simulation analysis software is dispersed in various professions by a small number of design and simulation personnel. The knowledge and experience are scattered among individuals, which is difficult to accumulate and share, which is not conducive to the accumulation of enterprise knowledge and the formation of expert knowledge base. In the face of complex simulation analysis objects, standard methods are needed to help designers quickly understand the analysis process steps, quickly complete cumbersome parameter input, complex working condition definition and simulation report generation, and accumulate and share simulation knowledge and experience.
  • the present invention provides a pre-processing template and a processing method for a high-speed traveling vehicle simulation model, and provides a standard method and a process for establishing a computing domain grid and determining boundary conditions.
  • the calculation domain grid is generated by dividing the calculation domain into smaller, non-overlapping sub-domains or cells (grids), the number of cells determines the calculation accuracy, and the too small grid makes the calculation amount very It's huge, so the fineness of the grid determines the accuracy of the simulation model, the necessary computer hardware, and the computation time.
  • Boundary conditions The mechanical properties of real fluids are very complex. Under different external conditions, such as different temperatures and different speeds, the mechanical properties will vary greatly. Therefore, it is impossible to adapt a fluid model to all working conditions. Therefore, different simulation models need to be established according to the boundary conditions determined by different working conditions, and different simulation analysis standards and processes are designed.
  • a high-speed traveling vehicle simulation model pre-processing template including a to-be-processed simulation model, a working condition definition module, a naming standardization and surface processing module, a grid processing module, and Boundary condition module;
  • the working condition definition module is configured to determine various working conditions of an operating state of the simulation model to be processed
  • the naming standardization and surface processing module includes a surface processing module and a name setting module;
  • the surface treatment module is configured to smooth the surface of the simulation model to be processed to form a single connected body
  • the name setting module is configured to divide the surface-processed simulation model according to a surface division rule, and name the divided surface;
  • the grid processing module establishes different grid processing flows according to different working conditions, and meshes the area between the computing domain and the outer surface of the named simulation model to simulate the running state of the vehicle. Analysis; the calculation domain is set by a corresponding working condition;
  • the boundary condition module establishes different boundary conditions for the simulation model after the grid processing according to different working conditions, and performs simulation analysis according to the boundary conditions.
  • the working condition of the working condition definition module is divided into a scaling model analysis working condition, a steady running operating condition and an unsteady running working condition according to a reduction ratio of the to-be-processed simulation model;
  • the scaling model analysis condition is divided into numerical wind tunnel analysis working condition and dynamic model analysis working condition according to the running state of the simulation model to be processed.
  • the surface division rule includes: performing surface division on the geometric part to be studied according to the geometric feature of the surface-processed simulation model, and dividing into at least one surface.
  • grid processing flow is divided into numerical wind tunnel analysis, dynamic model analysis, steady analysis and unsteady analysis according to different working conditions.
  • the second technical solution adopted by the present invention to solve the technical problem is: a method for processing a pre-processing template of a high-speed traveling vehicle simulation model, which is characterized in that it comprises the following steps:
  • step S2 includes:
  • A1 Clean the parts on the surface of the simulation model and remove the parts that do not affect the analysis
  • step S3 includes:
  • the simulation model of the calculation domain is established, and different division processes are selected according to different model scales and calculation condition types, and are mainly divided into: numerical wind tunnel analysis, dynamic model analysis, steady analysis and Unsteady analysis meshing.
  • an automatic mesh regeneration function is provided to solve the problem existing in the mesh division
  • the grid is subjected to various inspections prior to step b4), including quality inspection, negative volume, curvature, grid distribution inspection.
  • the dynamic model analysis and the establishment of the unsteady analysis grid include:
  • an automatic mesh regeneration function is provided to solve the problem existing in the mesh division
  • the grid is subjected to various inspections prior to step c3), including quality inspection, negative volume, curvature, grid distribution inspection;
  • the dynamic mesh setting and the dynamic mesh function test are further included after the step c3).
  • the invention has the beneficial effects that a simulation analysis template is established for a specific analysis process, and the analysis steps, operation instructions and help information are displayed on the interface to guide the designer to perform the simulation operation step by step, standardize the simulation analysis process, and simplify the manual operation on the interface. Improve work efficiency; show the knowledge and practical experience of professional simulation analysts, so that designers can take less detours. Pass By templated the specific analysis process, not only can the threshold of simulation analysis be reduced, the influence of individual differences on the accuracy of simulation analysis results can be reduced, and the formation of the knowledge base of enterprise simulation experts is also facilitated.
  • FIG. 1 is a flow chart of a method for processing a template before a vehicle simulation according to the present invention
  • FIG. 2 is a schematic structural diagram of a vehicle pre-processing processing template provided by the present invention.
  • FIG. 3 is a flow chart of another method for processing a pre-worker template according to the present invention.
  • FIG. 4 is a schematic diagram of computation domain partitioning of a vehicle pre-processing template provided by the present invention.
  • the invention relates to a pre-processing template for a high-speed driving vehicle simulation model, in particular to a CFD simulation analysis of a railway vehicle, the template encapsulating a vehicle analysis process and a specification, the function of which corresponds to the steps of the simulation analysis process. According to the interactive operation provided by the simulation template, the whole process of CFD simulation analysis of the railway vehicle model can be completed.
  • This patent is a simulation analysis experience obtained on the basis of a large number of wind tunnel, dynamic model and real vehicle test data.
  • the technical scheme of the present invention will be described in detail below, as shown in Fig. 1 and Fig. 2
  • the naming standardization and surface processing module can be divided into several different types of problems according to the calculation conditions:
  • the scaled model analysis refers to the simulation of the scaled model test by the CFD method to reduce the actual vehicle according to a certain proportion, mainly including numerical wind tunnel analysis and dynamic model analysis.
  • Numerical wind tunnel analysis mainly includes bicycle open line operation;
  • dynamic model analysis mainly includes train operation such as tunnel passage, tunnel intersection, and open line intersection.
  • the wind tunnel model of the present invention is a simulation analysis performed in CFD.
  • the model made according to the actual vehicle reduction ratio is placed on the test platform of the test section.
  • the wind resistance of the train, the gas field around the train and the flow field are measured during the test.
  • the wind tunnel here is an annular wind tunnel consisting of a wind turbine and a wind tunnel.
  • the air ducts are connected end to end, and the airflow circulates in the air duct.
  • the wind speed is generated and controlled by the fan to achieve a stable air flow field at different speeds.
  • the wind tunnel model parameters should include the reduction of the car body model, the composition of the car body model attachment, the size and shape of the roadbed, and the size and shape of the wind tunnel.
  • a numerical wind tunnel reduction ratio model is to be produced.
  • the test needs to include tests on the body parts such as the head car, the intermediate car, the tail car, the bogie, the pantograph, the through-pass, and the outer windshield. Therefore, the numerical wind tunnel reduction ratio model is shown in Figure 1.
  • the included vehicle body geometry model is a three-dimensional scaled model of three-car grouping, including the head car, the intermediate car, the tail car, the bogie, the pantograph, the through-pass, Body parts such as the outer windshield.
  • the size parameters of the car body are also inconsistent due to different lengths of the car and the cross section of the car body, but the model reduction ratio can be specified as 1:8, and the car body ground clearance is performed according to the actual vehicle.
  • the scale is reduced.
  • the length of the three-car is 9 to 10 meters, and the height of the car is about 0.4 meters.
  • the train When the train is in constant speed and the external environment is ideal, that is, when the external atmospheric pressure is constant and the air density is constant, the train is set to a constant flow field, so the volume (volume flow rate) qV flowing through the unit per unit time is constant.
  • the mass (density) ⁇ of the fluid per unit volume is also constant.
  • the vehicle body geometry model needs to clarify the vehicle grouping form, the composition of the car body attachment (if there is no windshield, bogie, pantograph, etc.), ground form (such as flat, roadbed, elevated).
  • the steady flow field analysis mainly includes train running conditions such as crosswinds on the open line of the bicycle and no crosswinds on the open line of the bicycle.
  • tunnel model parameters may include three aspects: a tunnel cross-sectional shape, a tunnel length, and a diversion form of the tunnel entrance and exit.
  • the unsteady flow field analysis includes the train running conditions such as crosswind, open line intersection, tunnel passage, tunnel intersection, and elevated passage.
  • the standardized name of the car body name will be based on the pre-set name rules.
  • the naming rules here include: naming the parts according to the surface of the car body, such as: doors, windows, cars, etc. The front head, body parking space, etc., where the naming rules can be defined as a universal and extensible template.
  • the parts that need to be removed include the nut that protrudes from the surface of the car body, the bogie of the bogie and the pantograph part, and the installation groove of the equipment compartment and the door and window parts, which can also be attributed to no shrinkage.
  • parts with a feature size of less than 5 cm can be simplified for deletion;
  • Redundant geometric features can be understood as coincident and isolated geometric features. For example, if two faces coincide or have a face that is hanging somewhere, one of them needs to be deleted. It is impossible to know what part is because the parts are cleaned up. This problem can occur in any location;
  • the so-called single connected body means that the calculation model is a closed surface, and there is no partition inside the curved surface.
  • the tolerance of the identification seam can be set to 1/10 of the minimum grid size, but generally should be greater than 0.1mm, that is, the gap larger than the tolerance value is the seam or Holes need to be patched.
  • the S5 model is exported and imported, and the single connectivity model is imported into the CAE pre-processing module;
  • S6 identifies geometric features and performs surface naming according to naming standardization rules
  • Identifying geometric features is to extract the feature lines of the model and important parts, and prepare for the surface naming.
  • the feature lines to be extracted are ridge lines with less than 30° angle of the surface, contours of doors and windows, and so on.
  • the feature line is segmented by the feature line to obtain the surface in the feature line, and then named, such as the surface inside the door outline is named as the door.
  • Face quality is greater than 0.5 and face proximity is greater than 0.1.
  • Face quality is twice the ratio of the inscribed circle radius of each side of the mesh to the radius of the circumscribed circle of each vertex.
  • Approach is the ratio of the distance between two adjacent grids to the grid size.
  • the so-called calculation domain is the outer boundary of the train flow field analysis.
  • the grid is generated in the space between the boundary and the train model.
  • This space is the calculation domain, that is, the space for the flow field simulation analysis.
  • the width dimension is generally the train model. 5 to 10 times the length of the feature in the width direction, the minimum distance between the longitudinal direction airflow inlet and the model should be greater than 3 times the length feature size, and the minimum distance between the airflow outlet and the model should be greater than 5 to 10 times the length feature size.
  • Grid processing may require different grid generation software such as ICEM, Tgrid, etc., so the grid needs to be converted between different software.
  • ICEM grid generation software
  • Tgrid grid needs to be converted between different software.
  • solver post-processing module conversion Star-CCM+ template is converted to ICEM, Tgrid, Fluent, Ensight, etc., the corresponding unit group name is unchanged, and the grid attribute is automatically converted. The cell group name does not change when the Fluent and Ensight read results are processed.
  • the input condition of the grid processing module is the geometric model processed by the surface, and different division processes are selected according to different calculation conditions, which are mainly divided into: numerical wind tunnel grid processing, dynamic model grid mathematics, steady state grid processing and Unsteady state mesh processing.
  • the model processed by the surface is mostly a dbs file.
  • S2 processes the model surface mesh and sets the surface mesh density according to the geometric features of the model.
  • the rule is to capture the minimum details of the model. Since the wind tunnel model is 1:8 scaled, the feature size can be as small as a millimeter, and for a vehicle model without shrinking, the minimum feature is on the order of centimeters;
  • the grid encryption zone and the given boundary layer parameters are set.
  • the encryption mesh size is related to the scale reduction of the model. For the whole vehicle model, it is 4 to 8 cm. For wind tunnel models, the number is 1 to 2 cm.
  • the layer of fluid attached to the solid surface is called the boundary layer.
  • the boundary layer is generally more clearly defined for the first layer. It needs to be 1 to 5 mm according to the ratio of the different models, and the thickness between the layers from the inside to the outside. The degree cannot be greater than the size of the face mesh of this layer of mesh.
  • S5 provides grid re-generation to solve problems in meshing
  • the grid re-generation function is to redistribute the grid density according to the flow condition in the flow field calculation process, that is, for the complicated flow position in the calculation process, the grid encryption or sparse is automatically performed according to the turbulence degree and the like, Improve computational efficiency and reduce the number of grids and improve calculation accuracy.
  • the quality of the grid has a great influence on the solution of the late flow field.
  • the high-quality grid is the basis of the high-precision flow field solution. Therefore, the grid quality check of the generated grid is needed.
  • the body mesh quality check standard generally the Cell quality is greater than 0.00001
  • Cell quality uses the Gaussian and least squares method to calculate the distance between the centroid of the volume mesh and the centroid of the adjacent mesh and the distance between the centroid of the surface mesh and the surface mesh.
  • a unit with no negative volume and negative curvature in the volume mesh, Volume change is generally greater than 0.00001
  • Volume change is the ratio of the volume of any body mesh to the adjacent largest volume mesh.
  • the negative volume mainly checks the distortion degree of the body mesh.
  • Different pre-processing softwares differ in the calculation of negative volume. For example, ICEM first constructs a Jacobian matrix in a certain direction for each mesh edge, and then orthogonalizes the matrix. If the value is 0, it is a negative volume. There should be no negative volume or negative curvature in the volume mesh.
  • the grid distribution check is mainly to analyze whether the transition between different density grids and grids of different regions is reasonable.
  • Grid optimization generally has a grid optimization command. By performing redistribution of the mesh portion with poor quality, the mesh quality can be improved, and the negative volume unit can be changed into a positive volume by modifying the normal direction of one side of the mesh.
  • the model processed by the surface is mostly a dbs file.
  • Typical working conditions generally include three conditions: open line intersection, tunnel passage, and tunnel intersection.
  • the open line rendezvous and the tunnel rendezvous are simulated when the two trains are in orbit and the rendezvous occurs. As shown in Figure 4, they are generally divided into seven computing domains, including A, B, C, D, E, F, and G.
  • the upper train runs from left to right and the lower train runs from right to left.
  • the grids in the calculation fields B and E are unchanged and the train runs forward at the same speed.
  • the calculation fields A and F will generate grid vacancies due to the forward running of the train, and a new grid fill will be automatically generated during the flow field solving process.
  • the vacant area, while the calculation fields C and D will automatically delete the coincident grid due to the forward movement of the train.
  • the calculation domain G encloses the six computational domains and is stationary.
  • the inner boundary is the outer boundary of the six computational domains.
  • the outer boundary is the outer boundary of the entire flow field.
  • S3 processes the model surface mesh and sets the surface mesh density according to the geometric features of the model.
  • the rule is to capture the minimum details of the model. Since the wind tunnel model is 1:8 scaled, the feature size can be as small as a millimeter, and for a vehicle model without shrinking, the minimum feature is on the order of centimeters;
  • the grid encryption zone and the given boundary layer parameters are set.
  • the encryption mesh size is related to the scale reduction of the model. For the whole vehicle model, it is 4 to 8 cm. For wind tunnel models, the number is 1 to 2 cm.
  • the layer of fluid attached to the solid surface is called the boundary layer.
  • the boundary layer is generally more clearly defined for the first layer. It needs to be 1 to 5 mm according to the ratio of the different models, and the thickness between the layers from the inside to the outside is also Cannot be larger than the size of the face mesh of this layer of mesh.
  • S6 provides grid re-generation to solve problems in meshing
  • the grid re-generation function is to redistribute the grid density according to the flow condition in the flow field calculation process, that is, for the complicated flow position in the calculation process, the grid encryption or sparse is automatically performed according to the turbulence degree and the like, Improve computational efficiency and reduce the number of grids and improve calculation accuracy.
  • the body mesh quality check standard generally the Cell quality is greater than 0.00001
  • Cell quality uses the Gaussian and least squares method to calculate the distance between the centroid of the volume mesh and the centroid of the adjacent mesh and the distance between the centroid of the surface mesh and the surface mesh.
  • Volume change is the ratio of the volume of any body mesh to the adjacent largest volume mesh.
  • the negative volume mainly checks the distortion degree of the body mesh.
  • Different pre-processing softwares differ in the calculation of negative volume. For example, ICEM first constructs a Jacobian matrix in a certain direction for each mesh edge, and then orthogonalizes the matrix. If the value is 0, it is a negative volume. There should be no negative volume or negative curvature in the volume mesh.
  • the grid distribution check is mainly to analyze whether the transition between different density grids and grids of different regions is reasonable.
  • Grid optimization generally has a grid optimization command. By performing redistribution of the mesh portion with poor quality, the mesh quality can be improved, and the negative volume unit can be changed into a positive volume by modifying the normal direction of one side of the mesh.
  • the moving mesh that is, the partial mesh in the computing domain, moves with time.
  • the dynamic mesh needs to be used, which is generally used in the state of train crossing, tunnel passing, etc., because these workers In the case of the following cars at different times, the change of the train position will require the grid near it to change along with it, so dynamic grid settings are required.
  • the mesh scale should be smaller than the product of the train speed and the calculated minimum time step.
  • the minimum mesh scale is not more than 0.1 m, and the corresponding time step can be obtained according to the train running speed.
  • the dynamic grid test is to move the train model in the entire calculation domain according to the specified speed and time step. At this time, the flow field is not solved, but whether the negative volume, grid interlace error, etc. will occur, if not Errors such as negative volume and network interleaving occur, indicating that the dynamic mesh test passes and a formal solution can be performed.
  • railway vehicles include EMUs and urban rail vehicles. Due to the different environments in which different types of vehicles operate, the operating conditions used in the simulation analysis are also different.
  • the boundary conditions can be divided into the following types according to different working conditions:
  • the scaling model analysis includes the following conditions:
  • the experimental platform is generally 1:8, three-car grouping/one-half car grouping, the model is placed on the roadbed, the roadbed is placed on the floor, and the model, roadbed and floor have no relative motion.
  • Dynamic model test platforms generally include subgrades, tracks, models, and the like.
  • the steady state analysis includes the working conditions: the running condition of the real car open line, the model is: the model with the real car 1:1, using the eight-car complete vehicle grouping, including the track, the roadbed, the ground, the train is still, The track, subgrade, and ground speed are the same as the incoming flow speed.
  • the general inlet is the speed inlet
  • the outlet is the pressure outlet
  • the wall is the physical wall surface.
  • the ground can be set as the moving ground and the fixed wall.
  • the surface of the train model is set as a physical wall surface, and special parts such as an air inlet and outlet of the air conditioner can be separately set, wherein the air inlet and outlet of the air conditioner is a pressure inlet.
  • the unsteady state analysis includes the following conditions: tunnel passage and intersection conditions, elevated passage conditions, open line intersection conditions, and three conditions in the unsteady state analysis.
  • the tunnel, ground, and elevated models are stationary.
  • the train moves at normal operating speed and adopts a 1:1 model with the actual vehicle.
  • the general inlet, outlet and wall are all pressure outlets, and the ground is a fixed wall surface.
  • the contact surface between different calculation domains is the interface surface, which realizes the data transmission between two adjacent surfaces.
  • the surface of the train model is set as a physical wall surface, and special parts such as an air inlet and outlet of the air conditioner can be separately set, wherein the air inlet and outlet of the air conditioner is a pressure inlet.
  • the simulation model pre-processing template of the invention establishes a simulation analysis template for a specific analysis process, and displays the analysis steps, operation instructions and help information on the interface to guide the designer to perform the simulation operation step by step, standardize the simulation analysis process; simplify the manual on the interface Operation, improved work Efficiency; show the knowledge and practical experience of professional simulation analysts, so that designers can take less detours.
  • the specific analysis process not only can the threshold of simulation analysis be reduced, the influence of individual differences on the accuracy of simulation analysis results can be reduced, and the formation of enterprise simulation expert knowledge base is also beneficial.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Hardware Design (AREA)
  • Evolutionary Computation (AREA)
  • Geometry (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Aerodynamic Tests, Hydrodynamic Tests, Wind Tunnels, And Water Tanks (AREA)

Abstract

公开了一种高速行驶交通工具仿真模型前处理模板及处理方法,包括待处理仿真模型、工况定义模块、命名标准化及曲面处理模块、网格处理模块和边界条件模块;工况定义模块用于确定待处理仿真模型运行状态的各种工况;命名标准化及曲面处理模块,包括表面处理模块和名称设定模块;表面处理模块用于,将待处理仿真模型的表面进行光滑处理;名称设定模块用于,将经过表面处理的仿真模型按照曲面划分规则进行曲面划分,并对划分的曲面进行命名;网格处理模块,根据不同的工况,建立不同的网格处理流程,进行模拟交通工具运行状态的仿真分析。该模板及方法可以简化仿真分析中繁琐的前处理工作,提高工作效率,减少人为错误。

Description

一种高速行驶交通工具仿真模型前处理模板及处理方法
本申请要求于2015年04月21日提交中国专利局、申请号为201510190610.7、发明名称为“一种高速行驶交通工具仿真模型前处理模板及处理方法”的中国专利申请的优先权,其全部内容通过引用结合在本申请中。
技术领域
本发明属于车量模型仿真处理技术领域,特别涉及一种高速行驶交通工具仿真模型前处理模板及处理方法。
背景技术
现代铁路车辆设计是一项复杂的系统工程,通常由多个专业的子系统组成,产品和项目的复杂度日益增加,优异的设计方案必然是在对其各个专业子系统进行了综合、协调之后才能产生,因此协同工作成为企业的一项迫切需求,也是企业信息化的主要目标。
从现代制造业的发展趋势来看,计算机仿真技术在新产品的开发过程中将发挥越来越重要的作用。与传统研发方法相比,结合了计算机仿真技术的现代研发方法可以节约大量的人力、物力、财力并且大大地缩短产品研发周期。随着有限元等仿真理论基础的成熟和计算机硬件的快速发展,仿真工作的计算精确度越来越高。所以在国际知名的铁路行业、车辆行业的企业,在新产品开发的初期,基于仿真的数字化样机已经开始替代大部分的物理样机作为产品设计的依据。
因此,基于流体力学的仿真模型越来越多的应用到列车设计中,而基于流体力学的仿真模型的建立主要就是建立流体力学的数学模型,它不仅包括数学方程的离散化及求解方法,还包括计算网格的建立、边界条件的处理,因此,高速行驶交通工具仿真模型前处理模板主要就是完成计算域网格的建立、边界条件的处理等,而由于学科仿真的专业复杂性,各种仿真分析软件分散在各专业由少数设计和仿真人员操作,知识和经验分散在个体,难以积累和共享,不利于企业知识积累和专家知识库的形成;同时 面对复杂的仿真分析对象,需要有标准的方法来帮助设计人员快速了解分析过程步骤,快速完成繁琐的参数输入、复杂的工况定义和仿真报告生成,并将仿真知识和经验积累和共享。
发明内容
为了克服上述缺点,本发明提供一种高速行驶交通工具仿真模型前处理模板及处理方法,提供了建立计算域网格和确定边界条件的标准方法及流程。
其中,计算域网格的生成是将计算域划分为较小的、不重叠的子域或单元(网格),单元的数目决定了计算精度,而太细小的网格又使计算量变得非常庞大,因此网格的细密程度决定了:仿真模型求解的精度、必需的计算机硬件和计算时间。
边界条件:真实流体的力学性质是很复杂的,在不同的外部条件下,比如温度不同、速度不同其力学特性会有很大的差异,因此要一个流体模型以适应所有工况是不可能的,因此需要根据不同的工况确定的边界条件建立不同的仿真模型,设计不同的仿真分析标准和流程。
本发明解决其技术问题所采用的技术方案之一是:一种高速行驶交通工具仿真模型前处理模板,包括待处理仿真模型、工况定义模块、命名标准化及曲面处理模块、网格处理模块和边界条件模块;
所述工况定义模块,用于确定待处理仿真模型运行状态的各种工况;
所述命名标准化及曲面处理模块,包括表面处理模块和名称设定模块;
所述表面处理模块,用于将所述待处理仿真模型的表面进行光滑处理,形成单连通体;
所述名称设定模块,用于将经过表面处理的仿真模型按照曲面划分规则进行曲面划分,并对划分的曲面进行命名;
所述网格处理模块,根据不同的工况,建立不同的网格处理流程,将计算域与命名后的仿真模型的外表面之间的区域进行网格划分以便进行模拟交通工具运行状态的仿真分析;所述计算域由相应的工况设定;
所述边界条件模块,根据不同的工况,为网格处理后的仿真模型建立不同的边界条件,并依据所述边界条件进行仿真分析。
进一步,所述工况定义模块的工况,根据待处理仿真模型的缩小比例划分为缩比模型分析工况、定常运行工况和非定常运行工况;
所述缩比模型分析工况按照待处理仿真模型运行状态划分为数值风洞分析工况和动模型分析工况。
进一步,所述曲面划分规则包括:根据经过表面处理的仿真模型的几何特征,对待研究的几何部位进行曲面划分,至少划分为一个曲面。
进一步,所述网格处理流程根据不同的工况划分为数值风洞分析、动模型分析、定常分析和非定常分析。
本发明解决其技术问题所采用的技术方案之二是:一种高速行驶交通工具仿真模型前处理模板的处理方法,其特征在于,包括如下步骤:
S1,首先结合待处理仿真模型运行状态的各工况对仿真分析进行分类;
S2,对待处理仿真模型的表面进行处理,生成单连通体的仿真模型;将经过表面处理的仿真模型按照曲面划分规则划分为至少一个曲面,并为划分的曲面进行命名;
S3,根据所述仿真分析分类设定命名后的仿真模型的计算域以及在所述计算域中进行网格划分;
S4,根据所述仿真分析分类设定边界条件。
进一步,所述步骤S2包括:
a1)对待处理仿真模型表面的零件清理,删除不影响分析的零部件;
a2)对仿真模型表面进行几何清理,清理冗余的点、线、面和体;
a3)再对仿真模型表面进行修补,分析清理后的模型,进行表面修补使之成为单连通体。
a4)将生成单连通体的仿真模型导入CAE前处理模块;
a5)将导入CAE前处理模块的仿真模型按照曲面划分规则划分为至少一个曲面,并为划分的曲面进行命名;
a6)在所述CAE前处理模块中对命名后的仿真模型进行检查与修复,修复不满足网格生成要求的曲面。
进一步,所述步骤S3包括:
d1)为修复完的仿真模型按照仿真模型要进行分析的工况建立计算 域;
d2)为建立完计算域的仿真模型生成网格。
进一步,在生成网格前,所述建立完计算域的仿真模型,根据不同的模型尺度和计算工况类型选择不同的划分流程,主要划分为:数值风洞分析、动模型分析、定常分析和非定常分析网格划分。
进一步,所述数值风洞分析和所述定常分析网格的建立包括:
b1)、处理模型面网格,根据模型几何特征进行面网格密度设置;
b2)、网格分布及附面层设置;
b3)、网格生成;
b4)、对网格进行质量优化;
优选地,在步骤b3)后,提供自动网格重新生成功能,以解决网格划分中存在的问题;
优选地,在步骤b4)前对网格进行各种检查,包括质量检查、负体积、曲率、网格分布检查。
进一步,所述动模型分析和非定常分析网格的建立包括:
c1)、处理模型面的网格,按照计算工况划分计算域,并根据所述车辆模型的几何特征进行面网格密度设置;
c2)、根据动模型分析和非定常分析对网格分布及附面层进行设置,并生成网格;
c3)、对网格进行质量优化;
优选地,在步骤c2)后,提供自动网格重新生成功能,以解决网格划分中存在的问题;
优选地,在步骤c3)前对网格进行各种检查,包括质量检查、负体积、曲率、网格分布检查;
优选地,在步骤c3)后还包括动网格设置和动网格功能测试。
本发明的有益效果是,针对具体分析过程建立仿真分析模板,将分析步骤、操作说明和帮助信息展现到界面上引导设计人员按部就班地进行仿真操作,规范仿真分析过程;简化界面上的手工操作,提高工作效率;将专业仿真分析专家的知识、实践经验展现出来,使设计人员少走弯路。通 过将具体分析过程模板化,不但可以降低仿真分析的门槛,减少人员个体差异对仿真分析结果准确性的影响,还有利于企业仿真专家知识库的形成。
附图说明
为了更清楚地说明本发明实施例或现有技术中的技术方案,下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍,显而易见地,下面描述中的附图仅仅是本发明的实施例,对于本领域普通技术人员来讲,在不付出创造性劳动的前提下,还可以根据提供的附图获得其他的附图。
图1是本发明提供的一种车辆仿真前处理模板的方法流程图;
图2是本发明提供的一种车辆仿真前处理模板的结构示意图;
图3是本发明提供的另一种车辆仿真前处理模板的方法流程图;
图4是本发明提供的一种车辆仿真前处理模板的计算域划分示意图。
具体实施方式
下面将结合本发明实施例中的附图,对本发明实施例中的技术方案进行清楚、完整地描述,显然,所描述的实施例仅仅是本发明一部分实施例,而不是全部的实施例。基于本发明中的实施例,本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例,都属于本发明保护的范围。
对高度行驶的交通工具,例如铁路机车、客车、货车、动车组、城轨、地铁等不同的车辆,通过开发通用的CFD模型通用前处理模板,可以对仿真分析中繁琐的前处理工作进行标准化和规范化,借助流程自动化工具在实现规范处理的同时,可以提高工作效率,减少人为错误,因此在研发和测试过程中建立CFD模型通用前处理模板尤为重要。
本发明涉及一种高速行驶交通工具仿真模型前处理模板,特别针对铁路车辆进行CFD仿真分析,模板封装了车辆分析过程和规范,其功能与仿真分析流程的步骤相对应。按照仿真模板所提供的交互式操作,即可完成铁路车辆模型CFD仿真分析的全过程。
本专利是在进行了大量的风洞、动模型、实车试验数据基础上得到的仿真分析经验,下面结合对本发明的技术方案进行详细描述,如图1和图2所 示的高速行驶交通工具仿真模型前处理模板的结构和处理流程。
1、所述命名标准化及曲面处理模块根据计算工况可以划分为几种不同类型的问题:
1)缩比模型分析工况
缩比模型分析指通过CFD方法对实车按照一定比例进行缩小而进行缩比模型试验的仿真模拟,主要包括数值风洞分析和动模型分析。数值风洞分析主要包括单车明线运行;动模型分析主要包括隧道通过、隧道交会、明线交会等列车运行情况。
铁路机车车辆,特别是高速列车,在运行中受到空气相对流动作用,产生阻力,机车车辆周围的气压和气流也在发生变化,特别是在列车通过隧道和列车相对开行(会车)时,这种气流作用会更加明显。为此,必须进行列车的气动力学的研究,研究的主要手段通常采用风洞试验。
为了满足车辆的风洞试验,因此需要建立风洞模型,在设定的风洞中安置车辆模型,研究气体流动及其与模型的相互作用,以了解车辆空气动力学特性的一种空气动力实验方法;而本发明的风洞模型是在CFD中进行的仿真分析。
按实车缩小比例制作的模型放置在试验段的试验平台上,试验时测定列车风阻力、列车周围气场和流场等。这里的风洞是一个环状的风道,由风机、风道组成。风道首尾相连,气流在风道中循环流动。风速由风机产生和控制,实现不同速度的稳定空气流场。
风洞模型参数应该包括车体模型缩小比例、车体模型附件的组成、路基尺寸与外形、风洞的尺寸和外形。
为了用于数值风洞试验而要制作数值风洞缩比模型,由于试验需要包含对:头车、中间车、尾车、转向架、受电弓、贯通道、外风挡等车体部件的试验,因此数值风洞缩比模型如图1所示,包括的车体几何模型为三节车编组的三维缩比模型,包括头车、中间车、尾车、转向架、受电弓、贯通道、外风挡等车体部件。
针对不同的车型由于车长、车体断面等不同,车体的尺寸参数也不一致,但是模型缩小比例可以规定为1:8,车体离地间隙按照实际车辆进行同 比例的缩小。一般来说三节车车长的9至10米,车高在0.4米左右。
2)定常运行工况
流体(气体、液体)流动时,若流体中任何一点的压力,速度和密度等物理量都不随时间变化,则这种流动就称为定常流动,也可称之为“稳态流动”或者“恒定流动”;反之,只要压力,速度和密度中任意一个物理量随时间而变化液体就是作非定常流动或者说液体作时变流动。
列车在恒速运动过程中并且外界环境为理想状态,即外界大气压恒定、空气密度恒定时,可是设定列车处于定常流场,因此流体每单位时间流过的体积(体积流量)qV为常量,流体每单位体积的质量(密度)ρ也是常量。
在定常流场分析中,车体几何模型需要明确车辆编组形式、车体附件的组成(如有无风挡、转向架、受电弓等)、地面形式(如平地、路基、高架)。
定常流场分析主要包括单车明线运行有侧风和单车明线运行无侧风等列车运行情况。
3)非定常运行工况
列车过隧道压缩波和膨胀波、隧道出口微气压波等都是高速列车系统中典型的非定常空气动力学问题,因此本发明对列车非定常流场分析不仅需要设定车体的几何模型,也需要设定隧道模型,其中,隧道模型参数可以包括隧道断面形状、隧道长度和隧道出入口的导流形式三方面内容。
非定常流场分析包括单车明显运行有侧风、明线交会、隧道通过、隧道交会、高架通过等列车运行情况。
2、针对上述不同的类型开发不同的建模流程,据此将命名标准化及曲面处理模块开发成具有多个分支可供选择的模板集合。
1)表面特征命名标准化和车体表面处理
表面特征命名标准化在CFD分析和后处理工作中起到重要的作用,是实现自动报告生成及自动数据处理的前提条件,同时良好的命名规则能为用户提供快速而且直观的信息,为仿真分析提供便利。
车体名称标准化命名将根据预先设定的名称规则进行,这里的命名规则包括:按照车体表面的部件命名,例如:车门、车窗、车厢等,还可以 车头、车身车位等,在这里命名规则可以定义为通用的可扩展的模板。
虽然不同类型的模型,例如风洞模型、定常模型、非定常模型等需要采用不同的处理方法,但是在这个步骤的目的是为后续的网格划分等处理准备好二维和三维的几何模型。所以一般都会经历如下几个步骤:
S1零件清理,删除、简化不影响分析的零部件;
S2对于整车气动性能分析,需要删除的零部件包括突出车体表面的螺帽、转向架和受电弓部位的管路,以及设备舱、门窗部位的安装槽,也可以归结为对于没有缩比似的整车模型,特征尺寸小于5cm的零部件均可进行删除简化;
S3几何清理,清理冗余的点、线、面和体问题,举例冗余的点、线、面和体指的是什么部分;
此处冗余的几何特征可以理解为重合的和孤立几何特征,比如两个面重合或者有孤悬某处的面的话就需要删除其中之一,无法明确是什么部分,因为由于零件清理完成后任何位置都有可能出现这种问题;
S4表面修补,分析清理后的模型,进行表面修补使之成为间单连通体;
所谓单连通体就是说计算模型为一个封闭的曲面,曲面内部没有隔断。
表面修补可以理解为见缝补缝,见洞补洞,识别缝的容差可以设置为最小网格尺寸的1/10,但一般应大于0.1mm,即认为大于此容差值的间隙就是缝或者洞,需要进行修补。
S5模型导出与导入,将单连通模型导入CAE前处理模块;
S6识别几何特征,根据命名标准化规则进行曲面命名;
识别几何特征就是提取出模型和重要部位的特征线,为曲面命名做准备,需要提取的特征线有曲面夹角小于30°的棱线,门窗轮廓线等。通过特征线分割模型得到特征线内的曲面,然后进行命名,如车门轮廓线内的曲面命名为车门。
S7模型检查与修复,检查导入模型的质量,修复不满足网格生成要求的曲面;
质量检查的标准是Face quality大于0.5,face proximity大于0.1。Face quality即网格各边的内切圆半径与各个顶点外接圆半径之比的两倍,face  proximity即两个相邻网格的距离与网格尺寸的比值。
S8计算域建立;
所谓计算域就是进行列车流场分析的外部边界,网格就在这个边界与列车模型之间的空间生成,这个空间就是计算域,即进行流场仿真分析的空间,其宽度尺寸一般为列车模型宽度方向特征长度的5至10倍,长度方向气流入口与模型最小距离应该大于长度特征尺寸的3倍,气流出口与模型最小距离应该大于长度特征尺寸的5至10倍。
3、网格处理模块开发
网格处理时可能会需要调用不同的网格生成软件如ICEM、Tgrid等,所以网格需要在不同的软件之间转换。对于前处理模块和求解器后处理模块转换:Star-CCM+模板转换到ICEM、Tgrid、Fluent、Ensight等,相应单元组名称不变,网格属性自动转换。在后处理Fluent、Ensight读取结果时,单元组名称不变。网格处理模块的输入条件是经过曲面处理的几何模型,根据不同的计算工况选择不同的划分流程,主要分为:数值风洞网格处理、动模型网格数理、定常状态网格处理和非定常状态网格处理。
(一)如图3所示,数值风洞网格处理和定常状态网格处理的步骤为:
S1读取已经命名好的计算域;
其中,经过曲面处理的模型,多为dbs文件。
S2处理模型面网格,根据模型几何特征进行面网格密度设置,其规则是以捕捉模型的最小细节为目标。对于风洞模型由于是1:8缩比的,所以特征尺寸最小可达毫米量级,而对于没有缩小的整车模型,最小特征在厘米量级;
S3网格分布及附面层设置;
具体地,设定网格加密区和给定附面层参数。
对于流场较为复杂的部位如头型、转向架、风挡、受电弓、尾车后部等部位需要进行加密,加密网格尺寸与模型缩小比例相关,对于整车模型在4至8厘米,对于风洞模型多在1至2厘米。
附于固体表面的一层流体称为附面层,附面层一般对第一层有较为明确的规定,根据不同模型比例需要在1至5毫米,且由内向外各层之间的厚 度也不能大于这层网格的面网格的尺寸。
S4网格生成;
S5提供网格重新生成功能,以解决网格划分中存在的问题;
其中,网格重新生成功能是在流场计算过程中根据流动情况对网格疏密进行重新分布,即针对计算过程中流动复杂的位置,根据湍流度等指标自动进行网格加密或者稀疏,如此提高计算效率降低网格数量,并提高计算精度。
S6网格质量检查、负体积、曲率、网格分布检查;
其中,网格的质量好坏对后期流场求解影响很大,高质量的网格是高精度流场求解的基础,因此需要对已生成的网格进行网格质量检查。此处为体网格质量检查标准:一般为Cell quality大于0.00001,Cell quality即采用高斯和最小二乘法计算体网格质心与相邻网格质心的距离和面网格质心与面网格的距离;体网格中不能有负体积和负曲率的单元,Volume change一般大于0.00001,Volume change即任一体网格与相邻的最大体网格的体积之比。网格分布没有统一的标准,根据实际情况具体分析。其中,当网格本身产生大变形造成自我体积的内面跑到外面时被判断为负体积,网格扭曲过大是产生负体积的主要原因,因此负体积主要考核体网格的扭曲程度。不同的前处理软件对负体积的计算方法有所区别,例如:ICEM首先将每个网格边按照一定方向构造雅可比矩阵,然后对矩阵正交化处理,如果值为0则为负体积。体网格中不能有负体积和负曲率。
其中,网格分布检查主要是分析不同密度网格、不同区域网格之间的过渡是否合理。
S7网格质量优化;
网格优化一般有网格优化命令,通过进行质量差的网格部分重新分布,可以提高网格质量,通过修改网格某一面的法线方向将负体积单元变为正体积。
(二)如图3所示,动模型网格处理和非定常状态网格处理的步骤为:
S1读取已经命名好的计算域;
其中,经过曲面处理的模型,多为dbs文件。
S2根据计算工况进行不同计算域的划分;
典型工况一般有明线交会、隧道通过、隧道交会三种工况。
明线交会和隧道交会是模拟两列车在轨道运行,并发生交会的情况,如图4示,一般分为7个计算域,包括A、B、C、D、E、F、G。
上方列车由左向右运行,下方列车由右向左运行。计算域B和E内网格不变与列车以相同的速度向前运行,计算域A和F由于列车向前运行会产生网格空缺,在流场求解过程中会自动生成新的网格填充空缺区域,而计算域C和D会自动删除由于列车向前运行而产生重合的网格,计算域G包裹着这个六个计算域且静止不动,内边界为六个计算域的外边界,外边界为整个流场的外边界。
进行隧道通过仿真时就只有4个计算域A、B、C、G,没有D、E、F。
S3处理模型面网格,根据模型几何特征进行面网格密度设置,其规则是以捕捉模型的最小细节为目标。对于风洞模型由于是1:8缩比的,所以特征尺寸最小可达毫米量级,而对于没有缩小的整车模型,最小特征在厘米量级;
S4网格分布及附面层设置;
具体地,设定网格加密区和给定附面层参数。
对于流场较为复杂的部位如头型、转向架、风挡、受电弓、尾车后部等部位需要进行加密,加密网格尺寸与模型缩小比例相关,对于整车模型在4至8厘米,对于风洞模型多在1至2厘米。
附于固体表面的一层流体称为附面层,附面层一般对第一层有较为明确的规定,根据不同模型比例需要在1至5毫米,且由内向外各层之间的厚度也不能大于这层网格的面网格的尺寸。
S5网格生成;
S6提供网格重新生成功能,以解决网格划分中存在的问题;
其中,网格重新生成功能是在流场计算过程中根据流动情况对网格疏密进行重新分布,即针对计算过程中流动复杂的位置,根据湍流度等指标自动进行网格加密或者稀疏,如此提高计算效率降低网格数量,并提高计算精度。
S7网格质量检查、负体积、曲率、网格分布检查;
此处为体网格质量检查标准:一般为Cell quality大于0.00001,Cell quality即采用高斯和最小二乘法计算体网格质心与相邻网格质心的距离和面网格质心与面网格的距离;体网格中不能有负体积和负曲率的单元,Volume change一般大于0.00001,Volume change即任一体网格与相邻的最大体网格的体积之比。网格分布没有统一的标准,根据实际情况具体分析。
其中,当网格本身产生大变形造成自我体积的内面跑到外面时被判断为负体积,网格扭曲过大是产生负体积的主要原因,因此负体积主要考核体网格的扭曲程度。不同的前处理软件对负体积的计算方法有所区别,例如:ICEM首先将每个网格边按照一定方向构造雅可比矩阵,然后对矩阵正交化处理,如果值为0则为负体积。体网格中不能有负体积和负曲率。
其中,网格分布检查主要是分析不同密度网格、不同区域网格之间的过渡是否合理。
S8网格质量优化;
网格优化一般有网格优化命令,通过进行质量差的网格部分重新分布,可以提高网格质量,通过修改网格某一面的法线方向将负体积单元变为正体积。
S9动网格设置,动网格功能测试;
动网格即计算域中部分网格是随着时间运动的,在分析计算模型随时间运动的状态时需要采用动网格,一般在列车交会、隧道通过等状态下会用到,因为这些工况下列车在不同的时刻位置不同,列车位置的变化会要求其附近的网格也跟着一起变化,因此需要进行动网格设置。
在进行动网格设置时,网格尺度应小于列车速度与计算最小时间步长的乘积,通常最小网格尺度不大于0.1m,相应的时间步长可以根据列车运行速度求得。
动网格测试就是将列车模型在整个计算域中按照规定的速度和时间步长进行运动,此时并不进行流场求解,而是检查是否会出现负体积、网格交错等错误,如果没有出现负体积和网络交错等错误,说明动网格测试通过,可以进行正式的求解。
4、边界条件设置
铁路车辆包括动车组,城轨车辆,由于不同类型的车辆运行所处的环境不同,所以在仿真分析过程中其使用的工况也各有不同。根据不同的工况将边界条件可分为如下几种类型:
1)缩比模型分析包括如下工况:
a)风洞分析工况
采用缩比模型,实验平台一般为1:8,三车编组/一节半车编组,模型置于路基之上,路基放置在地板上,模型、路基、地板没有相对运动。
b)动模型分析工况
采用缩比模型,实验平台一般为1:16或者1:8,三车编组,模型置于轨道之上,模型静止,轨道速度与来流速度相同。动模型试验平台一般包括路基、轨道、模型等。
2)定常状态分析包括的工况为:实车明线运行工况,建立的模型为:与实车1:1的模型,采用八车整车编组,还包括轨道、路基、地面,列车静止,轨道、路基、地面速度与来流速度相同。
在进行定常计算时,一般入口为速度入口,出口为压力出口,壁面为物理壁面,根据求解工况的不同,地面可以设置为移动地面和固定壁面。列车模型表面设置为物理壁面,特殊的部位如空调进出风口等可单独设定,其中,空调进出风口为压力入口。
3)非定常状态分析包括的工况为:隧道通过及交会工况、高架通过工况、明线交会工况,非定常状态分析中的三个工况中,隧道、地面、高架模型均静止,列车以正常运行速度运动,采用与实车1:1的模型。
在进行非定常计算时,一般入口、出口和壁面均为压力出口,地面为固定壁面,不同据计算域之间的接触面为interface面,实现相邻两个面之间的数据传递。列车模型表面设置为物理壁面,特殊的部位如空调进出风口等可单独设定,其中,空调进出风口为压力入口。
本发明的仿真模型前处理模板针对具体分析过程建立仿真分析模板,将分析步骤、操作说明和帮助信息展现到界面上引导设计人员按部就班地进行仿真操作,规范仿真分析过程;简化了界面上的手工操作,提高了工 作效率;将专业仿真分析专家的知识、实践经验展现出来,使设计人员少走弯路。通过将具体分析过程模板化,不但可以降低仿真分析的门槛,减少人员个体差异对仿真分析结果准确性的影响,还有利于企业仿真专家知识库的形成。
以上所述仅是本发明的较佳实施例而已,并非对本发明作任何形式上的限制,虽然本发明已以较佳实施例揭露如上,然而并非用以限定本发明,任何熟悉本专利的技术人员在不脱离本发明技术方案范围内,当可利用上述提示的技术内容作出些许更动或修饰为等同变化的等效实施例,但凡是未脱离本发明技术方案的内容,依据本发明的技术实质对以上实施例所作的任何简单修改、等同变化与修饰,均仍属于本发明方案的范围内。

Claims (10)

  1. 一种高速行驶交通工具仿真模型前处理模板,其特征在于,包括待处理仿真模型、工况定义模块、命名标准化及曲面处理模块、网格处理模块和边界条件模块;
    所述工况定义模块,用于确定待处理仿真模型运行状态的各种工况;
    所述命名标准化及曲面处理模块,包括表面处理模块和名称设定模块;
    所述表面处理模块,用于将所述待处理仿真模型的表面进行光滑处理,形成单连通体;
    所述名称设定模块,用于将经过表面处理的仿真模型按照曲面划分规则进行曲面划分,并对划分的曲面进行命名;
    所述网格处理模块,根据不同的工况,建立不同的网格处理流程,将计算域与命名后的仿真模型的外表面之间的区域进行网格划分以便进行模拟交通工具运行状态的仿真分析;所述计算域由相应的工况设定;
    所述边界条件模块,根据不同的工况,为网格处理后的仿真模型建立不同的边界条件,并依据所述边界条件进行仿真分析。
  2. 根据权利要求1所述的一种高速行驶交通工具仿真模型前处理模板,其特征在于,所述工况定义模块的工况,根据待处理仿真模型的缩小比例划分为缩比模型分析工况、定常运行工况和非定常运行工况;
    所述缩比模型分析工况按照待处理仿真模型运行状态划分为数值风洞分析工况和动模型分析工况。
  3. 根据权利要求1所述的一种高速行驶交通工具仿真模型前处理模板,其特征在于,所述曲面划分规则包括:根据经过表面处理的仿真模型的几何特征,对待研究的几何部位进行曲面划分,至少划分为一个曲面。
  4. 根据权利要求3所述的一种高速行驶交通工具仿真模型前处理模板,其特征在于,所述网格处理流程根据不同的工况划分为数值风洞分析、动模型分析、定常分析和非定常分析。
  5. 根据权利要求1-4任意一项所述的一种高速行驶交通工具仿真模 型前处理模板的处理方法,其特征在于,包括如下步骤:
    S1,首先结合待处理仿真模型运行状态的各工况对仿真分析进行分类;
    S2,对待处理仿真模型的表面进行处理,生成单连通体的仿真模型;将经过表面处理的仿真模型按照曲面划分规则划分为至少一个曲面,并为划分的曲面进行命名;
    S3,根据所述仿真分析分类设定命名后的仿真模型的计算域以及在所述计算域中进行网格划分;
    S4,根据所述仿真分析分类设定边界条件。
  6. 根据权利要求5所述的高速行驶交通工具仿真模型前处理模板的处理方法,其特征在于,所述步骤S2包括:
    a1)对待处理仿真模型表面的零件清理,删除不影响分析的零部件;
    a2)对仿真模型表面进行几何清理,清理冗余的点、线、面和体;
    a3)再对仿真模型表面进行修补,分析清理后的模型,进行表面修补使之成为单连通体;
    a4)将生成单连通体的仿真模型导入CAE前处理模块;
    a5)将导入CAE前处理模块的仿真模型按照曲面划分规则划分为至少一个曲面,并为划分的曲面进行命名;
    a6)在所述CAE前处理模块中对命名后的仿真模型进行检查与修复,修复不满足网格生成要求的曲面。
  7. 根据权利要求5所述的高速行驶交通工具仿真模型前处理模板的处理方法,其特征在于,所述步骤S3包括:
    d1)为修复完的仿真模型按照仿真模型要进行分析的工况建立计算域;
    d2)为建立完计算域的仿真模型生成网格。
  8. 根据权利要求7所述的高速行驶交通工具仿真模型前处理模板的处理方法,其特征在于,在生成网格前,所述建立完计算域的仿真模型,根据不同的模型尺度和计算工况类型选择不同的划分流程,主要划分为:数值风洞分析、动模型分析、定常分析和非定常分析网格划分。
  9. 根据权利要求8所述的高速行驶交通工具仿真模型前处理模板的处理方法,其特征在于,所述数值风洞分析和所述定常分析网格的建立包括:
    b1)、处理模型面网格,根据模型几何特征进行面网格密度设置;
    b2)、网格分布及附面层设置;
    b3)、网格生成;
    b4)、对网格进行质量优化;
    优选地,在步骤b3)后,提供自动网格重新生成功能,以解决网格划分中存在的问题;
    优选地,在步骤b4)前对网格进行各种检查,包括质量检查、负体积、曲率、网格分布检查。
  10. 根据权利要求8所述的高速行驶交通工具仿真模型前处理模板的处理方法,其特征在于,所述动模型分析和非定常分析网格的建立包括:
    c1)、处理模型面的网格,按照计算工况划分计算域,并根据所述车辆模型的几何特征进行面网格密度设置;
    c2)、根据动模型分析和非定常分析对网格分布及附面层进行设置,并生成网格;
    c3)、对网格进行质量优化;
    优选地,在步骤c2)后,提供自动网格重新生成功能,以解决网格划分中存在的问题;
    优选地,在步骤c3)前对网格进行各种检查,包括质量检查、负体积、曲率、网格分布检查;
    优选地,在步骤c3)后还包括动网格设置和动网格功能测试。
PCT/CN2015/094563 2015-04-21 2015-11-13 一种高速行驶交通工具仿真模型前处理模板及处理方法 WO2016169249A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201510190610.7A CN104866648B (zh) 2015-04-21 2015-04-21 一种高速行驶交通工具仿真模型前处理模板及处理方法
CN201510190610.7 2015-04-21

Publications (1)

Publication Number Publication Date
WO2016169249A1 true WO2016169249A1 (zh) 2016-10-27

Family

ID=53912473

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2015/094563 WO2016169249A1 (zh) 2015-04-21 2015-11-13 一种高速行驶交通工具仿真模型前处理模板及处理方法

Country Status (2)

Country Link
CN (1) CN104866648B (zh)
WO (1) WO2016169249A1 (zh)

Cited By (26)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107748822A (zh) * 2017-10-31 2018-03-02 华晨汽车集团控股有限公司 一种发动机附件仿真分析中的配重方法
CN108763806A (zh) * 2018-06-07 2018-11-06 谢红太 高速列车过特长双线隧道时列车及隧道表面所受静压的数值模拟分析方法
CN109214139A (zh) * 2018-11-16 2019-01-15 中车株洲电力机车有限公司 一种受电弓强度仿真计算快速建模方法
CN110197041A (zh) * 2019-06-10 2019-09-03 大连交通大学 一种多物理场工况下多目标表面细观结构优化方法及系统
CN110309575A (zh) * 2019-06-26 2019-10-08 武汉蓝恩汽车技术有限公司 商用车前下防护装置断裂失效评估方法及系统
CN110489835A (zh) * 2019-08-05 2019-11-22 安徽建筑大学 基于Ansys软件的自然通风与墙体蓄热耦合模拟方法
CN111241627A (zh) * 2020-01-03 2020-06-05 同济大学 一种计入胶层的异质车身门盖件模态数值模拟及优化方法
CN111753372A (zh) * 2020-06-11 2020-10-09 雷沃工程机械集团有限公司 一种挖掘机整机热平衡仿真方法及系统
CN112199907A (zh) * 2020-10-15 2021-01-08 乌江渡发电厂 一种基于cfd的水电站地下厂房多指标通风效果的分析方法
CN112257307A (zh) * 2020-09-29 2021-01-22 北京科技大学 一种水下履带装备行走阻力的仿真计算方法及系统
CN112597680A (zh) * 2020-12-21 2021-04-02 北京慧物科联科技有限公司 一种应用于卫浴洁具的自动化有限元建模方法
CN113076597A (zh) * 2021-04-13 2021-07-06 吉林大学 一种面向cfd仿真的重卡车身快速建模方法
CN113312728A (zh) * 2021-06-24 2021-08-27 南京航空航天大学 一种双列圆锥滚子轴承失油过程中的流场仿真方法及系统
CN113343351A (zh) * 2021-05-18 2021-09-03 重庆长安汽车股份有限公司 一种汽车制动盘热-结构耦合分析方法
CN113505544A (zh) * 2021-06-18 2021-10-15 清华大学 基于有限体积法自行车运动虚拟数值风洞系统
CN113536701A (zh) * 2020-04-14 2021-10-22 广州汽车集团股份有限公司 Cfd仿真模型自动设置方法、系统、计算机设备及存储介质
CN113673178A (zh) * 2021-07-16 2021-11-19 北京宇航系统工程研究所 一种基于cfd仿真的注气式蓄压器动特性获取方法
CN113701985A (zh) * 2021-08-16 2021-11-26 武汉一冶钢结构有限责任公司 一种风洞收缩段肋板曲面测量方法
CN113820095A (zh) * 2021-08-23 2021-12-21 北京强度环境研究所 捆绑火箭风洞缩比模型组件
CN113849933A (zh) * 2021-09-27 2021-12-28 陕西柴油机重工有限公司 一种大功率船用柴油机润滑系统一维仿真方法
CN114970183A (zh) * 2022-06-08 2022-08-30 一汽解放汽车有限公司 风速仿真的评估方法、装置、计算机设备、存储介质
CN115859706A (zh) * 2022-11-01 2023-03-28 宝钢工程技术集团有限公司 一种基于宝数云的连铸设备有限元仿真流程封装工具
CN115994410A (zh) * 2023-03-22 2023-04-21 中国人民解放军国防科技大学 基于八叉树细化四面体网格的飞行器仿真驱动设计方法
CN116227295A (zh) * 2023-03-10 2023-06-06 宁夏宝丰昱能科技有限公司 电池包有限元建模方法、装置、计算机设备及存储介质
CN116776648A (zh) * 2023-08-22 2023-09-19 中汽研(天津)汽车工程研究院有限公司 车型热管理系统仿真模型标定和结构优化方法
CN116822024A (zh) * 2023-06-28 2023-09-29 西南交通大学 一种铁路桥上多线列车最不利交会位置的确定方法

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104866648B (zh) * 2015-04-21 2018-10-30 中车青岛四方机车车辆股份有限公司 一种高速行驶交通工具仿真模型前处理模板及处理方法
CN105243224A (zh) * 2015-10-28 2016-01-13 唐山轨道客车有限责任公司 客室内流场的参数化仿真方法、装置及列车
CN110795877B (zh) * 2019-10-23 2021-09-07 北京航空航天大学 一种基于流固耦合的受电弓作动电机扭矩补偿量计算方法
CN111008473A (zh) * 2019-12-03 2020-04-14 南方电网科学研究院有限责任公司 一种电力设备的仿真分析方法、装置及存储介质
CN113268845A (zh) * 2020-02-14 2021-08-17 广州汽车集团股份有限公司 一种汽车空气动力学仿真方法、优化仿真方法及装置
CN112380701A (zh) * 2020-11-16 2021-02-19 摩登汽车(盐城)有限公司 仿真加速度波形的对标方法

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102222121A (zh) * 2010-04-13 2011-10-19 延锋伟世通汽车饰件系统有限公司 用于模态仿真的有限元建模方法
CN102622467A (zh) * 2012-02-17 2012-08-01 上海同岩土木工程科技有限公司 包含多种构件的三维有限元网格自动剖分方法
CN102880759A (zh) * 2012-09-27 2013-01-16 西南交通大学 高速列车大系统耦合动力学仿真平台
CN103870623A (zh) * 2012-12-18 2014-06-18 南车青岛四方机车车辆股份有限公司 一种车辆模型前处理模板
CN104866648A (zh) * 2015-04-21 2015-08-26 南车青岛四方机车车辆股份有限公司 一种高速行驶交通工具仿真模型前处理模板及处理方法

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
GB0507618D0 (en) * 2005-04-15 2005-05-25 Lms Internat Nv Method and system for dynamic analysis of complex systems

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102222121A (zh) * 2010-04-13 2011-10-19 延锋伟世通汽车饰件系统有限公司 用于模态仿真的有限元建模方法
CN102622467A (zh) * 2012-02-17 2012-08-01 上海同岩土木工程科技有限公司 包含多种构件的三维有限元网格自动剖分方法
CN102880759A (zh) * 2012-09-27 2013-01-16 西南交通大学 高速列车大系统耦合动力学仿真平台
CN103870623A (zh) * 2012-12-18 2014-06-18 南车青岛四方机车车辆股份有限公司 一种车辆模型前处理模板
CN104866648A (zh) * 2015-04-21 2015-08-26 南车青岛四方机车车辆股份有限公司 一种高速行驶交通工具仿真模型前处理模板及处理方法

Cited By (43)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107748822A (zh) * 2017-10-31 2018-03-02 华晨汽车集团控股有限公司 一种发动机附件仿真分析中的配重方法
CN107748822B (zh) * 2017-10-31 2023-10-31 华晨汽车集团控股有限公司 一种发动机附件仿真分析中的配重方法
CN108763806A (zh) * 2018-06-07 2018-11-06 谢红太 高速列车过特长双线隧道时列车及隧道表面所受静压的数值模拟分析方法
CN109214139A (zh) * 2018-11-16 2019-01-15 中车株洲电力机车有限公司 一种受电弓强度仿真计算快速建模方法
CN110197041B (zh) * 2019-06-10 2023-07-21 大连交通大学 一种多物理场工况下多目标表面细观结构优化方法及系统
CN110197041A (zh) * 2019-06-10 2019-09-03 大连交通大学 一种多物理场工况下多目标表面细观结构优化方法及系统
CN110309575B (zh) * 2019-06-26 2023-06-02 武汉蓝恩汽车技术有限公司 商用车前下防护装置断裂失效评估方法及系统
CN110309575A (zh) * 2019-06-26 2019-10-08 武汉蓝恩汽车技术有限公司 商用车前下防护装置断裂失效评估方法及系统
CN110489835A (zh) * 2019-08-05 2019-11-22 安徽建筑大学 基于Ansys软件的自然通风与墙体蓄热耦合模拟方法
CN110489835B (zh) * 2019-08-05 2022-10-28 安徽建筑大学 基于Ansys软件的自然通风与墙体蓄热耦合模拟方法
CN111241627A (zh) * 2020-01-03 2020-06-05 同济大学 一种计入胶层的异质车身门盖件模态数值模拟及优化方法
CN111241627B (zh) * 2020-01-03 2024-04-23 同济大学 一种计入胶层的异质车身门盖件模态数值模拟及优化方法
CN113536701A (zh) * 2020-04-14 2021-10-22 广州汽车集团股份有限公司 Cfd仿真模型自动设置方法、系统、计算机设备及存储介质
CN111753372A (zh) * 2020-06-11 2020-10-09 雷沃工程机械集团有限公司 一种挖掘机整机热平衡仿真方法及系统
CN111753372B (zh) * 2020-06-11 2023-09-15 雷沃重工集团有限公司 一种挖掘机整机热平衡仿真方法及系统
CN112257307B (zh) * 2020-09-29 2023-12-08 北京科技大学 一种水下履带装备行走阻力的仿真计算方法及系统
CN112257307A (zh) * 2020-09-29 2021-01-22 北京科技大学 一种水下履带装备行走阻力的仿真计算方法及系统
CN112199907A (zh) * 2020-10-15 2021-01-08 乌江渡发电厂 一种基于cfd的水电站地下厂房多指标通风效果的分析方法
CN112199907B (zh) * 2020-10-15 2022-12-09 乌江渡发电厂 一种基于cfd的水电站地下厂房多指标通风效果的分析方法
CN112597680A (zh) * 2020-12-21 2021-04-02 北京慧物科联科技有限公司 一种应用于卫浴洁具的自动化有限元建模方法
CN112597680B (zh) * 2020-12-21 2023-11-24 北京慧物科联科技有限公司 一种应用于卫浴洁具的自动化有限元建模方法
CN113076597A (zh) * 2021-04-13 2021-07-06 吉林大学 一种面向cfd仿真的重卡车身快速建模方法
CN113343351A (zh) * 2021-05-18 2021-09-03 重庆长安汽车股份有限公司 一种汽车制动盘热-结构耦合分析方法
CN113505544A (zh) * 2021-06-18 2021-10-15 清华大学 基于有限体积法自行车运动虚拟数值风洞系统
CN113312728B (zh) * 2021-06-24 2024-06-11 南京航空航天大学 一种双列圆锥滚子轴承失油过程中的流场仿真方法及系统
CN113312728A (zh) * 2021-06-24 2021-08-27 南京航空航天大学 一种双列圆锥滚子轴承失油过程中的流场仿真方法及系统
CN113673178B (zh) * 2021-07-16 2024-05-03 北京宇航系统工程研究所 一种基于cfd仿真的注气式蓄压器动特性获取方法
CN113673178A (zh) * 2021-07-16 2021-11-19 北京宇航系统工程研究所 一种基于cfd仿真的注气式蓄压器动特性获取方法
CN113701985A (zh) * 2021-08-16 2021-11-26 武汉一冶钢结构有限责任公司 一种风洞收缩段肋板曲面测量方法
CN113701985B (zh) * 2021-08-16 2023-08-29 武汉一冶钢结构有限责任公司 一种风洞收缩段肋板曲面测量方法
CN113820095B (zh) * 2021-08-23 2023-05-23 北京强度环境研究所 捆绑火箭风洞缩比模型组件
CN113820095A (zh) * 2021-08-23 2021-12-21 北京强度环境研究所 捆绑火箭风洞缩比模型组件
CN113849933A (zh) * 2021-09-27 2021-12-28 陕西柴油机重工有限公司 一种大功率船用柴油机润滑系统一维仿真方法
CN113849933B (zh) * 2021-09-27 2024-04-05 陕西柴油机重工有限公司 一种大功率船用柴油机润滑系统一维仿真方法
CN114970183B (zh) * 2022-06-08 2024-06-04 一汽解放汽车有限公司 风速仿真的评估方法、装置、计算机设备、存储介质
CN114970183A (zh) * 2022-06-08 2022-08-30 一汽解放汽车有限公司 风速仿真的评估方法、装置、计算机设备、存储介质
CN115859706A (zh) * 2022-11-01 2023-03-28 宝钢工程技术集团有限公司 一种基于宝数云的连铸设备有限元仿真流程封装工具
CN116227295B (zh) * 2023-03-10 2023-11-28 宁夏宝丰昱能科技有限公司 电池包有限元建模方法、装置、计算机设备及存储介质
CN116227295A (zh) * 2023-03-10 2023-06-06 宁夏宝丰昱能科技有限公司 电池包有限元建模方法、装置、计算机设备及存储介质
CN115994410A (zh) * 2023-03-22 2023-04-21 中国人民解放军国防科技大学 基于八叉树细化四面体网格的飞行器仿真驱动设计方法
CN116822024A (zh) * 2023-06-28 2023-09-29 西南交通大学 一种铁路桥上多线列车最不利交会位置的确定方法
CN116776648B (zh) * 2023-08-22 2023-10-20 中汽研(天津)汽车工程研究院有限公司 车型热管理系统仿真模型标定和结构优化方法
CN116776648A (zh) * 2023-08-22 2023-09-19 中汽研(天津)汽车工程研究院有限公司 车型热管理系统仿真模型标定和结构优化方法

Also Published As

Publication number Publication date
CN104866648A (zh) 2015-08-26
CN104866648B (zh) 2018-10-30

Similar Documents

Publication Publication Date Title
WO2016169249A1 (zh) 一种高速行驶交通工具仿真模型前处理模板及处理方法
Wang et al. The effect of the ground condition on high-speed train slipstream
Chen et al. Dynamic analysis of the effect of nose length on train aerodynamic performance
Liu et al. An alternative algorithm of tunnel piston effect by replacing three-dimensional model with two-dimensional model
Diedrichs On computational fluid dynamics modelling of crosswind effects for high-speed rolling stock
Mohebbi et al. Analysis of the effects of lateral wind on a high speed train on a double routed railway track with porous shelters
Paz et al. Numerical methodology for evaluating the effect of sleepers in the underbody flow of a high-speed train
Liang et al. Numerical simulation of pressure transients caused by high-speed train passage through a railway station
Wang et al. Effect of moving ground on the aerodynamics of a generic automotive model: The DrivAer-Estate
Fu et al. Numerical analysis of the slipstream development around a high-speed train in a double-track tunnel
Fu et al. Turbulence models and model closure coefficients sensitivity of NASCAR Racecar RANS CFD aerodynamic predictions
CN108763806B (zh) 高速列车过特长双线隧道时列车及隧道表面所受静压的数值模拟分析方法
Ekman et al. Accuracy and speed for scale-resolving simulations of the DrivAer reference model
Wang et al. The impact of rails on high-speed train slipstream and wake
Li et al. Influence of blockage ratio on slipstreams in a high-speed railway tunnel
Zhou et al. Numerical investigation of the evolution of aerodynamic behaviour when a high-speed train accelerates under crosswind conditions
CN101906758A (zh) 高速铁路中声屏障脉动力大小的确定方法及应用
Nayani et al. Numerical study of the high-speed leg of a wind tunnel
Wang et al. Impact of the train heights on the aerodynamic behaviour of a high-speed train
Ramlan et al. Comparison between Solidworks and Ansys CFX flow simulation on aerodynamic studies
Shaharuddin et al. Numerical study for flow over a realistic generic model, DrivAer, using URANS
Mario et al. Numerical and experimental analysis of drag force in medium speed train design
Ji et al. Aerodynamic Shape Design of Pantograph Network Monitoring Device on High-Speed Trains
Lun et al. Numerical modelling of train aerodynamics in confined space
Paradot et al. A comparison of the numerical simulation and experimental investigation of the flow around a high speed train

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 15889732

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 15889732

Country of ref document: EP

Kind code of ref document: A1