JP5277129B2 - Fluid analysis device - Google Patents

Description

  The present invention relates to a numerical simulation analysis of a fluid using a personal computer or a large computer, and more particularly to a fluid analysis apparatus that executes a numerical simulation of a flow around a moving object having an arbitrary boundary shape placed in an analysis space.

  With the development of computers and numerical analysis methods, in-cylinder phenomena in reciprocating engines (hereinafter referred to as engines) such as gasoline engines using numerical simulation of flow and combustion, and engine development are being carried out.

  The engine is provided with an intake valve for controlling the amount of air sucked from the outside, an exhaust valve for controlling the amount of gas discharged to the outside, and a piston for taking out work. During operation of the engine, These repeatedly move and stop. For this reason, the internal space shape of the engine changes every moment. When analyzing the interior of the engine by numerical simulation, it is necessary to reproduce the change in the spatial shape accompanying the movement of the piston and valve by some method.

  As one of the methods for reproducing such a change in the space shape, a boundary condition embedding method (hereinafter referred to as IB method) has been devised. In the fluid numerical simulation, the analysis space is divided into a plurality of polyhedrons (hereinafter, calculation cells) for analysis. The IB method uses surface data of a moving object (for example, an intake valve, an exhaust valve, and a piston) separately from a calculation cell, and expresses a change in shape of the analysis space by moving the surface data in the analysis space. It is. The change in the flow due to the moving object is reflected by adding an interaction force term to the equation of motion of the fluid. Specifically, the flow analysis is performed by solving equation (1).

Here, ρ is the density of the fluid, p is the pressure, τ is the shear stress tensor, vec (u) is the average velocity vector, and vec (F _s ) is the interaction force vector. The average velocity vector vec (u) and the interaction force term vec (F _s ) are evaluated by the equations (2) and (3).

Here, θ is the volume fraction of the object occupying the calculation cell (hereinafter referred to as object volume fraction), vec (u _F ) is the fluid velocity vector, vec (u _P ) is the velocity vector of the object, and Δt is the simulation condition. And is determined so as to realize the stability of the analysis and the required calculation accuracy. The volume fraction θ of the object is defined by Equation (4), where V _k is the volume of each calculation cell and V _M is the volume of the object occupying the calculation cell.

  The object volume fraction defined by Expression (4) is calculated based on data regarding a given moving object (hereinafter referred to as moving object data) and data regarding a calculation cell (hereinafter referred to as calculation grid data).

  The IB method is advantageous in that it can reduce the man-hours for creating the calculation cell, but it has a disadvantage in that the volumetric load calculation based on the calculation grid data and the moving body data is heavy. As a means for solving this disadvantage, there is a method of calculating an interaction force term based on information of virtual particles arranged based on an object shape (see, for example, Non-Patent Document 1).

S.V.Apte, “A multilevel formulation to simulate particulate flows”, CTR Annual Research Report 2004, pp201-208 (2004)

  According to the literature, a method of calculating the volume fraction of an object in each calculation cell based on the virtual particle data used to distribute the multiple virtual particles inside the moving object and approximate the shape of the moving object. Has been. According to this method, it is not necessary to perform volume calculation based on geometric calculation, and the calculation load can be reduced. However, in this method, the volume fraction of the object is 1 in the lattice completely filled with the moving object, where the volume fraction is not 0 in the calculation cell where the moving object does not exist near the boundary surface of the moving object. It may not be possible. This is considered to be caused by the fact that the accurate position information of the boundary of the object is lost because the calculation is based only on the virtual particle data used for shape approximation. As a result, the fluid simulation error is increased.

  The present invention solves such problems of the prior art, and its purpose is to predict the volume fraction of an object in a calculation cell located near the surface of a moving object with high accuracy, thereby analyzing time. It is intended to provide a fluid analysis apparatus that realizes improvement in fluid analysis accuracy that prevents a decrease in accuracy and a prolonged calculation time.

  In order to achieve the above object, in the present invention, lattice point data represented by position coordinates of a plurality of lattice points scattered in an analysis space, a plurality of cell surface data and a plurality of cells represented by using a plurality of lattice point data Represented by calculation grid data composed of calculation cell data represented by surface data, multiple object point data provided on the object surface to represent an object placed in the analysis space, and multiple object point data Means for storing object data composed of object minute surface data are provided.

In the fluid analysis device for analyzing the flow around the object,
Means for storing virtual particle data composed of volume data determined based on position coordinates, object volume and lattice data of a plurality of virtual particles distributed inside the object placed in the analysis space;
Grid point determination means for determining whether the grid point is inside or outside the moving object based on the object point data and the object micro-plane data;
The lattice point of the volume of the object, which is the ratio of the volume of the object included in each calculation cell and the volume of each calculation cell, determined to be inside the object by the lattice point determination means in the cell on the boundary surface A calculation based on volume data of virtual particles belonging to a cell on the boundary surface, which is a calculation cell composed of both an internal point that is and an external point that is a lattice point determined to be outside the object,
A fluid analysis device characterized by the above is created.

  According to the present invention, a plurality of calculation cells provided in the analysis space are calculated cells (hereinafter referred to as internal cells) in which 100% is occupied by moving objects, and calculation cells (hereinafter referred to as internal cells) in which the surfaces of moving objects exist , Cells on the boundary surface), and calculation cells that do not include moving objects inside the cell (hereinafter referred to as external cells), the object volume fraction of the internal cell is 1 and the object volume fraction of the external cell is 0. Can be set. In addition, the object volume fraction of the cell on the boundary surface can be calculated based on the position and volume data of virtual particles distributed inside the moving object and near the surface of the moving object. Can be improved. As a result, it is possible to achieve both prevention of fluid analysis accuracy reduction and prevention of prolonged calculation time.

1 is an overall schematic configuration diagram of a fluid analysis device in an embodiment of the present invention. It is a functional block diagram of the process performed with a fluid analysis apparatus. It is a flowchart of the process performed with a fluid analysis apparatus. It is a flowchart of the process in a flow and an object movement analysis means. It is a functional block diagram of the process performed by a moving object calculation means. It is a flowchart of the process in a moving object calculation means. It is a flowchart of the process in a lattice point division means. It is a flowchart of the process in a calculation cell classification means. It is a flowchart of the process in a virtual particle classification means. It is a flowchart of the process in a volume fraction and interaction force calculation means. It is the figure which extracted a part of expression of the moving object which used the object micro surface. It is the figure which showed the structure of moving object data. It is the figure which took out one of the calculation grids created with the cube. It is the figure which showed the structure of calculation grid data. It is a figure which shows the virtual particle distributed in the two-dimensional analysis space. It is the figure which showed the structure of virtual particle data. It is a figure which shows the classification | category of a lattice point and a calculation cell. It is a figure explaining a virtual particle generation processing method. It is a figure of the input screen of virtual particle generation conditions. It is a display screen of the result analyzed in the embodiment of the present invention.

  An embodiment of the present invention will be described with reference to FIGS. The configuration of the fluid analysis apparatus, data to be used, processing means and processing method, and data input / output will be described in this order.

  FIG. 1 is a configuration diagram of a fluid analysis apparatus 1000. The fluid analysis apparatus 1000 includes an input device 1001 such as a keyboard and a mouse, an external storage medium 1002 such as a hard disk or a compact disk in which grid data used for fluid analysis, initial conditions, boundary conditions, moving object data, simulation results, and the like are recorded. It comprises a display device 1003 such as a display for displaying results and settings necessary for executing the simulation, and an analysis device 1004 for executing the simulation. The analysis device 1004 includes a main storage device 1005 for storing data input from the input device 1001, data input to and output from the external storage medium 1002, a program executed by the central processing unit 1006, and simulation processing. The central processing unit 1006 executes a program that describes The fluid analysis apparatus is configured as described above.

  Next, data used in the fluid analysis device will be described. The data will be described in the order of moving object data representing a moving object, calculation grid data generated by dividing the analysis space, and virtual particle data used for calculating the object volume fraction, which is the ratio of the object to each calculation cell.

Moving object data will be described with reference to FIGS. The moving object data represents the position and moving speed of the object surface, and needs to correspond to a moving object having a complicated shape. For this reason, the moving object is represented by a set of a plurality of polygons. Hereinafter, the polygon used to represent the object surface is referred to as an object minute surface 4201. FIG. 11 shows a part of the surface 4200 of the moving object. The object surface 4200 is composed of a large number of object minute surfaces 4201. Further, the object minute surface 4201 is composed of three object points 4202. In the fluid analysis apparatus 1000, these are expressed by a combination of numerical data. The object point data 4203 representing the object point 4202 and the object minute surface data 4204 representing the object minute surface 4201 will be described with reference to FIG. The object point data 4203 includes an object point number NB _g for distinguishing each object point, the coordinates of the object point, and the moving speed of the object point. In this embodiment, the total number of object points is represented as NB _gm . The coordinates are data indicating the position in the analysis space, and the moving speed of the object point is data indicating the moving speed of each object point in the analysis space. The object minute surface data 4204 includes an object surface number NB _s for distinguishing each object minute surface 4201 and an object point number of the object point 4202 constituting the object minute surface 4201. In this embodiment, the total number of object micro surfaces is represented as NB _sm . The object surface number is data given as an integer. In addition, although it demonstrated using the triangle in FIG. 11, the shape of a micro surface is not restricted to this, It can also be made into polygons, such as a rectangle and an octagon. Further, the format of the moving object data is not limited to the format shown in FIG.

Subsequently, the calculation grid data will be described with reference to FIGS. 13 and 14. As described above, the space is divided into a plurality of polyhedrons when fluid analysis is performed by simulation. This polyhedron is called a calculation cell. FIG. 13 shows a calculation cell 4100 created as a cube as an example. The calculation cell 4100 includes six cell surfaces 4101, and each cell surface 4101 includes four lattice points 4102. In the fluid analysis apparatus 1000, these are expressed by a combination of numerical data. The lattice point data 4103 representing the lattice point 4102, the cell surface data 4104 representing the cell surface 4101, and the cell data 4105 representing the calculation cell 4100 will be described with reference to FIG. The lattice point data 4103 includes a lattice point number N _g for distinguishing each lattice point, the coordinates of the lattice point, and an object internal flag. In this embodiment, the total number of lattice points is N _gm . The coordinates of the lattice points are data indicating the position of each lattice point in the analysis space. The object internal flag will be described with reference to FIG. FIG. 17 shows a calculation cell 4100 and a moving object surface 4200 in a two-dimensional analysis space. From the relationship with the surface 4200 of the moving object, there are two types of lattice points 4102: lattice points outside the moving object (hereinafter referred to as external points) 4112 and lattice points (hereinafter referred to as internal points) 4111 inside the moving object. Can be classified. Inside the object flag if the grid point is internal point is data indicating whether the external point, set in this embodiment, if the grid point is internal point T _rue, and F _Alse if external point Is done. Cell surface data 4104 representing the cell surface 4101 is composed of a cell surface numbers N _s for distinguishing each cell surfaces, in the lattice point number N _g constituting the cell surface. In this embodiment, the total number of cell surfaces is represented as N _sm . The calculation cell data 4105 representing the calculation cell 4100 includes a cell number _Nc for distinguishing each calculation cell, a cell surface number constituting the calculation cell, a cell volume, a cell position flag, and a volume of an object occupying each cell. Consists of rates. In this embodiment, the total number of cells is expressed as N _cm . The cell position flag will be described with reference to FIG. FIG. 17 shows the calculation cell 4100 and the surface 4200 of the moving object in the two-dimensional analysis space as described above. The calculation cell 4100 has a calculation cell (hereinafter referred to as an external cell) 4107 in which the entire calculation cell is located outside the object because of the relationship with the object surface 4200, and a calculation cell (hereinafter referred to as an internal cell) in which the entire calculation cell is included inside the object. 4108, a calculation cell (cell on the boundary surface) 4106 including an object boundary inside the calculation cell can be classified. The cell position flag is data indicating which of three types of calculation cells are classified. In this embodiment, the calculation cell is 0 if the calculation cell is an internal cell, 1 if the calculation cell is an external cell, or a cell on the boundary surface. If there is, 2 is set. In addition, although it demonstrated using the cubic cell in FIG. 13, the shape of a calculation cell is not restricted to this, It can also be made into polyhedrons, such as a tetrahedron and an octahedron. Further, the format of the calculation grid data is not limited to the format shown in FIG.

Next, virtual particle data used for volume calculation will be described with reference to FIG. FIG. 15 shows a calculation cell 4100, an object surface 4200, and virtual particles 4300 in a two-dimensional analysis space. As shown in FIG. 15, the virtual particles are characterized in that they are distributed inside a closed curved surface formed on the surface of the moving object (a closed curve if the analysis space is a two-dimensional space). The virtual particle is expressed by a combination of numerical data in the fluid analyzing apparatus 1000. The virtual particle data 4301 representing the virtual particle 4300 includes a virtual particle number N _p for distinguishing each virtual particle, a position coordinate of the virtual particle, a moving speed of the virtual particle, a volume of the virtual particle, a micro surface number to which the virtual particle belongs, Consists of belonging cell number and object minute surface flag. In this embodiment, the total number of virtual particles is represented as _Npm . The position coordinates and moving speed of the virtual particles are given by coordinate axes provided in the analysis space. The object minute surface number NB _s is stored in the assigned minute surface number. For example, it can be set as the number of the object minute surface 4201 that is the closest. As the belonging cell number, a calculation cell number _Nc to which the virtual particle belongs is set. The object minute surface flag indicates whether the virtual particle belongs to the cell 4106 on the boundary surface. When belonging to the interface on the cell 4106, T _rue is set, when belonging to other than the boundary surface on the cell 4106 F _Alse is set.

In this embodiment, the moving object is a rigid body. It is assumed that the moving speed of the moving object is vec (u _P ), and the moving speed of each object point and each virtual particle is also vec (u _P ).

  This is the end of the description of the data, followed by a description of the processing. Processing will be described with reference to FIG. 2 and FIG. 3, and then the details of each processing will be described with reference to flowcharts and block diagrams.

  FIG. 2 is a functional block diagram showing an analysis function executed by the fluid analysis apparatus 1000. The analysis function includes an input data storage unit 100 for storing data input from the input device 1001 and the external storage medium 1002, an analysis unit 200 for executing volume analysis of the object and fluid analysis in each calculation cell, and an analysis result. The result display means 300 for displaying on the display device 1003 is configured.

  The input data storage means 100 has initial conditions for physical quantities (velocity, temperature, chemical species concentration, pressure, turbulence, etc.) to be analyzed by the fluid analyzing apparatus 1000 and settings necessary for simulation (for example, calculation time, time step width, etc.) ), Fluid analysis condition storage means 101 for storing boundary conditions for the physical quantity equation, and calculation grid data storage means 102 for storing calculation grid data generated by dividing the analysis space for fluid analysis into a plurality of polyhedra. , Moving object data storage means 103 for storing moving object data and virtual particle data storage means 104 for storing virtual particle data. The analysis unit 200 includes a flow field analysis unit 201 that performs flow field analysis based on calculation of object movement, fluid analysis conditions, calculation grid data, object volume fraction, and the like, calculation grid data, object plane data, and virtual particles. It comprises moving object calculation means 202 for calculating the object volume fraction and interaction force based on the data, and virtual particle generation means 203 for generating virtual particles based on the virtual particle data.

  FIG. 3 is a flowchart showing processing by the analysis function shown in FIG. First, in step S1000, data necessary for analysis is input from the input device 1001 and the external storage medium 1002. Subsequently, the process proceeds to step S2000 to execute calculation of the object volume fraction and fluid analysis. Subsequently, the process proceeds to step S3000, and a process of displaying the analysis result on a display device 1003 such as a display is executed.

  Next, the process of step S2000 shown in FIG. 3 will be described with reference to FIG. FIG. 4 is a flowchart showing the process of the flow / object movement analysis means S2000. In step S2100, analysis is started. In step S2200, it is determined whether virtual particles used for volume calculation are arranged in the analysis space. If it is determined in step S2200 that the virtual particles have been arranged (YES), the process proceeds to step S2400. If it is determined in step S2200 that the virtual particles are not arranged (NO), the process proceeds to step S2300, and virtual particles are generated inside the moving object and in the vicinity of the surface of the moving object, and then the process proceeds to step S2400. In step S2400, fluid analysis is started based on the initial conditions. In step S2500, the calculation of the volume fraction and flow field of the object at time t is started. Subsequently, the process proceeds to step S2600, where the interaction force reflecting the volume fraction of the moving object occupying each calculation cell and the influence of the moving object is calculated, and the process proceeds to step S2700. In step S2700, fluid analysis is performed based on the volume fraction and interaction force of the object calculated in step S2600, the input calculation grid data, initial condition, boundary condition data, and the old time step calculation result. In this process, the moving object is also moved. In step S2800, it is determined whether to end the analysis based on the data input and held in the fluid analysis device storage unit 101. If it is determined in step S2800 that the analysis is to be terminated (YES), the flow proceeds to step S2990 and the flow / object movement calculation is terminated. If it is determined in step S2800 that the fluid analysis is not terminated (NO), the process proceeds to step S2900, and the analysis time t is then updated to a new time, and the process proceeds to step S2500. Here, the processing content of step S2300 is demonstrated using FIG. 18 and FIG. FIG. 18 shows a moving object 4200, a calculation cell 4100, and virtual particles 4300 placed in a two-dimensional analysis space, and FIG. 19 shows an input screen for virtual particle generation conditions. As shown in FIG. 18, the virtual particles 4300 are distributed inside a closed curved surface formed on the surface of the moving object 4200 (a closed curve if the analysis space is a two-dimensional space). The range Lp in which the virtual particles are distributed is determined, and the virtual particles 4300 are randomly distributed between the virtual surface 4400 determined inside the object and the surface of the moving object 4200 determined thereby. Here, the number of virtual particles to be generated and the range in which the virtual particles are generated are input using a text file stored in the input device 1001 or the external storage medium 1002. FIG. 19 shows an example of a condition input screen 4500 displayed on the display device 1003. At least a total number of particles to be generated and a numerical value (distance from the surface of the moving object) for designating the range in which the particles are generated are input. . The number of generated particles is a number that specifies the total number of virtual particles generated in the closed curved surface formed on the surface of the moving object 4200, and is specified as 0 when not generated. The larger the number of virtual particles, the higher the accuracy. However, increasing the number of virtual particles increases the calculation load, so it is determined by a balance between calculation accuracy and calculation time. The generation range of virtual particles is a value that specifies a range in which virtual particles are generated, and can be specified by a distance Lp from the surface of the moving object. The value of Lp needs to be in the distance order of lattice points. For example, an average value of distances between lattice points or a value that is a constant multiple of the average value may be determined as the value of Lp. Here, the particle generation processing has been described using a two-dimensional analysis space, but the same processing can be performed for a three-dimensional space.

  Next, the process of step S2600 shown in FIG. 4 will be described with reference to FIGS.

  FIG. 5 is a functional block diagram of the moving object calculation means shown in FIG. The moving object calculation unit 202 includes a lattice point sorting unit 2021, a calculation cell sorting unit 2022, a virtual particle sorting unit 2023, and an object volume / interaction force calculation unit 2024. The grid point sorting unit 2021 determines whether the grid point 4102 is an internal point based on the moving object data and the calculated grid data. The determination is made by a method such as a ray intersection method. The calculation cell determination unit 2022 is a unit that classifies each calculation cell 4100 into an internal cell 4108, an external cell 4107, and an on-boundary cell 4106 based on the object plane data and the result of the grid point classification unit 2021. The virtual particle sorting unit 2023 is a unit that searches for virtual particles belonging to the cell 4106 on the boundary surface based on the result of the grid point sorting unit 2023 and the virtual particle data. The determination is made from the virtual particle position and the position of the lattice point. The object volume / interaction force calculation means 2024 is a means for calculating an interaction force representing the volume fraction of the object occupying each calculation cell 4100 and the effect of the object on the fluid using the determination result of each process. is there.

  FIG. 6 is a flowchart showing processing in the moving object calculation means shown in FIG. First, in step S2610, it is determined whether each grid point 4102 is an internal point. In step S2620, the calculation cell 4100 is divided into an internal cell 4108, an external cell 4107, and a boundary surface cell 4106. Subsequently, in step S2640, a process for searching for virtual particles for volume calculation located in the boundary surface cell 4106 is performed. Next, it progresses to step S2660 and an object volume fraction and interaction force are calculated.

  Next, the processing from step S2610 to step S2640 shown in FIG. 6 will be described with reference to FIGS.

FIG. 7 is a flowchart showing the process of step S2610 shown in FIG. In step S2611, the grid point segmentation process is started, and then the process proceeds to step S2612, where the Mth (1 ≦ M ≦ N _gm ) grid point (hereinafter referred to as Mth grid point) segment is started. Then, it progresses to step S2613 and it is determined whether the M-th grid point is an internal point. The process of step S2613 is realized using a ray intersection method or the like. Subsequently, if it is determined in step S2613 that the object is outside the object (NO), the process proceeds to step S2614, the object internal flag of the _Mth lattice point is set to False, and the process proceeds to step S2616. If it is determined that the internal point (YES) at step S2613, sets an internal flag in the M-th grid point proceeds to step S2615 to T _rue, subsequently, the flow proceeds to step S2616. In step S2616, it is determined whether classification of all grid points has been completed. If it is determined in step S2616 that the process has been completed, the process advances to step S2618 to end the grid point division process. If it is determined in step S2616 that the classification of all grid points has not been completed, the process proceeds to step S2617, the value of M is updated (for example, M is replaced with the value of M + 1), and the process proceeds to step S2612.

FIG. 8 is a flowchart showing the process of step S2620 shown in FIG. First, it progresses to step S2621 and a calculation cell division process is started. Next, the process proceeds to step S2622, and determination of a cell having a cell number K (1 ≦ K ≦ N _cm ) (hereinafter referred to as “K” cell) is started. Then, it progresses to step S2623 and it is determined whether the Kth cell is the foreign cell 4107. FIG. It is possible to use an object internal flag of a lattice point constituting the cell. If all the object internal flags of the lattice points constituting the cell are _False , it is an external cell. If it is determined in step S2623 that the cell is an external cell (YES), the process proceeds to step S2625, the cell position flag of the Kth cell is set as an external cell, and then the process proceeds to step S2628. If it is determined in step S2623 that the cell is not an external cell (NO), the process advances to step S2624 to determine whether the Kth cell is an internal cell. For the determination, an object internal flag of a lattice point constituting the cell can be used. If all the object internal flags of the lattice points constituting the cell are _True , the cell is an internal cell. If it is determined in step 2624 that the cell is an internal cell (YES), the flow proceeds to step S2626, the cell position flag of the Kth cell is set as the internal cell, and the flow proceeds to step S2628. If it is determined in step S2624 that the cell is not an internal cell (NO), the process proceeds to step S2627, the cell position flag is set as a cell on the boundary surface, and then the process proceeds to step S2628. In step S2628, it is determined whether all calculation cells have been classified. If it is determined that classification of all grids has been completed (YES), the process proceeds to step S2630, and the calculation cell sorting process is terminated. If it is determined in step S2628 that all classifications are not completed (NO), the process proceeds to step S2629, the value of K is updated (for example, K is replaced with a value of K + 1), and the process proceeds to step S2622.

FIG. 9 is a flowchart showing the process of step S2640 shown in FIG. First, the process proceeds to step S2641 to start virtual particle classification. Next, the process proceeds to step S2642, and processing of virtual particles (hereinafter referred to as J-th virtual particles) whose virtual particle number is J-th (1 ≦ J ≦ N _pm ) is started. In step S2643, the cell number to which the J-th virtual particle belongs is searched based on the coordinates of the J-th virtual particle and the calculation grid data, and the calculated cell number is set based on the result. Then, it progresses to step S2644 and it is determined whether J number virtual particle is located in a cell on a boundary surface. For the determination, the cell position flag of the calculation cell having the number held as the belonging calculation cell number as the cell number can be used. If it is determined in step S2644 that the cell is not located on the boundary surface cell (NO), the flow proceeds to step S2646, the boundary surface flag of the virtual particle data 4301 is set to _False, and the flow proceeds to step S2648. If it is determined in step S2644 that the cell is located on the boundary surface cell (YES), the process proceeds to step S2645 to update N _pc which is a variable for storing the number of virtual particles in the boundary surface cell, and the process proceeds to step S2647. Set the boundary flag to _True . Then, it progresses to step S2648. In step S2648, it is determined whether classification of all virtual particles has been completed. If it is determined that classification of all virtual particles has been completed (YES), the process proceeds to step S2650, and the virtual particle search process is terminated. If it is determined in step S2648 that classification of all virtual particles has not been completed (NO), the process proceeds to step S2649, and the value of J is updated (for example, the value of J is replaced with J + 1, and the process proceeds to step S2642. .

FIG. 10 is a flowchart showing the process of step S2660 shown in FIG. First, the process proceeds to step S2661, where calculation of volume fraction and interaction force is started. Next, the process proceeds to step S2662, and calculation of a calculation cell (hereinafter referred to as I-th cell) whose cell number is I (1 ≦ I ≦ N _cm ) is started. Next, proceeding to step S2663, it is determined whether the number I cell is an internal cell. For this determination, the cell position flag of the cell data 4105 can be used. If it is determined in step S2663 that the cell is an internal cell (YES), the flow proceeds to step S2665, the object volume fraction is set to 1, and the object moving speed vec (u _P ) and the above equation (3) are used for mutual processing. The acting force is calculated, and the process proceeds to step S2668. If it is determined in step S2663 that the cell is not an internal cell (NO), the process proceeds to step S2664 to determine whether the cell number I is an external cell. If it is determined in step S2664 that the cell is an object external cell (YES), the process proceeds to step S2667, the object volume fraction and the interaction force are set to 0, and the process proceeds to step S2668. If it is determined in step S2664 that the object is not an external cell (NO), the process proceeds to step S2666, and the volume fraction and interaction force are calculated based on the virtual particle data. A specific example of the process in step S2666 will be described below. The total number of virtual particles belonging to the calculation cell divided into cells on the boundary surface is N _ST , the number of virtual particles contained in the I-th cell is N _PI , the total volume of moving objects is V _m , and the volume of the I-th cell is V _I , V _{IN is} the total volume of the calculation cells divided into internal cells, vec (u _Pj ) is the velocity of the j-th virtual particle, vec (u _P ) is the moving velocity of the moving object, and ρ is the fluid density . Based on this definition, the volume fraction θ _I and the interaction force vec (F _SI ) of the cell No. _I are obtained by equations (5) to (7).

  Here, Δt is a time interval in the simulation, and is determined according to the stability of calculation and the required calculation accuracy. The sum total of formula (6) is calculated for the particles belonging to the cell number I.

  In the formulas (5) and (6), the range affected by the virtual particles is limited to a point, but the range affected by the virtual particles can be expanded by using the distribution function, and the calculation accuracy is further increased. Improvement can be expected. For example, the calculation can be performed from Expression (8) to Expression (10).

Here, vec (x) and vec (x _p ) are vectors representing positions in the analysis space, vec (x _{p, j} ) is a vector representing positions of virtual particles in the analysis space, and σ is a range in which particles are distributed Is a real number given by a constant multiple of. Further, the sum of Expression (8) and Expression (9) is calculated for the particles belonging to the cell number I. The above is only an example, and any method can be selected for the calculation of the volume fraction in step S2666 without being limited to these methods.

  FIG. 20 is a display image 4600 of the volume fraction calculated using the fluid analysis apparatus according to the embodiment of the present invention. The calculation cell 4100, the moving object 4200, the virtual particle 4300, and each calculation cell in the two-dimensional analysis space 4700 are shown. The volume fraction calculation result in is shown. According to the present invention, the external cell 4107 can calculate the volume fraction of 0, the internal cell 4108 can calculate the volume fraction of 1, and the boundary cell 4106 can calculate the volume fraction of greater than 0 and less than 1. It can be displayed that the object volume ratio in 4107 is 0, the object volume ratio in the internal cell 4108 is 1, and the object volume ratio in the cell 4106 on the boundary surface is larger than 0 and smaller than 1. Further, the result of the virtual particles 4300 being distributed inside the surface of the moving object 4200 and in the vicinity of the surface of the moving object 4200 can be displayed on the screen.

100 Input data storage means 101 Fluid analysis condition storage means 102 Calculation grid data storage means 103 Moving object data storage means 104 Virtual particle data storage means 200 Flow / object movement analysis means 201 Flow field analysis means 202 Moving object calculation means 203 Virtual particle generation Means 300 Result display means 1000 Fluid analysis device 1001 Input device 1002 External storage medium 1003 Display device 1004 Analysis device 1005 Main storage device 1006 Central processing unit 2021 Grid point sorting means 2022 Calculation cell sorting means 2023 Virtual particle sorting means 2024 Volume fraction / Interaction force calculation means 4100 Calculation cell 4101 Cell surface 4103 Lattice point data 4104 Cell surface data 4105 Cell data 4106 Cell on boundary surface 4107 External cell 4108 Internal cell 4111 Internal point 4112 External Point 4200 Surface of moving object 4201 Object micro surface 4202 Object point 4203 Object point data 4204 Object micro surface data 4300 Virtual particle 4301 Virtual particle data 4400 Virtual particle distribution range 4500 Input screen image 4600 Output screen image

Claims (8)

A plurality of grid point data represented by the position coordinates of a plurality of grid points scattered in the analysis space;
A plurality of cell surface data represented using the plurality of grid point data;
Calculation grid data composed of calculation cell data represented by the plurality of cell surface data;
Object point data represented by position coordinates of a plurality of points scattered on the object surface to represent an object placed in the analysis space;
Means for storing object data composed of object microsurface data represented by the plurality of object point data,
In the fluid analysis device for analyzing the flow around the object,
Means for storing virtual particle data composed of volume data determined based on position coordinates, object volume and lattice data of a plurality of virtual particles distributed inside the object placed in the analysis space;
Grid point determination means for determining whether the grid point is inside or outside the moving object based on the object point data and the object micro-plane data;
The volume fraction of the object, which is the ratio of the volume of the object contained in each calculation cell and the volume of each calculation cell, is determined between the internal point that is a grid point determined by the grid point determination means and the outside of the object. In the cell on the boundary surface, which is a calculation cell configured using both external points that are determined to be lattice points, calculation based on volume data of virtual particles belonging to the cell on the boundary surface,
A fluid analysis device characterized by the above. The fluid analysis apparatus according to claim 1,
Based on the volume data V _c of one calculation cell classified as a cell on the boundary surface, the number of virtual particles N _pc belonging to one calculation cell classified as the cell on the boundary surface, and the volume V _p of the virtual particle, Obtaining an object volume fraction θ _c of the cell on the boundary surface;
A fluid analysis device characterized by the above. In the fluid analysis apparatus according to claim 1 and claim 2,
A calculation means for obtaining a sum V _sum C of the volume of the internal cell which is a calculation cell composed only of the internal point;
A calculation means for obtaining a sum N _{t of} virtual particles belonging to the calculation cell classified as the cell on the boundary surface;
Based on the volume V _{m of the} object, the V _sum C, and the N _t , the volume V _p of each virtual particle belonging to the calculation cell classified as the lattice on the boundary surface is obtained by the equation (A) V _p = V _sum C / Nt formula (A)
A fluid analysis device characterized by the above. The fluid analyzing apparatus according to claim 3,
The number of virtual particles N _pc belonging to one calculation cell classified as the boundary cell, the volume data V _c of one calculation cell classified as the boundary cell, and the volume V _p of each virtual particle Based on the equation (B), the object volume fraction θ _c of the cell on the boundary surface is calculated based on θ _c = N _pc * V _p / V _c (B)
A fluid analysis device characterized by the above. The fluid analysis device according to claim 1, wherein
The volume fraction in each calculation cell classified as an internal cell, which is a calculation cell composed only of the internal points, is calculated as 1, and the volume fraction of the object in each calculation cell classified as the external cell To 0,
A fluid analysis device characterized by the above. The fluid analysis apparatus according to claim 1,
The object moves in the analysis space at an arbitrary speed;
A fluid analysis device characterized by the above. The fluid analysis device according to claim 1,
Means for inputting and storing the number M _{p of} virtual particles generated inside the object;
Means for inputting a distance L _p from the object surface indicating a range in which the virtual particles are generated, and means for storing;
A fluid analyzing apparatus that generates virtual particles based on the M _p and the L _p . The fluid analyzing apparatus according to claim 1,
Comprising display means for displaying the object surface and the virtual particles generated inside the object;
A fluid analysis device characterized by the above.

Images