KR102298706B1 - Fluid analysis devices, methods, and computer programs capable of reducing memory usage - Google Patents

Abstract

Disclosed is a fluid analysis method capable of reducing memory usage during fluid analysis performed by a computer. generating a plurality of grid elements for a fluid and a structure through which the fluid flows; selecting a target grid element included in the analysis area from among a plurality of grid elements; Allocating a memory space for a plurality of grid elements and analyzing the flow using the memory space.

Description

Fluid analysis device, method and computer program that can reduce memory usage {FLUID ANALYSIS DEVICES, METHODS, AND COMPUTER PROGRAMS CAPABLE OF REDUCING MEMORY USAGE}

The present invention relates to a fluid analysis apparatus, method, and computer program that can reduce the use of memory during fluid analysis, and more particularly, to a fluid analysis device capable of reducing the memory usage required for fluid analysis using a grid element, It relates to methods and computer programs.

In general, to analyze the flow phenomenon, researchers use FDM (Finite Difference Method, Finite Difference Method), which divides the object to be analyzed into a mesh shape and allocates 3D memory to each grid. That is, the spatial domain is discretized into a very small volume mesh or grid, and a motion equation is established for each grid and calculated using an appropriate numerical algorithm.

At this time, in order to analyze the flow phenomenon of the molten metal casting having a complex shape, a large amount of calculation grid is required compared to other analysis, and a large amount of main memory used for calculation is also required. As a result, the number of grids that can be analyzed may be limited depending on the size of the main memory of the computer for analysis, and the more complex the casting, the more computation time may be required to analyze the flow phenomenon.

Therefore, it is possible to analyze the complex shape of the casting site, and a memory optimization study is required to reduce the loss by minimizing the use of memory.

In particular, when the flow field and temperature analysis are not combined, it is necessary to develop a fluid analysis method that can be applied by different handling of the grid corresponding to the molten metal and the grid corresponding to the mold among the grid elements.

An embodiment of the present invention is to provide a fluid analysis apparatus, method, and computer program capable of reducing the use of memory by different handling of grid elements for each object of fluid analysis such as a pressure field or heat flow.

An embodiment of the present invention is to provide a fluid analysis apparatus, method, and computer program capable of reducing the memory allocated to unnecessary grid elements during fluid analysis.

An embodiment of the present invention is to provide a fluid analysis apparatus, method, and computer program capable of reducing fluid analysis time during fluid analysis.

According to one aspect of the present invention, there is provided a fluid analysis method capable of reducing memory usage during fluid analysis performed by a computer, comprising: generating a plurality of grid elements for a fluid and a structure in which the fluid flows; selecting a target grid element from among the plurality of grid elements; A fluid analysis method capable of reducing memory usage is provided, comprising allocating a memory space to the plurality of grid elements and analyzing a flow using the memory space.

In this case, the fluid may be a molten metal in a molten state, and the structure may be a mold.

In this case, the fluid analysis may be a velocity field or a pressure field analysis.

In this case, the selecting of the target grid element may include selecting a first grid element corresponding to the fluid.

In this case, the selecting of the target grid element may further include selecting a second grid element in contact with the first grid element from among the structures.

In this case, allocating a memory space to the plurality of grid elements may include: assigning individual indexes to the target grid elements; It may include assigning the same index to a third grid element other than the target grid element, and matching the flow variable to the target grid element with reference to the individual index.

In this case, the method may further include matching a constant to the third grid element to which the same index is assigned.

In this case, the flow variable may include at least one of a speed, a pressure, and a filling degree.

In this case, the method may further include analyzing the heat flow of the fluid, and the analyzing of the heat flow of the fluid may include allocating a memory space to the first, second, and third grid elements.

According to another aspect of the present invention, there is provided a fluid analysis apparatus for analyzing a fluid using the above-described fluid analysis method, comprising: a lattice generating unit generating a plurality of lattice elements with respect to the fluid and a structure in which the fluid flows; a grid selection unit for selecting a target grid element from among the plurality of grid elements; A fluid analysis apparatus capable of reducing memory usage is provided, including a memory allocator for allocating a memory space for the plurality of grid elements and a flow analysis unit for analyzing a flow using the memory space.

According to another aspect of the present invention, there is provided a computer program that can reduce the use of memory, stored in a medium in order to execute the above-described fluid analysis method using a computer.

According to an embodiment of the present invention, memory usage can be reduced by classifying the grid elements and performing the analysis except for the grid elements that do not have an effect on the fluid analysis.

According to an embodiment of the present invention, time required for fluid analysis can be saved by reducing memory usage, and fluid analysis for a more complex structure can be smoothly performed.

1 is a block diagram of a fluid analysis apparatus according to an embodiment of the present invention.
2 is a perspective view showing a fluid and a structure in which a grid element is generated by the fluid analysis apparatus according to an embodiment of the present invention.
FIG. 3 is a front view of FIG. 2 ;
4 is a flowchart illustrating a fluid analysis method according to an embodiment of the present invention.
5 is a flowchart illustrating a step of selecting a grid element in a fluid analysis method according to an embodiment of the present invention.
6 is a flowchart illustrating a step of allocating a memory space in a fluid analysis method according to an embodiment of the present invention.
7 is a view showing that the apparatus using the fluid analysis method according to an embodiment of the present invention generates a grid element for the entire area.
8 is a diagram illustrating that the apparatus using the fluid analysis method according to an embodiment of the present invention assigns individual indexes to target grid elements.
9 is an example of a memory reference table showing the individual indexes shown in FIG. 8 and the corresponding integrated indexes.
FIG. 10 is a diagram showing that a device in which an error due to cross-reference occurs during fluid analysis assigns individual indexes to target grid elements.
FIG. 11 is an example of a memory reference table showing individual indexes shown in FIG. 10 and integrated indexes corresponding thereto.

Hereinafter, with reference to the accompanying drawings, the embodiments of the present invention will be described in detail so that those of ordinary skill in the art to which the present invention pertains can easily implement them. The present invention may be embodied in many different forms and is not limited to the embodiments described herein. In order to clearly explain the present invention in the drawings, parts irrelevant to the description are omitted, and the same reference numerals are assigned to the same or similar components throughout the specification.

In addition, terms such as “…unit”, “…group”, and “…module” described in the specification mean a unit that processes at least one function or operation, which may be implemented by hardware or software or a combination of hardware and software. can Here, the hardware may include a data processing device including a CPU or other processor. In addition, software driven by hardware may refer to a running process, an object, an executable file, a thread of execution, a program, and the like.

Fluid analysis apparatus, method, and computer program (hereinafter 'fluid analysis apparatus, method and computer program') capable of reducing memory usage according to an embodiment of the present invention can reduce memory usage during fluid analysis related to computational fluid dynamics. It is a fluid analysis apparatus, method and computer program that can be used.

In detail, it is an apparatus and method for minimizing the memory allocated to grid elements that do not affect the analysis during fluid analysis performed by a computer based on the grid model.

At this time, the meaning of analyzing the fluid may mean that in a fluid flowing through a structure such as a pipe or a mold, an expected behavior or distribution of physical properties of the fluid is simulated in advance through a computer.

1 is a block diagram of a fluid analysis apparatus according to an embodiment of the present invention.

Referring to FIG. 1 , a fluid analysis apparatus 100 according to an embodiment of the present invention includes an input unit 150 , a grid generation unit 110 , a grid selection unit 120 , a memory allocation unit 130 , and a flow analysis unit. (140). Although the drawing shows that the input unit 150 is a separate component from the fluid analysis apparatus 100 , the input unit 150 may be included in the fluid analysis apparatus 100 .

In an embodiment of the present invention, in order for the fluid analysis apparatus 100 to perform a fluid analysis, data related to a fluid analysis to be analyzed first may be input through the input unit 150 . Here, the data may include model information, variable variables, and the like.

In this case, the model information is information about the flow prediction model.

In detail, the fluid analysis apparatus 100 according to an embodiment of the present invention may predict a flow that meets the conditions of an actual fluid flow system by using one or more flow prediction models. Here, the model information constitutes a flow prediction model and may include at least one or more equations for obtaining a solution related to an analysis according to a flow variable.

Meanwhile, the flow variable is information constituting the fluid flow system, and may include a type of fluid, an initial inflow velocity of the fluid, initial physical properties of the fluid, a shape of a structure or a coefficient of friction, and the like. As an example, the flow variable may include at least one of a speed, a pressure, and a degree of filling.

The user may input all or part of the information to be inputted by inputting all of such information into the input unit 150 or by retrieving previously inputted information from the storage unit (not shown).

2 is a perspective view showing a fluid and a structure in which a grid element is generated by the fluid analysis apparatus according to an embodiment of the present invention. 3 is a front view of FIG. 2 ; 4 is a flowchart illustrating a fluid analysis method according to an embodiment of the present invention. 5 is a flowchart illustrating a step of selecting a grid element in a fluid analysis method according to an embodiment of the present invention. 6 is a flowchart illustrating a step of allocating a memory space in a fluid analysis method according to an embodiment of the present invention. 7 is a view showing that the fluid analysis apparatus according to an embodiment of the present invention generates a grid element for the entire area.

In one embodiment of the present invention, referring to FIG. 3 , the grid generating unit 110 may generate a plurality of grid elements that separate a fluid and a structure, which are objects of fluid analysis, into a predetermined grid unit. (S10) ) As described above, in an embodiment of the present invention, fluid analysis is performed by a flow prediction model, and a grid element becomes a basic unit of this flow prediction model and can interact with other grid elements.

Referring to FIGS. 2 and 3 , the grid generating unit 110 may divide the fluid 10 , the structure 30 , and all objects in the region 50 excluding them into grid elements of a predetermined size.

In this case, the lattice element may be formed in three dimensions, and may be formed of a polyhedron having a regular hexagonal shape in which all sides have the same length. However, the shape of the grid element is not limited thereto. However, in this specification, a one-dimensional square lattice element will be assumed as shown in FIG. 3 to avoid repeated description and to help understanding.

Also, referring to FIG. 7 , the grid generating unit 110 may assign an integrated index to each of the generated grid elements for identification. In this way, the integrated index is given to the grid elements to match the flow variables corresponding to each grid element.

As an example, the integrated index may be numbered so that the value sequentially increases while moving from left to right as shown in the figure. However, the integrated index assignment method is not limited thereto, and may be assigned with various regularities.

3 and 5 , in an embodiment of the present invention, the grid selector 120 may select the target grid elements 10 and 32 from among the plurality of grid elements generated as described above. (S20)

In this case, the target grid elements 10 and 32 mean grid elements that have a great influence on fluid analysis. The fluid analysis apparatus 100 according to an embodiment of the present invention distinguishes the target grid elements 10 and 32 that have a great influence on the fluid analysis and other grid elements, that is, the third grid elements 34 and 50, The object of the invention can be achieved by reducing the memory allocated to the third grid elements 34 and 50 .

For example, when analyzing the pressure field or velocity field of the molten metal in the casting process, most 34 except for the interface among the grid elements included in the mold do not move, and have no effect on the pressure and velocity of the fluid. Therefore, efficiency can be promoted by excluding it from the analysis. This efficiency improvement will be particularly effective in fluid analysis of incompressible fluids.

However, in contrast to this, in the case of thermal fluid analysis, not only the molten metal but also all lattice elements corresponding to the mold must be considered in the analysis, so it is impossible to exclude the lattice elements 34 of the mold or the lattice elements that can be excluded will be very limited.

Therefore, the fluid analysis apparatus 100 according to an embodiment of the present invention can be effectively applied to a case in which the structure has no effect on the flow, such as a pressure field or a velocity field analysis of a fluid.

In an embodiment of the present invention, referring to FIGS. 2 and 5 , the grid selector 120 may include a grid element corresponding to a fluid as a target grid element, that is, the first grid element 10 ( S21 ). ) If the casting process, the grid element included in the molten metal may be the first grid element (10). Since the fluid analysis is for examining the flow pattern of the fluid, it will be natural that the first grid element 10 is included as the target grid element.

The grid selector 120 may further include a second grid element 32 in contact with the first grid element 10 among a plurality of grid elements constituting the structure 30 as a target grid element. (S22)

The second grid element 32 is selected as the target grid element in order to prevent cross-reference errors that may occur when the flow analysis unit 140 analyzes the flow ( S40 ), which will be described later.

In more detail, when calculating the velocity value or velocity gradient of the first lattice element 10 when calculating the velocity field of fluid analysis, information on adjacent lattice elements may be required. At this time, as shown in FIG. 10 , when only the first grid element 10 corresponding to the fluid is designated as the target grid element and an individual index is given, in the case of the outermost first grid element 10 (for example, FIG. 10 ) In the grid element number 22 or 31), cross-referencing errors may occur because adjacent grid elements include the same index (-1 in FIG. 10) without discrimination. This will be described in detail in the section describing the memory allocator 130 .

In order to prevent such an error, another conditional statement such as an if statement is required at the time of coding, and in this case, the purpose of the present invention for reducing the amount of calculation may be somewhat tarnished.

8 is a diagram illustrating that the apparatus using the fluid analysis method according to an embodiment of the present invention assigns individual indexes to target grid elements. 9 is an example of a memory reference table showing the individual indexes shown in FIG. 8 and the corresponding integrated indexes. FIG. 10 is a diagram showing that a device in which an error due to cross-reference occurs during fluid analysis assigns individual indexes to target grid elements. FIG. 11 is an example of a memory reference table showing individual indexes shown in FIG. 10 and integrated indexes corresponding thereto.

In one embodiment of the present invention, referring to FIG. 6 , the memory allocator 130 may allocate a memory space to a plurality of grid elements. (S30) Here, allocating a memory space means that each grid element follows. It may mean pre-allocating a memory space to be used when performing fluid analysis (S40).

In this case, the memory allocator 130 may assign individual indexes to the target grid elements 10 and 32 in order to effectively allocate a memory space for each target grid element 10 and 32 . (S31)

Referring to FIG. 8 , in the drawing, an integrated index is displayed above the grid elements that have been hatched (first grid element) and shaded (second grid element), and individual indexes are displayed below.

In this way, the memory allocator 130 may assign individual indexes to the first and second grid elements 10 and 32 . For example, as shown, individual indexes may be assigned to sequentially increase index numbers while moving from left to right in the same manner as the integrated index. In addition, indexes may be assigned so that no individual indexes are identical to each other.

The memory allocator 130 may assign the same index to the third grid element 34 other than the target grid elements 10 and 32. (S32) As described above, the third grid element 34 is used for fluid analysis. It has no significant effect and does not need to be specifically identified. Only the fact that the corresponding grid element is the third grid element 34 may be identified. For example, in FIG. 8 , the same index of '-1' is assigned to all of the third grid elements 34 .

In an embodiment of the present invention, the memory allocator 130 may match the floating variables corresponding to the individual first and second grid elements 10 and 32 to the allocated memory space with reference to the individual index. ( S33)

In this case, the memory allocator 130 may call the floating variable from the storage unit (not shown). Here, the storage unit may receive and store the flow variables for all grid elements to which the integrated index is given from the input unit 110 in advance.

In detail, as described above, in the case of thermal fluid analysis, flow parameters for all grid elements 10, 32, 34 and 50 included in the mold are required. Therefore, the storage unit may contain information on the entire grid element to be applicable to all fluid analysis.

At this time, the fluid analysis apparatus 100 according to an embodiment of the present invention performs thermal fluid analysis including all grid elements, and then selectively targets grid elements 10 and 32 only in the case of pressure field or velocity field analysis. It is also possible to reduce memory usage by performing the analysis by applying .

Referring to FIGS. 8 and 9 , the target grid element has both an integrated index and an individual index, and in this case, the individual index and the integrated index correspond to each other one-to-one. Therefore, the individual index can function as an address to retrieve the fluid variable for each grid element from the storage unit.

As such, the flow variable matched to the memory space may be used as an input variable when performing fluid analysis (S40), which will be described later. In this case, the memory allocator 130 may generate a memory reference table as shown in FIG. 9 in order to easily call the flow variable of the target grid element corresponding thereto from all grid elements to which the integrated index is assigned.

In more detail, since the integrated index of the grid element to which the individual index '6' is assigned is '21', the memory allocator 130 may call the flow variable of the grid element to which the integrated index '21' is assigned.

In addition, the memory allocator 130 may require a flow variable of each grid element and adjacent grid elements in some cases. In the same way, floating variables can be called using the memory reference table.

As an example, referring again to FIGS. 8 and 9 , the process of calling the flow variable on the right side of the grid element to which the individual index '6' is assigned will be described as follows. The integrated index corresponding to the individual index '6' is '21', the right side of the integrated index '21' is '22', and the individual index corresponding to the integrated index '22' is '7', so the individual index '7' The flow variable of this given grid element is called. That is, in this process, cross-referencing occurs at least three times.

Therefore, when only the first grid element 10 is selected as the target grid element to be analyzed, an error due to cross-reference may occur.

For example, referring to FIGS. 10 and 11 , when the flow variable on the right side of the grid element to which the individual index '3' is assigned is called, the individual index '3' corresponds to the integrated index '22', and the integrated index '22' The right side of ' is '23', and the individual index corresponding to '23' becomes -1 (example of the same index value). In this case, since the same index has no distinguishing power, the memory allocator 130 may generate a cross-referencing error.

Therefore, it is preferable in terms of calculation efficiency to include not only the first grid element 10 corresponding to the fluid as the target grid element, but also the second grid element 32 in contact with the fluid.

Meanwhile, in the case of the memory space of the third grid element 34 to which the same index is assigned, the memory allocator 130 may match an appropriate constant. (S34)

For example, in the case of velocity field analysis, all of the constants stored in the memory space of the third grid element 34 may have a value of '0' indicating a stationary state.

The flow analysis unit 140 of the fluid analysis apparatus 100 according to an embodiment of the present invention may analyze the flow using a memory space allocated for each grid element.

Through the process discussed above, it is possible to convert nonlinear partial differential equations (e.g., Navier-Stokes Equations) into algebraic equations by discretizing the fluid analysis target into lattice elements. That is, each lattice element unit The fluid flow problem can be analyzed by using the algorithm of numerical methods, at this time, the flow variables in the memory allocated for numerical analysis for each grid element will be used, and calculations for each grid element are performed. In the process, the allocated memory space for big data to be processed will be used.

Referring back to FIG. 1 , the processed result may be expressed in a form that can be confirmed by the user through the output unit 160 . If necessary, an image rendering technique may be applied to visually express the flow of a fluid.

As described above, in the fluid analysis apparatus and method according to an embodiment of the present invention, memory usage can be reduced by allocating memory to a part of a grid element corresponding to a structure and preventing unnecessary calculations from occurring.

As a result, there is an advantage that can significantly reduce the time required for fluid analysis involving complex calculations. In particular, the inventor of the present invention was able to obtain the effect of reducing the memory usage by 60% compared to usual in the fluid analysis based on a one-dimensional grid element through the fluid analysis apparatus and method according to an embodiment of the present invention. If this is applied to a 3D grid element, the memory usage reduction effect will be even greater.

Although one embodiment of the present invention has been described above, the spirit of the present invention is not limited to the embodiments presented herein, and those skilled in the art who understand the spirit of the present invention can add components within the scope of the same spirit. , changes, deletions, additions, etc. may easily suggest other embodiments, but this will also fall within the scope of the present invention.

10 First grid element 30 Target grid element
32 Second Grid Element 34 Third Grid Element
50 Structure surrounding environment 100 Fluid analysis device

Claims (11)

As a fluid analysis method that can reduce memory usage in fluid analysis performed by a computer,
generating a plurality of grid elements for a fluid and a structure in which the fluid flows;
selecting a target grid element from among the plurality of grid elements;
allocating a memory space for the plurality of grid elements; and
Analyzing the flow using the memory space,
The target grid element,
a first grid element corresponding to the fluid and a second grid element in contact with the first grid element among the structures;
Allocating a memory space for the plurality of grid elements comprises:
A fluid analysis method capable of reducing memory usage by matching flow variables only with respect to the target grid element, and matching constants with respect to a third grid element excluding the target grid element among the plurality of grid elements. The method of claim 1,
The fluid is a molten metal in a molten state, and the structure is a mold, a fluid analysis method that can reduce memory usage. The method of claim 1,
The fluid analysis is a velocity field or pressure field analysis, a fluid analysis method that can reduce memory usage. delete delete The method of claim 1,
Allocating a memory space for the plurality of grid elements comprises:
assigning an integrated index to all of the plurality of grid elements;
assigning individual indexes to the first grid element and the second grid element among all the plurality of grid elements;
assigning the same index to the third grid element; and
Comprising the step of matching the flow variable to the target grid element using the integrated index and the individual index, the fluid analysis method that can reduce the use of memory. 7. The method of claim 6,
A fluid analysis method capable of reducing memory usage, further comprising the step of matching a constant to the third grid element to which the same index is assigned. 7. The method of claim 6,
The flow variable includes at least one of speed, pressure, and a degree of filling, a fluid analysis method capable of reducing memory usage. The method of claim 1,
Further comprising the step of analyzing the heat flow of the fluid,
Analyzing the heat flow of the fluid comprises:
A fluid analysis method capable of reducing memory usage, comprising allocating a memory space for the first, second and third grid elements. A fluid analysis device for analyzing a fluid using the fluid analysis method according to any one of claims 1 to 3 and 6 to 9, comprising:
a grid generating unit generating a plurality of grid elements with respect to the fluid and the structure in which the fluid flows;
a grid selection unit for selecting a target grid element from among the plurality of grid elements;
a memory allocator for allocating a memory space for the plurality of grid elements; and
A fluid analysis device capable of reducing memory usage, including a flow analysis unit that analyzes the flow using the memory space. A computer program capable of reducing memory usage, stored in a medium for executing the method according to any one of claims 1 to 3 and 6 to 9 using a computer.

Images