A clean room air outlet and orifice plate arrangement air flow organization optimization design method based on CFD

Abstract

This invention relates to the field of cleanroom airflow organization optimization design technology, and discloses a CFD-based cleanroom air outlet and perforated plate arrangement airflow organization optimization design method, including: preliminary optimization of FFU position; structural simplification of porous components; establishment of a three-dimensional physical model and mesh generation; input of actual parameters for calculation; and adjustment of perforated plate parameters and FFU equipment position when airflow does not meet design requirements. This CFD-based cleanroom air outlet and perforated plate arrangement airflow organization optimization design method reduces the difficulty of cleanroom airflow organization simulation operation by employing an FFU equipment arrangement method for the best cleanroom airflow organization under empty cleanroom conditions, an orifice plate parameter optimization method for large airflow deflection angles, an FFU equipment position optimization method to improve crosswinds and local vortices on the equipment sides, and a simplified modeling and parameter calculation method for porous components such as DCC and perforated plates. It provides a fast, efficient, and accurate technical path for obtaining excellent cleanroom airflow organization.

Description

Technical Field

[0001] This invention relates to the field of cleanroom airflow organization optimization design technology, specifically a CFD-based cleanroom air outlet and perforated plate arrangement airflow organization optimization design method. Background Technology

[0002] The production of semiconductor devices has very strict requirements for the production environment, involving hundreds of manufacturing processes. Due to the special nature of its production technology, the cleanroom environment will directly affect the yield rate of chip manufacturing processes.

[0003] A well-organized airflow system in a cleanroom can ensure the cleanliness level of the room and effectively prevent contamination and cross-contamination during the production process. However, the airflow organization of a room cannot be visually displayed during the design process, and designers cannot directly check whether the airflow organization is reasonable. Therefore, some inappropriate or even erroneous designs are inevitable. To address such issues, CFD software is often used during the design process to simulate and demonstrate the airflow organization.

[0004] However, cleanrooms have a large number of FFU devices, and the airflow organization effect varies greatly depending on the arrangement. The existing FFU arrangements in engineering projects mostly rely on experience and it is difficult to guarantee that they have a good airflow organization.

[0005] The complex porous structure of cleanrooms, such as DCC (dry coil), louvered air vents, and perforated plates in raised floors, reduces the efficiency of engineers in modeling and calculation.

[0006] When high-velocity crosswinds occur on both sides of cleanroom equipment, cross-contamination between nearby installations will occur, and eddies will weaken the cleanroom's ability to remove contaminants.

[0007] When the airflow in a cleanroom is tilted, the air age inside the cleanroom increases, creating dead zones in some areas. This airflow pattern is extremely unfavorable for the control of pollutants in the cleanroom.

[0008] Meanwhile, when airflow patterns that do not meet the requirements occur, it is difficult for engineers to accurately and quickly adjust the cleanroom structure. These problems place extremely high demands on engineers and limit the convenient, efficient and accurate application of CFD numerical simulation technology in engineering.

[0009] Therefore, a CFD-based method for optimizing the airflow organization of cleanroom air outlets and perforated plates is proposed to address the aforementioned problems. Summary of the Invention

[0010] (a) Technical problems to be solved

[0011] To address the shortcomings of existing technologies, this invention provides a CFD-based method for optimizing airflow organization in cleanroom vents and perforated plate arrangements, thus resolving many of the deficiencies in existing technologies.

[0012] (II) Technical Solution

[0013] The technical solution of this invention to solve the above-mentioned technical problems is as follows: A method for optimizing airflow organization based on CFD cleanroom air outlets and perforated plate arrangement, comprising:

[0014] Step 1: Determine the location of equipment in the cleanroom and the actual number of FFUs (Fan Filter Units) based on the actual preliminary design drawings;

[0015] Step 2: Determine the area that the FFU can affect based on the FFU deployment rate. When the FFU cannot affect the entire clean area, the location where the FFU can affect the largest area is taken as the FFU deployment location.

[0016] When an FFU can affect the entire clean area, the FFU placement location should be the area where the FFU has the greatest high-intensity impact.

[0017] Step 3: Based on the actual solution requirements, if the influence of DCC on the cleanroom air temperature is considered, simplify the complex duct network structure of DCC into a flat plate with the same actual size, and simplify the perforated raised floor and air outlet into a plane with the same actual size.

[0018] If the impact of DCC on cleanroom air temperature is not considered, the complex duct network structure, porous raised floor and air outlets of DCC are simplified to a plane with the same actual size.

[0019] Step 4: Based on the above model simplification, establish a three-dimensional physical model of the cleanroom according to the FFU floor plan, cleanroom structure diagram, raised floor layout diagram, etc.

[0020] Step 5: Import the established 3D model into the mesh generation software for mesh generation, and then use Fluent software to solve the required content and select the corresponding governing equations.

[0021] Determine the model boundary type based on the actual working conditions, input the corresponding parameters, and select a suitable solution algorithm for calculation;

[0022] Step 6: When the residual of the solved governing equation decreases to the required range or the monitored physical quantity remains unchanged or changes by an extremely small amount, the solution can be considered to have converged and the calculation ends.

[0023] Step 7: Import the calculation results into the post-processing software to obtain the cleanroom flow field distribution. When the cleanroom airflow deflection angle is large, adjust the cleanroom orifice plate arrangement to improve the cleanroom flow field. Repeat steps 5 to 7 until the airflow meets the cleanliness level requirements.

[0024] When the crosswind speed on the equipment surface is high and the vortex on the side of the equipment is large, adjust the position of the FFU to reduce the crosswind speed and the vortex area.

[0025] Repeat steps four through seven until the airflow meets the cleanliness requirements.

[0026] The beneficial effects of this invention are:

[0027] (1) This invention provides a preliminary optimization scheme for FFU, which is the best form of cleanroom airflow organization when the cleanroom is empty and the same FFU layout rate is achieved. This greatly facilitates engineers to quickly, efficiently and accurately arrange FFU equipment.

[0028] (2) This invention provides a simplified modeling method for porous structures in cleanrooms, which simplifies the modeling operations for DCC, air outlets, perforated plates, etc.

[0029] Based on the simplified model, a simplified calculation method for solving porous media or porous step models is presented, which reduces the operational difficulty of cleanroom airflow organization simulation and improves the efficiency of cleanroom modeling and calculation for engineering designers.

[0030] (3) This invention provides an airflow organization optimization method for crosswinds and local vortices on the surface of cleanroom equipment. After optimization, the crosswind speed and vortex area are significantly reduced.

[0031] To address the issue of airflow tilt in cleanrooms, based on the ratio K1 of the average air velocity at 0.4m below the raised floor and 0.5m from the return air duct to the average velocity at 0.1m above the raised floor, a method is proposed to adjust the airflow tilt by using the perforation ratio and arrangement of the perforated plate.

[0032] The above measures enable engineers to quickly, efficiently, and accurately apply CFD airflow organization simulation technology.

[0033] (4) This invention provides a simulation optimization process for cleanroom airflow organization, including corresponding steps such as modeling, solving, and optimization, which improves the efficiency of CFD numerical simulation for engineering design and construction personnel and makes cleanroom airflow organization simulation more systematic.

[0034] The implementation of this technology can shorten the construction period and ensure that the cleanroom has a better flow field, resulting in good socio-economic benefits.

[0035] Based on the above technical solution, the present invention can be further improved as follows.

[0036] Furthermore, in step two, the initial position optimization of the FFU specifically includes:

[0037] a1. Calculate the FFU deployment rate based on the number of FFUs and the area of ​​the clean area;

[0038] a2. The area affected by an FFU is its own area and four adjacent areas of the same area. The intensity of the impact of an FFU on an area without an FFU is divided into two levels: Level 1 is when there is only one FFU in front of, behind, to the left or right of the area, and Level 2 is when there are two FFUs in front of, behind, to the left or right of the area.

[0039] a3. When the FFU deployment rate is low (not greater than 40%) and the FFU cannot fully affect the clean area, select the FFU deployment location with the largest area of ​​Level 1 influence intensity.

[0040] a4. When the FFU deployment rate is high (greater than 40%), and the FFU can have a comprehensive impact on the clean area, select the FFU deployment location with the largest area of ​​the Level 2 impact intensity area while ensuring comprehensive impact.

[0041] Furthermore, in step three, the DCC, orifice plate, and air outlet are simplified. When performing heat coupling, the DCC can be selected as a porous medium model.

[0042] When thermal coupling is not performed, DCC can choose the porous step model;

[0043] For air vents and raised floors without directly loaded heat source components, they are simplified as planes of a porous step model.

[0044] Furthermore, after simplification using porous media or porous step structures, when adjusting the airflow resistance through the porous structure, for those without simplified components, adjusting the arrangement of the porous structure requires complex processes such as remodeling, meshing, and naming. This simplified solution eliminates the need to repeat the modeling, meshing, and naming processes; the resistance change can be completed simply by modifying the relevant coefficients in the boundary conditions.

[0045] Furthermore, it provides several simplified calculation methods for perforated plates;

[0046] When DCC needs to load a heat source term, it is simplified to a solid plate according to Dancy's law model. Based on the actual test results of the resistance and flow velocity relationship, the permeability α and the inertial drag coefficient C2 are obtained by fitting the following formula.

[0047]

[0048] Furthermore, when DCC is not loaded with a heat source term, it is simplified to a plane along with porous components such as orifice plates and air vents. According to the porous step condition, based on the relationship between resistance and flow velocity obtained from actual tests, the permeability α and inertial drag coefficient C2 are obtained by fitting according to the following formula.

[0049]

[0050] Furthermore, when fluid flows through a perforated plate, if Re > 30 and the orifice ratio does not exceed 50%, the plate's resistance characteristics are only related to the orifice ratio and are independent of the plate thickness, orifice diameter, and Re number. Under the condition of uniform orifice arrangement, the relationship between the orifice ratio n and the resistance coefficient ε is:

[0051] ε=25426×n -2.051 R 2 =0.9931

[0052] The magnitude of the local resistance ΔP is calculated based on the drag coefficient ε. j :

[0053]

[0054] Ignoring the friction resistance along the orifice plate, that is:

[0055] ΔP=ΔP j

[0056] According to the source term S of the total resistance and momentum equation i Relationship between them:

[0057] ΔP=S i *L

[0058] In the formula, L is the thickness of the orifice plate perpendicular to the airflow direction.

[0059] get:

[0060]

[0061] Since the viscous resistance of the fluid in the cleanroom is very small, the influence of the first-order velocity term can be ignored when α = 1, and the inertial drag coefficient C2 can be calculated through the local resistance.

[0062]

[0063] Furthermore, the relationship between the orifice plate resistance coefficient and the opening ratio was obtained through extensive experimental fitting and regression. Based on the regression formula, the velocity first-order coefficient was further optimized considering the characteristics of the cleanroom orifice plate.

[0064] The method was verified by simulation and experiment, and it can well reflect the pressure drop and flow velocity changes of gas flowing through the porous plate.

[0065] Furthermore, in step five, Fluent includes multiple turbulence models, among which the realizable k-ε model performs well for rotating flows, boundary layer flows with strong adverse pressure gradients, flow separation, and secondary flows.

[0066] Therefore, the turbulence model selected in this study is the realizable k-ε model;

[0067] The realizable k-ε turbulence model includes various wall functions. The Menter-Lechner wall function shows good adaptability to different Re ranges and different boundary layer thicknesses.

[0068] The finite volume method is used in the flow field solution. The simple algorithm is used for pressure and velocity coupling. The momentum and energy equations are based on the second-order upwind dispersion scheme, and the turbulent kinetic energy k and turbulent dissipation rate ε are based on the first-order upwind dispersion scheme.

[0069] The airflow control equations are Reynolds-averaged Navier-Stokes equations, including the continuity equation, momentum equation, and energy equation in tensor form:

[0070]

[0071]

[0072]

[0073] in, Represents Reynold stress. This represents turbulent heat flux.

[0074] Furthermore, in step seven, when the crosswind intensity on the upper surface of the equipment is high and the vortex area on the side of the equipment is large after the production process equipment is arranged in the clean room, the position of the FFU air outlet should be adjusted. The adjustment principle is to place the air outlet off the upper surface of the equipment and repeat steps four to seven until the requirements are met.

[0075] Furthermore, in step seven, the unidirectional flow cleanroom has high requirements for the cleanroom flow line. When the FFU arrangement rate exceeds 40% and the cleanroom flow line deflects and does not meet the requirements, the perforation rate of the cleanroom perforated plate is arranged according to the ratio K1 value of the average air velocity at 0.4m below the raised floor and 0.5m from the return air duct to the average velocity of the cross section at 0.1m above the upper surface of the raised floor.

[0076]

[0077] Among them, V x V represents the average air velocity at a point 0.4m below the raised floor of the cleanroom and 0.5m from the return air duct. b The average flow velocity is the value of the cross-section at the top surface of the elevated floor, which is 0.1m above the floor surface.

[0078] Through extensive experiments and simulations, I have found that streamline tilt is mainly affected by the ratio of airflow in the lower mezzanine to the velocity near the perforated plate in the cleanroom, and is almost unaffected by the FFU outlet velocity.

[0079] At the same time, this conclusion still holds true regardless of whether it is a single-sided return air duct or a double-sided return air duct, or what the lower mezzanine height is.

[0080] When the cleanroom is relatively short from the return air duct (distance from the nearest return air duct does not exceed 6m), the perforated plates can be laid out in two non-uniform sections; when the cleanroom is relatively long from the return air duct, the non-uniform perforations can be arranged in multiple sections; the relationship between the perforated plate arrangement and the K1 value is as follows:

[0081]

[0082]

[0083] Furthermore, when the cleanroom flow line is tilted, multi-directional simulations are conducted on different cleanroom structural forms, different lower mezzanine heights, different room depths, different FFU arrangement rates, and different FFU flow velocities. It is found that the relationship between the K1 value and the floor opening rate arrangement is still applicable. The simulation results are compared with the experimental results to obtain a quick optimization strategy for selecting the perforated plate arrangement based on the K1 value range, which greatly improves the efficiency of airflow organization optimization for engineering designers. Attached Figure Description

[0084] Figure 1 Flowchart for optimizing airflow organization in cleanrooms;

[0085] Figure 2 This is a schematic diagram showing the area and intensity of FFU influence.

[0086] Figure 3 FFU floor plan;

[0087] Figure 4 A simplified diagram of the orifice plate model;

[0088] Figure 5 Fitting a graph to the parameter data;

[0089] Figure 6 The diagram shows the flow field model before optimization in Case 1.

[0090] Figure 7 The optimized flow field model diagram for Case 1;

[0091] Figure 8 This is a schematic diagram of grid division;

[0092] Figure 9 To optimize the front flow field cross-sectional diagram;

[0093] Figure 10 To optimize the flow field cross-sectional diagram;

[0094] Figure 11 Here is the model structure diagram for Case 2;

[0095] Figure 12 To optimize the front flow field streamline diagram;

[0096] Figure 13 To optimize the flow field streamline diagram.

[0097] In the diagram: 1. Return air duct; 2. DCC; 3. Equipment; 4. FFU; 5. Orifice plate; 6. Flat plate; 7. Plane. Detailed Implementation

[0098] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0099] In the embodiments, Figure 1 A flowchart for the design methodology of cleanroom airflow organization optimization includes the following steps:

[0100] Step 1: Determine the location of equipment (such as dry coils, process production equipment, etc.) and the actual number of FFUs in the cleanroom based on the actual preliminary design drawings.

[0101] Step 2: Determine the area that an FFU can affect based on the FFU deployment rate. When an FFU cannot affect the entire clean area, the location with the largest affected area is the FFU placement location; when an FFU can affect the entire clean area, the location with the largest high-intensity affected area is the FFU placement location. This includes the following:

[0102] a1. Calculate the FFU deployment rate based on the number of FFUs and the area of ​​the clean area;

[0103] a2、FFU( Figure 2 The area affected by the location indicated by "X" is its own area plus the area of ​​four adjacent areas of the same size. Figure 2 (The positions indicated by "X", "1", and "2" in the text); the influence intensity of FFU on areas without FFU equipment is divided into two levels, level 1 ( Figure 2 The location indicated by "1" in the middle is where there is only one FFU in front of, behind, left or right of this area, which is level 2. Figure 2 The location indicated by "2" is where there are two FFU devices in the front and back (or left and right) positions of the area.

[0104] a3. When the FFU deployment rate is low (generally no more than 40%) and the FFU cannot fully affect the clean area, select the FFU deployment location with the largest area of ​​Level 1 influence intensity.

[0105] a4. When the FFU deployment rate is high (greater than 40%), and the FFU can have a comprehensive impact on the clean area, select the FFU deployment location with the largest area of ​​the Level 2 impact intensity area while ensuring comprehensive impact.

[0106] Figure 2 The diagram shows two layout methods when both have a 30% layout rate. Since FFUs cannot fully affect the clean area, the FFU with the largest area of ​​influence intensity at level 1 has better airflow (left side).

[0107] Step 3: Based on the actual solution requirements, if the influence of DCC on the cleanroom air temperature is considered, simplify the complex duct network structure of DCC into a flat plate of the same size as the actual size, and simplify the perforated raised floor and air outlets into planes of the same size as the actual size; if the influence of DCC on the cleanroom air temperature is not considered, simplify the complex duct network structure of DCC, the perforated raised floor and air outlets into planes of the same size as the actual size.

[0108] Step 4: Based on the above model simplification, establish a three-dimensional physical model of the cleanroom according to the FFU plan, cleanroom structure diagram, raised floor layout diagram, etc.

[0109] Step 5: Import the established 3D model into the mesh generation software, discretize the model, set the mesh size and mesh type, and perform mesh generation. Import the meshed model into Fluent software, select the appropriate governing equations according to the required solution, determine the model boundary type based on the actual working conditions and input the corresponding parameters, and select a suitable solution algorithm for calculation.

[0110] Step 6: When the residuals of the control equations decrease to the required range (e.g., less than 10e-4 for continuity equations, less than 10e-6 for energy equations) or the monitored physical quantities remain unchanged or change by an extremely small amount, the solution can be considered to have converged and the calculation is complete.

[0111] If the residuals or monitoring values ​​fail to meet the convergence requirements, the grid settings, boundary conditions, and solution algorithm in step four need to be adjusted until the convergence conditions are met.

[0112] Step 7: Import the calculation results into the post-processing software to obtain the cleanroom flow field distribution. When the cleanroom airflow deflection angle is large, adjust the cleanroom perforated plate arrangement to improve the cleanroom flow field. Repeat steps 5 to 7 until the airflow meets the cleanliness level requirements.

[0113] When the crosswind velocity on the equipment surface is high and the vortex on the side of the equipment is large, adjust the position of the FFU to reduce the crosswind velocity and vortex area. Repeat steps four through seven until the airflow meets the cleanliness requirements.

[0114] For airflow optimization in cleanrooms with large crosswind velocities and vortex areas, taking an electronic chip factory as an example, its clean area is approximately 1150㎡, with an upper mezzanine height of 2m, a cleanroom height of 4.5m, a lower mezzanine height of 6.4m, a cleanliness level of Class 100, and a perforated plate opening rate of 17%. Its air purification system consists of a combined fresh air handling unit (MAU) + fan filter unit (FFU) + dry coil (DCC). Fresh air is treated by the MAU and then sent to the upper mezzanine. The FFU then sends the fresh air into the cleanroom at a certain velocity, replacing the cleanroom air through the perforated floor plate to the lower mezzanine. The air from the lower mezzanine is then cooled to a certain temperature by the DCC and sent into the return air duct until it returns to the upper mezzanine to mix with the fresh air, and the same cycle repeats.

[0115] (1) Confirmation of drawing information

[0116] First, the cleanroom structure and the layout of DCC process equipment were determined based on the original drawings, with a total of 278 FFUs.

[0117] (2) Calculating the FFU placement rate to approximately 35%, when the FFUs cannot fully impact the clean area, select the FFU placement location with the largest area in the Level 1 impact intensity zone, such as... Figure 3 As shown.

[0118] (3) Simplification of porous model

[0119] If DCC needs to load cold / heat source terms at this location, it can be simplified to a porous medium Dancy's law model. If no heat source is needed, the model can be simplified to a porous step model to further simplify the model and improve the solution efficiency. In this case, the porous components can be named as surfaces, and there is no need to perform volume division.

[0120] This case study simulates the airflow patterns in a cleanroom, temporarily disregarding temperature variations. Therefore, the DCC and elevated perforated plate are simplified to a planar model (multi-perforated step model), as follows: Figure 4 The diagram shown is a simplified representation of a perforated plate. It is simplified to a planar plane along with other perforated components such as perforated plates and air vents. Taking DCC as an example, its thickness is 0.2m. Based on the perforated step condition and actual measured resistance and velocity values, the following is obtained: Figure 4 The resistance versus flow velocity fitting relationship is shown below:

[0121] ΔP = 5.242 × v + 3.866 × v 2

[0122] The permeability α and the inertial drag coefficient C2 are obtained by fitting the following formula.

[0123]

[0124] The calculation yielded:

[0125] α=0.00000068295; C2=31.72.

[0126] When fluid flows through a perforated plate, the plate's resistance characteristics depend only on the orifice ratio and are independent of the plate thickness, orifice diameter, and Reynolds number. Under the condition of uniform orifice arrangement, the relationship between the orifice ratio and the resistance coefficient is as follows:

[0127] ε=25426×n^(-2.051)R^2=0.9931

[0128] In the formula, ε is the local resistance coefficient and n is the orifice plate opening ratio.

[0129] The magnitude of the local resistance ΔP is calculated based on the drag coefficient ε. j :

[0130]

[0131] Ignoring the friction resistance along the orifice plate, that is:

[0132] ΔP=ΔP j

[0133] According to the source term S of the total resistance and momentum equation i Relationship between them:

[0134] ΔP=S i *L

[0135] In the formula, L is the thickness of the orifice plate perpendicular to the airflow direction.

[0136] get:

[0137]

[0138] Since the viscous resistance of the fluid in the cleanroom is very small, the influence of the first-order velocity term can be ignored when α = 1, and the inertial drag coefficient C2 can be calculated from the local resistance.

[0139]

[0140] Calculated

[0141]

[0142] (4) Model building

[0143] Based on the information in the drawings and the simplified content of the model, establish as follows Figure 6 The numerical simulation model shown.

[0144] (5) Mesh generation and parameter settings

[0145] Import the model into mesh generation software for mesh generation, such as... Figure 8 As shown, the parameters are imported into Fluent and set. The airflow control equation is the Reynolds-averaged Navier-Stokes equation, which includes the continuity equation, momentum equation, and energy equation in tensor form.

[0146]

[0147]

[0148]

[0149] in, Represents Reynold stress. This represents turbulent heat flux.

[0150] Fluent includes several turbulence models, among which the realizable k-ε model performs well for rotating flows, boundary layer flows with strong adverse pressure gradients, flow separation, and secondary flows. Therefore, the realizable k-ε model is selected for this case study. The realizable k-ε turbulence model includes various wall functions. Studies have shown that the Menter-Lechner wall function exhibits good adaptability to different Re values ​​and boundary layer thicknesses. The finite volume method is used for flow field solving. The pressure and velocity coupling uses the simple algorithm. The momentum and energy equations employ a second-order upwind dispersion scheme, while the turbulent kinetic energy k and turbulent dissipation rate ε use a first-order upwind dispersion scheme. The table below shows the specific boundary condition parameter settings:

[0151]

[0152]

[0153] (6) Post-processing and optimization settings

[0154] Based on the above settings, a simulation is performed. After the convergence condition is met, the solution is imported into post-processing software such as CFD-post to obtain information such as velocity contour maps or tilt contour maps.

[0155] ① When there is a significant crosswind above the equipment or a large vortex area on the side of the equipment, the position of the FFU should be adjusted. The purpose of the adjustment is to reduce the intensity and range of the crosswind and vortex on both sides of the equipment. The principle of FFU adjustment is to avoid the FFU vent blowing directly above the equipment, and to offset most of the FFU vent area from the upper surface of the equipment. Repeat steps (4)(5)(6) until the streamline meets the requirements.

[0156] Based on the above principles, the location of FFUs in the cleanroom was optimized, and the optimized model is as follows: Figure 7 As shown, Figure 9 , Figure 10 The diagram shows the velocity cloud diagrams of the cross section before and after optimization. It can be seen that after the above optimization principles, the crosswinds and vortices on both sides of the equipment are almost eliminated, which greatly improves the airflow streamline quality and meets the airflow requirements of the cleanroom.

[0157] ②Based on the different values ​​of K1, which is the ratio of the average air velocity at 0.4m below the raised floor and 0.5m from the return air duct to the average velocity at 0.1m above the raised floor, this invention provides a method for arranging the perforated plates of the raised floor.

[0158] Studies have shown that streamline inclination is mainly affected by the ratio of airflow in the lower mezzanine to the velocity near the perforated plate in the cleanroom, and is almost unaffected by the FFU outlet velocity. This conclusion remains valid regardless of whether the return air duct is single-sided or double-sided, or the height of the lower mezzanine. When the cleanroom is short from the return air duct (distance not exceeding 6m), it can be laid out in two non-uniform sections; when the cleanroom is long from the return air duct, the non-uniform openings can be arranged in multiple sections.

[0159]

[0160]

[0161] To better illustrate the streamline deflection optimization method obtained from the research, another model was selected for analysis, such as... Figure 11 The diagram shows the model. The lower duct height is 0.9m, the cleanroom net height is 4.5m, the upper duct height is 2m, the farthest distance from the cleanroom to the return air duct is 9m, the perforated plate arrangement rate is 50%, and the opening rate is 17%. Due to the symmetry of its flow field, only the flow field on one side is analyzed.

[0162] The following steps (3)(4)(5)(6) yield the following result: Figure 12 The streamline diagram shows a significantly large inclination angle of the flow field velocity within the cleanroom. Calculations show that the average inclination angle of the section at a height of 1m from the raised floor is 34.2°, with K1 = 14 ≥ 9. Based on the K1 value range in the table, the perforated plates in the cleanroom are arranged in three sections with opening ratios of 10%-17%-25%. The average opening ratio of all perforated plates remains the same as before optimization, at 17%, effectively ensuring the same pressure difference between the cleanroom and the upper mezzanine before and after optimization. Figure 13 The diagram shows the optimized flow field streamlines. After optimization using method ② above, the inclination angle of the 1m height section was reduced from 34.2° to 7.9°, a reduction of 77%, which greatly improved the streamline quality of the cleanroom.

[0163] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0164] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for optimizing airflow organization based on CFD-based cleanroom air outlet and perforated plate arrangement, characterized in that, include: Step 1: Determine the location of equipment and the actual number of fan filter units (FFUs) in the cleanroom based on the actual preliminary design drawings; Step 2: Determine the area that the FFU can affect based on the FFU deployment rate. When the FFU cannot affect the entire clean area, the location where the FFU can affect the largest area is taken as the FFU deployment location. When an FFU can affect the entire clean area, the FFU placement location should be the area where the FFU has the greatest high-intensity impact. Step 3: Based on the actual solution requirements, if the influence of DCC on the cleanroom air temperature is considered, simplify the complex duct network structure of DCC into a flat plate with the same actual size, and simplify the perforated raised floor and air outlet into a plane with the same actual size. If the impact of DCC on cleanroom air temperature is not considered, the complex duct network structure, porous raised floor and air outlets of DCC are simplified to a plane with the same actual size. Step 4: Based on the above steps, establish a three-dimensional physical model of the cleanroom according to the FFU floor plan, cleanroom structure diagram, and raised floor layout diagram; Step 5: Import the established 3D model into the mesh generation software for mesh generation, and then use Fluent software to select the appropriate governing equations according to the required solution content. Determine the model boundary type based on the actual working conditions, input the corresponding parameters, and select a suitable solution algorithm for calculation; Step 6: When the residual of the solved governing equation decreases to the required range or the monitored physical quantity remains unchanged or changes by an extremely small amount, the solution can be considered to have converged and the calculation ends. Step 7: Import the calculation results into the post-processing software to obtain the cleanroom flow field distribution. When the cleanroom airflow deflection angle is large, adjust the cleanroom orifice plate arrangement to improve the cleanroom flow field. Repeat steps 5 to 7 until the airflow meets the cleanliness level requirements. When the crosswind speed on the equipment surface is high and the vortex on the side of the equipment is large, adjust the position of the FFU to reduce the crosswind speed and the vortex area. Repeat steps four through seven until the airflow meets the cleanliness requirements.

2. The method for optimizing airflow organization based on CFD cleanroom air outlets and perforated plate arrangement according to claim 1, characterized in that: Step two, the initial location optimization of the FFU, specifically includes: a1. Calculate the FFU deployment rate based on the number of FFUs and the area of ​​the clean area; a2. The area affected by an FFU is its own area and four adjacent areas of the same area. The intensity of the impact of an FFU on an area without an FFU is divided into two levels: Level 1 is when there is only one FFU in front of or behind the area or to the left or right of the area; Level 2 is when there are two FFUs in front of or behind the area or to the left or right of the area. a3. When the FFU deployment rate is low, i.e. no more than 40%, and the FFU cannot fully affect the clean area, select the FFU deployment location with the largest area of ​​the Level 1 influence intensity zone. a4. When the FFU deployment rate is high, i.e. greater than 40%, and the FFU has a comprehensive impact on the clean area, the FFU deployment location with the largest area of ​​the Level 2 impact intensity area should be selected while ensuring the comprehensive impact.

3. The method for optimizing airflow organization based on CFD cleanroom air outlets and perforated plate arrangement according to claim 1, characterized in that: In step three, the DCC, perforated plate, and air vent are simplified. When performing thermal coupling, DCC can select a porous media model; When thermal coupling is not performed, DCC can choose the porous step model; For air vents and raised floors without directly loaded heat source components, they are simplified as planes of a porous step model.

4. The method for optimizing airflow organization based on CFD cleanroom air outlets and perforated plate arrangement according to claim 3, characterized in that: After simplification using porous media or porous steps, when adjusting the airflow resistance through the porous structure, the resistance change can be achieved simply by modifying the relevant coefficients in the boundary conditions.

5. The method for optimizing airflow organization based on CFD cleanroom air outlets and perforated plate arrangement according to claim 3, characterized in that: Provides various simplified calculation methods for perforated plates; When DCC needs to load a heat source, it is simplified to a solid plate according to Dancy's law model. Based on the actual test results of the resistance and flow velocity relationship, the permeability α and the inertial drag coefficient C2 are obtained by fitting the following formula. 。 6. The method for optimizing airflow organization based on CFD cleanroom air outlets and perforated plate arrangement according to claim 3, characterized in that: When DCC is not loaded with a heat source, it is simplified to a plane along with porous components such as orifice plates and air outlets. According to the porous step condition, based on the relationship between resistance and flow velocity obtained from actual tests, the permeability α and inertial drag coefficient C2 are obtained by fitting according to the following formula. 。 7. The method for optimizing airflow organization based on CFD cleanroom air outlets and perforated plate arrangement according to claim 1, characterized in that: When fluid flows through a perforated plate, if Re > 30 and the orifice ratio does not exceed 50%, the plate's resistance characteristics are only related to the orifice ratio and are independent of the plate thickness, orifice diameter, and Re number. Under the condition of uniform orifice arrangement, the relationship between the orifice ratio n and the resistance coefficient ε is: ε=25426×n -2.051 R 2 =0.9931, The magnitude of the local resistance is calculated based on the drag coefficient ε. : ， Ignoring the friction resistance along the orifice plate, that is: ， According to the source term of the total resistance and momentum equation Relationship between them: ， In the formula, L is the thickness of the orifice plate perpendicular to the airflow direction; get: = ， Since the viscous resistance of the fluid in the cleanroom is very small, the influence of the first velocity term can be ignored when α=1, and the inertial drag coefficient C2 can be calculated through the local resistance. 。 8. The method for optimizing airflow organization based on CFD cleanroom air outlets and perforated plate arrangement according to claim 7, characterized in that: The relationship between the orifice plate resistance coefficient and the opening ratio was obtained through extensive experimental fitting and regression. Based on the regression formula, the velocity first-order coefficient was further optimized considering the characteristics of the cleanroom orifice plate. The CFD-based cleanroom air outlet and perforated plate arrangement airflow organization optimization design method has been mutually verified by simulation and experiment, and has a good ability to reflect the pressure drop and flow velocity changes of gas flowing through the perforated plate.

9. The method for optimizing airflow organization based on CFD cleanroom air outlets and perforated plate arrangement according to claim 1, characterized in that: In step five, Fluent includes multiple turbulence models, among which the realizable k-ε model performs well for rotating flows, boundary layer flows with strong adverse pressure gradients, flow separation, and secondary flows. Therefore, the turbulence model selected in this study is the realizable k-ε model; The realizable k-ε turbulence model includes various wall functions. The Menter-Lechner wall function shows good adaptability to different Re ranges and different boundary layer thicknesses. The finite volume method is used in the flow field solution. The simple algorithm is used for pressure and velocity coupling. The momentum and energy equations are based on the second-order upwind dispersion scheme, and the turbulent kinetic energy k and turbulent dissipation rate ε are based on the first-order upwind dispersion scheme. The governing equations for the airflow are the Reynolds-averaged Navier-Stokes equations, which include the continuity equation, momentum equation, and energy equation in tensor form: ， ， ， in, Represents Reynold stress. This represents turbulent heat flux.

10. The method for optimizing airflow organization based on CFD cleanroom air outlets and perforated plate arrangement according to claim 1, characterized in that: In step seven, when the crosswind intensity on the upper surface of the equipment is high and the vortex area on the side of the equipment is large after the production process equipment is arranged in the clean room, the position of the FFU air outlet should be adjusted. The principle of adjustment is to place the air outlet off the upper surface of the equipment and repeat steps four to seven until the requirements are met.

11. The method for optimizing airflow organization based on CFD cleanroom air outlets and perforated plate arrangement according to claim 1, characterized in that: In step seven, the unidirectional flow cleanroom has high requirements for the cleanroom flow line. When the FFU arrangement rate exceeds 40% and the cleanroom flow line deflects and does not meet the requirements, the cleanroom perforation plate opening rate is arranged according to the ratio K1 value of the average air velocity at 0.4m below the raised floor and 0.5m from the return air duct to the average velocity of the cross section at 0.1m above the upper surface of the raised floor. ， in, This indicates the average air velocity at a point 0.4m below the raised floor of the cleanroom and 0.5m from the return air duct. The average flow velocity is the value of the cross-section at the top surface of the elevated floor, which is 0.1m above the floor surface. When the cleanroom is relatively short from the return air duct, i.e., the distance to the nearest return air duct is no more than 6m, it can be laid out in two non-uniform sections; when the cleanroom is relatively long from the return air duct, the non-uniform openings can be arranged in multiple sections.

12. The method for optimizing airflow organization based on CFD cleanroom air outlets and perforated plate arrangement according to claim 11, characterized in that: When the cleanroom flow line is tilted, multi-directional simulations are performed on different cleanroom structures, different lower mezzanine heights, different room depths, different FFU arrangement rates, and different FFU flow velocities. The results show that the relationship between the K1 value and the floor opening rate arrangement is still applicable. The simulation results are compared with the experimental results to obtain a quick optimization strategy for selecting the perforated plate arrangement based on the K1 value range.

Images