KR20180115631A - Optimum design method for fluid system - Google Patents

Abstract

The present invention relates to an optimum design method for a fluid system. The optimum design method comprises: a step of setting a design area including the inlet and the outlet of fluid so as to design a flow path through the flow of fluid; a step of discretizing the design area; a step of setting a boundary condition corresponding to the operating condition of the design area; and a step of performing a phase optimal design. As a result, it is possible to obtain a flow path which satisfies the objective function and the constraint condition set by a user in a given design area without an initial shape, and to prevent flow loss and increase efficiency in an operating environment. Also, it has the effect of shortening the design time of a basic design stage in a development process.

Description

{Optimum design method for fluid system}

The present invention relates to a method of optimizing a fluid system, and more particularly, to a fluid system optimization method capable of obtaining a flow path satisfying an objective function and constraint conditions set by a user in a given design area without an initial shape, The present invention relates to a method for optimizing a fluid system.

The fluid flow path generally consists of a straight pipe and a bent pipe. It is known that the curved part has a complicated flow structure and therefore, about 50 to 60% of the total pressure loss of the flow path occurs in the bending part. Therefore, it is very important to design the bending section to minimize the flow loss of the components in order to increase the efficiency of the fluid system in designing the flow path of the fluid. In the past, the design of the channel was made based on the experience and intuition of the designer, and the design of the channel was improved through trial and error. In this case, it takes a long time from design to application and it is difficult to get out of the initial design form. Therefore, in order to improve these problems drastically, it is necessary to design a high-efficiency fluid system that can improve the efficiency in the operating environment by improving the flow loss in the concept designing stage.

In the prior art, the fluid system design for reducing the pressure loss of the fluid has been largely made as shown in Figs. 1 and 2. Fig. 1 shows a guide vane provided in the bending portion of the flow path, and Fig. 2 shows a modification of the shape of the inner wall and the outer wall of the flow path. As shown in FIG. 1, the flow path with the guide vane reduces the large pressure gradient generated in the bending portion but increases the weight of the blade to increase the stress due to the centrifugal force. As shown in FIG. 2, when the inner wall and outer wall shape of the flow path are modified, the pressure gradient inside the flow path and the occurrence of the secondary flow are reduced through the change of the cross-sectional area of the flow path. However, this is limited because it is designed by optimizing design variables or finite design variables through the experience and intuition of designers.

In order to solve such problems, shape optimization is recently used. The shape optimization is based on mathematical methods to determine the design variables that satisfy the constraints with the objective function maximized or minimized for a given design parameter and boundary condition. Design method. However, in the case of shape optimization, the initial shape of the shape does not change much during the design process because the shape is optimized based on the basic shape of the given channel.

Korean Patent Application Publication No. 10-2005-0007941 Korea Patent Office Registration No. 10-1072597

Therefore, it is an object of the present invention to provide a fluid system that can obtain a flow path that satisfies an objective function and a constraint set by a user in a given design area without an initial shape, And to provide a design method.

It also provides a method for optimizing the fluid system that can shorten the design time in the basic design stage in the development process.

The above-mentioned object is achieved by a method of designing a fluid, comprising: setting a design area including an inlet and an outlet of a fluid so as to design a flow path through a flow of fluid; Discretizing the design area; Setting a boundary condition corresponding to an operating condition of the design area; And performing a topology optimization of the fluid system.

Here, the step of setting the design region sets a two-dimensional or three-dimensional region in which a channel shape is not defined as a design region, and the step of discretizing the design region includes: Is assumed to be a porous medium and it is desirable to discretize the design area into finite number of elements.

Wherein performing the phase optimal design comprises: performing a flow analysis for a given boundary condition; Interpreting the sensitivity of each element to an objective function; And performing an iterative calculation until the objective function meets a convergence condition while changing a design parameter of each element according to the sensitivity, wherein the design parameter is a transmittance, and based on the sensitivity, It is preferable to increase the transmittance of the design regions contributing to the flow loss by adjusting the transmittance and to solidify the region where the flow rate is decreased due to the increase of the flow rate due to the increased transmittance,

Further, after performing the phase optimal design, deriving and extracting a final design phase; Reconstructing the design phase surface; Wherein the step of extracting and extracting a final design phase comprises extracting a final design phase as point data and reconstructing the design phase surface comprises: It is preferable to uniformly reconstruct the point data of the state.

In the step of verifying the design phase through computational analysis, it is preferable to perform computational analysis using a shape converted to CAD (computer aided design).

According to the structure of the present invention described above, it is possible to obtain a flow path that satisfies the objective function and the constraint condition set by the user in a given design area without an initial shape, and can improve fluidity The system optimum design method can be applied.

Also, it has the effect of shortening the design time in the basic design stage in the development process.

1 and 2 are views showing a conventional method of optimizing a fluid system,
Figures 3 and 4 are flow charts of the fluid system optimization design method,
5 is a view showing a design area for the optimum design of the internal flow path bending part,
FIG. 6 is a diagram illustrating a final phase extraction and a CAD conversion process derived after the phase optimal design,
FIG. 7 is a final shape of the internal passage bending portion derived through the phase optimum design,
8 is a diagram showing a flow domain setting for performing computational analysis on the derived shape,
9 is a view showing a pressure distribution in an intermediate plane in the optimized bending portion,
10 is a view showing a distribution of voltage and a velocity field in an intermediate plane,
11 is a graph showing the pressure distribution in the bending portion.

BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, a method of optimizing a fluid system according to an embodiment of the present invention will be described in detail with reference to the drawings.

The present invention relates to a method of optimizing a fluid system using a phase optimal design and more specifically to a method of optimizing a fluid system by assuming a fluid system as a porous material and formulating a finite element model, and a method of designing a fluid system using topology optimization. According to the present invention, it is possible to derive a flow path of any fluid system, such as gas or liquid, which can minimize pressure loss within a given design area. The optimal design method of the fluid system is as follows.

As shown in FIGS. 3 and 4, first, a design area is set (1).

A design space including the inlet and outlet of the fluid is set so that the flow path can be designed through the fluid flow. The inlet and outlet of the fluid can be set according to the shape of the flow path, and it is set appropriately according to the flow path and the flow path of the fluid. At this time, the setting of the design area can set the inlet and the outlet of the fluid from the box-shaped space in which a separate channel shape is not defined. FIG. 5 is a view showing a design region for the optimum design of the inner flow path bending portion, and this design region can be designed to optimize both two-dimensional or three-dimensional regions.

In the conventional case, the design was performed while adjusting the parameters by dividing the elements from the shape of the flow path prepared for the optimum flow path design. In the case of designing the flow path while controlling the variables as described above, it is necessary to repeat the computation analysis by controlling the variables in the parameterization process several tens to several hundred times, which causes the design work time to take more than a week, .

However, unlike the conventional art, when the basic flow path shape is not present and only the space for installing the fluid system is provided, an optimized flow path shape can be obtained without any parameterization.

Disseminate the design domain (2).

And discretization of the design area set through the first step. This is an operation for elementizing a design domain. The present invention assumes a given design domain as a porous medium by using a phase optimal design method, and discretizes a design domain into finite elements. In this way, the objective function becomes the maximum or the minimum, and at the same time, the individual element becomes the design variable so as to satisfy the constraints. The design variables correspond to the transmittance of each element, and the smaller the size of these elements, the smoother the final derived shape. At this time, the shape of the element is also called a mesh, and it is possible to apply both the alignment grid and the non-alignment grid.

Set boundary conditions (3).

The boundary condition corresponding to the operating condition of the selected fluid design area is set. Here, the boundary condition means conditions such as velocity and pressure at the interface such as the inlet and outlet of the fluid, and these conditions are additionally set.

Phase optimal design is performed (4).

The phase optimal design consists of performing the flow analysis for a given boundary condition, analyzing the sensitivity of each element to the objective function, and changing the design parameters of each element according to the sensitivity, when the objective function satisfies the convergence condition And then performing the iterative calculation up to the time t. Here, the design parameter means the transmittance.

The topology optimization consists of objective functions and constraints that are to be improved for a given fluid system. The objective function in the present invention is to minimize the inlet and outlet pressure drop of the fluid system, that is to minimize the flow loss, and the constraint means a constant that can not be changed in the design.

Also, the search for the direction of the design variables to find the optimal phase design is called the sensitivity analysis. The objective function of the fluid system assumed as a porous material in the present invention corresponds to minimizing the pressure difference between the inlet and the outlet of the system. Entering the boundary condition in a given fluid system, an optimum shape minimizing the pressure drop can be obtained. The phase optimal design of the fluid system can be formulated as follows to derive a phase that minimizes inlet and outlet pressure drops using a given design area: Where J is the objective function, R is the state equation, u is the velocity, p is the pressure, α is the design variable (transmittance), v is the kinematic viscosity and D is the shear strain.

<Formula 1>

The present invention can be applied to incompressibility and compressibility by using the steady-state Navier-Stokes governing equation as a porous medium as shown in Equation (2).

<Formula 2>

When the permeability (α) is 0, the flow is not in resistance and the momentum equation is expressed by the steady-state Navier-Stokes equation. As the permeability increases, the fluid in the porous medium becomes resistive, and when the permeability is sufficiently large, the velocity of the fluid approaches zero and the region is recognized as a solid. That is, the optimum shape is derived by adjusting the transmittance of each element in the design region. Numerical analysis using the finite volume method is also performed for the flow analysis. The finite volume method is a method of approximating the phenomenon by dividing the target region into finite number of elements. In this step, the continuous adjoint method is used to calculate the sensitivity of the objective function. The Lagrangian function ( L ) for each element is shown in Equation 3.

<Formula 3>

The objective function ( J ) is the pressure difference at the interface of a given fluid system, as shown in equation (4).

<Formula 4>

To optimize the Lagrangian function ( L ), the conditions described in Eq. 5 must be satisfied.

&Lt; EMI ID =

The sensitivity of each element to the objective function is calculated by Equation (6).

&Lt; EMI ID =

The design phase of the fluid system is derived by repeating the step of changing the transmittance of each element as shown in Equation 7 using the steepest descent method through the sensitivity calculated by Equation 6. Where n is the number of repetitions.

Equation (7)

Using this equation, the sensitivity of the objective function of each design area is calculated, and the transmittance of each element is adjusted in the design area based on the calculated sensitivity. This increases the transmittance of the design areas contributing to the flow loss, and the flow rate increases due to the increased flow load due to the increased transmittance. Thereby solidifying a region where the fluid can not pass.

The final design phase is derived and extracted (5, 6).

By varying the transmittance through the equation of four steps, the iterative calculation is performed until the objective function satisfies the convergence condition, and the optimum design phase satisfying the convergence condition is derived. The final design phase is then extracted as point data.

Reconstruct the surface of the final design phase (7).

Since the surfaces of the point data extracted through steps 5 and 6 are in an uneven state, the surface of the final design phase is reconstructed to uniformly form the surface of the point data. This is the step of smoothing the surface for conversion to CAD (computer aided design).

It is verified by computer analysis (8).

The performance of the final design phase is verified by computerized analysis using the CAD converted shape. The flow path obtained through the CAD conversion can be confirmed from FIG. FIG. 6 shows the final phase extraction and CAD conversion process derived after the fluid system topology optimization. The point data is extracted from the final phase, and the surface is triangulated and CAD conversion is performed. The internal flow path bending portion finally obtained through the above steps can be seen from FIG. 7, and the point where the pressure is lowered when the fluid moves is removed through solidification to obtain a flow path shape minimizing the flow loss . 8 is a diagram showing a flow domain setting for performing computational analysis on the derived shape.

FIG. 9 shows the pressure distribution in the intermediate plane in the optimized bending section as a result of the computer analysis, and FIG. 10 shows the distribution of the voltage and the velocity field in the intermediate plane in the optimized bending section. The bend section designed through the present invention reduces fluid cross-sectional area prior to fluid entry into the bend section, thereby increasing flow momentum and thereby relieving pressure gradient within the bend section. In the bending part, the pressure gradient is partially relaxed due to the increase or decrease of the cross-sectional area of the passage part, and the strength of the secondary flow is decreased. Since the cross-sectional area of the passage increases gradually after passing through the bending part, the size of the region where flow seperation occurs is reduced. In addition, it can be seen that the phenomenon of flow reattachment disappears as the flow follows the shape of the inner wall well.

11 is a graph showing the pressure distribution in the bending portion. The pressure distribution inside the flow path obtained through the computer analysis is shown. When the inlet pressure is the same, the pressure loss of the bending portion designed by the present invention is remarkably reduced.

Conventionally, a shape has to be present because the shape is optimized based on the basic shape of a given channel using the shape optimum design, and the initial shape of the shape has not changed much during the design process. However, according to the present invention, it is possible to obtain a flow path that satisfies an objective function and constraint conditions set by a user in a given design area without an initial shape by using a phase optimal design, Effect can be obtained.

Claims (8)

In a fluid system optimal design method,
Setting a design area including an inlet and an outlet of the fluid so as to design the flow path through the flow of the fluid;
Discretizing the design area;
Setting a boundary condition corresponding to an operating condition of the design area;
And performing a topology optimization of the fluid system. The method according to claim 1,
The step of setting the design area includes:
Dimensional region or a three-dimensional region in which a separate flow path shape is not defined is set as a design region. The method according to claim 1,
The step of discretizing the design area comprises:
Wherein the design area is assumed to be a porous medium using a topology optimization method and the design area is discretized into finite elements. The method according to claim 1,
Wherein performing the phase optimal design comprises:
Performing a flow analysis for a given boundary condition;
Interpreting the sensitivity of each element to an objective function;
And performing the iterative calculation until the objective function meets the convergence condition while changing the design parameter of each element according to the sensitivity. 5. The method of claim 4,
The design parameter is the transmittance,
The permeability of each element in the design region is adjusted based on the sensitivity to increase the transmittance of the design regions contributing to the flow loss and the region where the flow can not pass due to the increased flow rate due to the increased flow rate, Is solidified. &Lt; / RTI &gt; The method according to claim 1,
After performing the phase optimal design,
Deriving and extracting a final design phase;
Reconstructing the design phase surface;
And verifying the design phase through a computational analysis. The method according to claim 6,
The step of deriving and extracting the final design phase comprises:
The final design phase is extracted as point data,
Wherein reconstructing the design phase surface comprises:
Wherein the point data in the non-uniform state is uniformly reconstructed. The method according to claim 6,
Wherein the step of verifying the design phase through computational analysis comprises:
Wherein a computer aided design (CAD) shape is used to perform computational analysis.

Images