KR980009957A - Design Method of Cooling Fan for Noise Reduction and Noise Prediction Method of Cooling Fan - Google Patents

Abstract

The present invention is designed to design the axial fan based on the basic design specifications such as the air flow rate and the pressure rise and the rotation speed of the motor, and also to predict the noise level at the operating point in advance in designing, And a method for predicting noise radiation of a cooling fan.

The present invention makes it possible to obtain a cooling fan capable of minimizing noise generation so that the noise of the cooling tower can be reduced to an optimum state and the ambient noise can be predicted so that the cooling tower can be used at the time of noise- .

Description

Design Method of Cooling Fan for Noise Reduction and Noise Prediction Method of Cooling Fan

FIG. 1 is a hardware configuration diagram to which the present invention is applied; FIG.

FIGS. 2 (a) and 2 (b) are flow charts for performing a cooling fan designing method for noise reduction according to the present invention. FIG.

FIG. 3 is a graph showing the noise state according to the noise prediction. FIG.

FIGS. 4 and 5 also illustrate a design of a cooling fan according to the present invention. FIG.

6 is a flow chart for performing a method of predicting noise radiation of a cooling fan according to the present invention.

DESCRIPTION OF THE REFERENCE NUMERALS

10: key input unit 20:

30: memory unit 40: display unit

The present invention relates to a cooling fan, and more particularly, to design an axial fan based on basic design specifications such as an air flow rate, an increase in pressure, and a rotation speed of a motor, and at the same time, A method for designing a cooling fan for noise reduction and a method for predicting a noise emission of a cooling fan.

Generally, there is a great deal of noise around the cooling towers that are responsible for the cooling of one or several buildings.

Therefore, a sound absorber is installed around the cooling tower and a noise can be reduced by installing a silencer in a path where sound can be propagated.

However, this is not done by measuring the noise according to the surrounding environment and installing the soundproof wall or installing the silencer but without measuring the noise to the surrounding environment, so that the proper noise preventing function according to the surrounding environment is not provided.

SUMMARY OF THE INVENTION It is an object of the present invention to provide a method of designing a cooling fan for noise reduction so as to reduce a noise of a cooling tower to an optimal state by obtaining a cooling fan capable of minimizing noise generation. have.

It is another object of the present invention to provide a method for predicting noise emission of a cooling fan that can be used at the time of soundproofing of a cooling tower by predicting the noise around the cooling tower.

According to an aspect of the present invention, there is provided a method for designing a cooling fan for noise reduction, the method comprising: a first step of determining a variable input and an initial value necessary for designing a fan; A third step of determining whether the values obtained in the second step converge; and a third step of determining whether or not the values obtained in the second step converge; and, when converging in the third step, And a fourth step of obtaining a fan design specification.

The method of predicting noise radiation of a cooling fan according to the present invention includes a first step of inputting a specification of a cooling tower, an environment of an installation layout and an installation space, a second step of considering a direction characteristic according to a point noise source assuming that the cooling tower is a point- A third step of calculating a noise level reduction according to an increase in distance from a noise source after determining a power level of the point noise source, a third step of calculating a noise level reduction according to an environmental condition, And a fifth step of calculating a noise level at a listener position.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will now be described in detail with reference to the accompanying drawings.

FIG. 1 is a hardware block diagram for implementing a method for designing a cooling fan for noise reduction and a method for predicting noise emission of a cooling fan according to the present invention. The hardware block diagram includes a key input unit 10 for inputting various keys, The control unit 20 controls the overall system according to the key input of the key input unit 10 and is responsible for noise prediction and fan design. In accordance with various operations and controls performed in the control unit 20 according to the key input of the key input unit 10, A memory unit 30 for storing the obtained data and a display unit 40 for displaying the shape of the fan designed according to the control of the controller 20 according to the key input of the key input unit 10. [

A cooling fan designing method for noise reduction according to the present invention operated in the above-configured hardware will be described with reference to the flowcharts of FIGS. 2 (A) and 2 (B).

First, a fuzzy set according to Heller Number, Chamber Angle, Aerodynamic Load Factor, Solidity, and Increment (Hub Ratio, Tip / Hub Lift Factor) (hub Ratio: √) and determines an initial value such as the lift coefficient _{C L) T, C L)} H of the tip / hub (S11).

Diameter coefficient:? ^X ? 1.865d _T [? (P /?)] ^1/4 Q ^-1/2

Speed coefficient:? ^X = (N / 28.5) Q ^1/2 (? (P /?)) ^3/4

Diameter _{coefficient: d T = (60 / Nπ} ) √2 △ pσ * 2δ * 2 / ρ

^{(Q: m 3 / s,} △ p: kg / m 3, N: rpm)

After inputting the air flow rate, pressure rise and rotation speed, initial values such as inlet condition, polytropic efficiency, workpiece factor and the like are determined, and determination of the type (Vortex Type) and tip / Lift factor at the hub is selected (S12).

Here, the calculation of the polytropic efficiency is as follows.

In the first calculation process, the efficiency at the design point is obtained by substituting the assumed values arbitrarily and recalculating the efficiency derived from the loss calculation through the entire design process. This means pure aerodynamic efficiency excluding the influence of the compression ratio in one end.

E = C _P ? T _O =? _{U m} (? C _u ) _m

And, the work parameter indicates the decrease of energy transfer due to the change of the axial velocity due to the boundary layer of the casing wall surface.

? = 0.86 or 0.98

After the step S12 is performed, the diameter of the tip is determined with the energy transfer estimated from the Euler's Equation (S13).

Then, the linear velocity component: determines the revolution speed component at the average radius, determines the distribution of the revolution speed component according to the velocity triangle at the average radius: the relative velocity at the inlet / outlet and the relative velocity, A three-dimensional flow field is obtained (S14).

Then, the velocity triangle according to the radius is completed (Vortex Design) (S15).

The step (S15) is a work for determining the speed triangle along the radial direction, and it is possible to neglect the interference of the adjacent oil surface by considering the oil surface independently, but the pressure force is applied to each other and the mechanical balance is established.

Radial equilibrium _{equation: dh o / dr - T (} ds / dr) = C a (dc a / dr) + (C μ / r) (d (rC μ) / dr)

1 d dc²a

--- --- (r²C²μ) + ----- = 0

r² dr dr

The revolution speed component at the entrance and exit: c _u1 = ar ⁿ - b / r

c _u2 = ar ⁿ + b / r

If there is no revolution speed component at the inlet side, the exponent n is -1.

Then, the solidity is determined from the lift coefficient in the tip / hub (S16).

σ = c / s

s: pitch between wings

c: Code length

As this value increases, the direction of flow becomes more efficient, while the friction loss increases due to the increased surface area. In the opposite case, the friction loss is reduced, but the horn is prone to separation on the wing surface. This value affects performance and blower as well as wing count and wing shape.

1) Method by the aerodynamic load _{factor: σ = 2 (tanβ 1 -} tanβ 2) cos 2 β 2 / ψ L cosβ M

tan? _M = (1/2) (tan? ₁ + tan? ₂ )

2) Method based on the lift coefficient: σ = (2 / C ₁ ) cos β _M (tan β ₁ - tan β ₂ )

Then, the flow angle and blade angle beta ₁ are determined (S17).

This uses Howeel's empirical formula.

Chamber angle:? =? ¹ ₁ -? ¹ ₂ =? ₁ -? ₂ +?

Stagger angle: ξ = β ¹ ₁ - θ / 2

Deviation angle:? = M? m (? ₁ -? ₂ +?)

At this time, m = 0.23 + 0.1 (β 2/50) δ = m (β 1 - β 2) / (1-m )

After the step S17 is performed, the shape of the cross section of the blades is determined (S18).

Then, loss and efficiency are calculated (S19).

(1) Loss of blade profile (Blade Profilc loss)

━

_{ωp = 2 (θ 2 / c} ) σ (cos 2 β 1 / cos 3 β 2) (2H 2 / 3H 2 -1) [1- (θ 2 / c) σH 2 / cosβ 2] -3

_{D eq = (cosβ 2 / cosβ} 1) [1.12 + 0.61cos 2 β 1 (s / c) (tanβ 1 - tan (β 2)]

? ₂ / c = 0.00138exp (1.1127D _eq ) + 0.0025

_{H 2 = 1.26 + 0.795 (D} eq - 1) 1.681

θ ₂ : Momentum thickness at the cascade exit

D _eq : equivalent diffusion coefficient

H ₂ : boundary layer shape coefficient

(2) Secondary Flow Loss (Secondary Flow Loss)

Loss model of IIorlock and Lakshiminarayana

----

_{s m} = ω 0.1336s / H _m (tanβ _sω1 - _sω2 tanβ) ² cos ² β _m1 / cosβ _M

(3) End wall loss

-

ω _σ = R _W ω _p

R _w = (S _m (r ₁ + r _H ) / r _m H) cos ζ ((sin (│β ₁ - β ₂ │ / 2) / │β ₁ - β ₂ )

(4) Tip clearance loss

?? = 0.7 (? / H)? / Cos? _M [1.0 + 1.0 AR / cos? N

At this time, ψ = Δp ₀ / (ρω _2TM / 2), φ = ω _T1 / U _T , AR = aspect ratio

After the step S19 is performed, convergence is tested (S20).

This is to check whether the value obtained according to each of the above steps converges to a desired value. The desired value is a value for designing a cooling fan in which noise can be minimized. If the value is converged, the process proceeds to the next step S21 and converges The process returns to step S13 and repeats the steps S13-S19.

In this step S20, the next value is calculated and it is determined whether or not convergence occurs.

--------- --- --- --- ---

Voltage loss: △ PL power = (ω _p + ω _e + ω _s + ωl) ρω²m2 / 2 (rotor)

--------- --- --- ---

△ PL power = (ω _p + ω _e + ω _s ) ρ ω 2 m 2/2 (stator)

Efficiency: η = 1 - ΔP _L / ΔPid

Where △ Pid = ρω²₁ / 2 (1 - cos²β₂ / cos²β₁)

Total efficiency of blower: η = Λη ratio + (1 - Λ) η stator

In step S21, the aerodynamic load factor? L in the hub obtained from the calculation result of each of the steps S12-S20, the solidity? H in the hub, the aerodynamic load coefficient (Ha = cos? 2 / cos? 1) and the chamber angle? H in the hub are obtained (S21).

The conditions and relations of the respective coefficients are as follows.

1) The lift coefficient does not exceed 1 for the entire radius, and increases from tip to hub.

2) The aerodynamic load factor ψL should be in the range of 0.7-1

3) Solidity a should be in the range of 0.6 to 1.5.

4) Number of Hallers in the hub Ha) H should be greater than 0.75

5) The camber angle θ) H in the hub should be less than 20 degrees

The solidity is calculated by varying the lift coefficient in the tip and hub and linearly distributing it along the radius.

The rules for satisfying the above conditions are as follows (S22).

1) If ψL in the hub is greater than 1, increase the CL in the hub.

2) If ψL in the hub is smaller than 1, reduce the CL in the hub.

3) If σ at the hub is greater than 1.5, increase the CL at the hub.

4) If σ at the hub is less than 1.5, decrease the CL at the hub.

5) If ψL at the tip is less than 0.7, increase the CL at the tip.

6) If ψL at the tip is greater than 0.7, decrease the CL at the tip.

7) If σ at the tip is greater than 0.6, reduce the CL at the tip.

8) If σ at the tip is less than 0.6, increase the CL at the tip.

9) If the number of Hallers in the hub is greater than 0.75, the hub ratio is reduced.

10) If the Haller's Haller number is less than 0.75, increase the hub ratio.

11) If the camber angle is greater than 20 degrees on the hub, reduce the hub ratio.

12) If the camber angle at the hub is less than 20 degrees, increase the hub ratio.

Use the above rules to get each membership.

On the other hand, after performing the step S22, the union of the memberships is lost (S23).

1) ψL) Membership μ 1 for determining △ CL) H from the union of the two memberships obtained by H and σH is obtained.

2) We obtain the membership μ2 to determine ΔCL) T from the union of the two memberships obtained by ψL) T and σT.

3) We obtain the membership μ3 for determining △ ν from the union of two memberships obtained by Ha and θ) H.

Then, the increment based on the area centering method is determined (S24).

μ 1, μ 2, and μ 3 are determined by the gravity center method to determine the increment Δ CL) H, Δ CL) T, Δ √.

Then, the convergence state is checked (S25). If convergence does not occur, a new design parameter is obtained (S26). When the convergence is achieved, a new design parameter is obtained.

A process of designing the cooling fan according to the cooling fan designing method will be described as follows.

First, execute the data file (axt.bat) necessary for the fan design. At the initial screen, input the following items.

The user inputs the basic design specifications (CMM [mm³ / min]), the voltage rise (mmAq), and the revolutions per minute (rpm) of the motor. After entering, press the enter key to change the input value from green to blue.

On the other hand, the number of wings of the fan should be specified as the last specification.

Originally, this should be decided in the design program. However, since the number of blades is a design factor sensitive to the performance and noise of the fan, it is included in the basic design specification since there is not a satisfactory design formula or an empirical formula for the determination.

After finishing the input, if you want to proceed with the design using the specification, enter 1 in the next window and enter 0 if you want to re-enter the input.

Once in the design, the program finds the optimal design point by varying the lift coefficient at the hub ratio, the tip, and the lift coefficient at the hub. In this state, the screen is erased and each value converges toward a constant value at every step.

On convergence, the screen briefly shows the results of the design. The result according to the radius is stored as ASCII file in the file (AXT.OUT) which stores the design according to the radial direction of the designed fan, and the value in the average radius stores the specification in the average radius of the designed fan And stored in the file (AXFM.OUT).

At this time, pressing any key will show the result of noise prediction. This is shown in FIG.

This represents the frequency by grabbing the X axis on a log scale. Originally, the total noise is not consistent with the Turbulent Boundary Layer Noise, but it does not take into account the laminar boundary layer noise.

On the other hand, if any key is pressed in this state, the state changes to the fourth state. This shows the 3D shape of the finished fan. You can see the angle of the fan by pressing the direction keys (Left, Right, Up, Down) and Page Up and Page down keys.

The direction key causes an angle change of 10, and the page up / down key causes an angle change of 30 degrees. In the numeric panel of the keyboard, press the + key to zoom and the - key to reverse.

The following is an example of use.

Example 1

When the screen for initial input is displayed, enter the following.

Air volume: 2200.0 ([CMM])

Pressure rise: 85.0 ([mmAq])

Rotation speed: 1000.0 ([rpm])

Number of wings: 12

The three-dimensional shape of the fan based on the input value is the same as the fourth aspect, and the prediction about the noise spectrum is the same as the third aspect.

Usage example 2

When the screen for initial input is displayed, enter the following.

Air volume: 1500.0 ([CMM])

Pressure rise: 30.0 ([mmAq])

Number of revolutions: 1200.0 ([rpm])

Number of wings: 6

The three-dimensional shape of the fan according to the input value is the same as the fifth aspect.

The following will describe the noise radiation prediction method.

The purpose of the noise emission prediction method is to predict the noise level (SPL) according to the surrounding environment according to the distance from the noise source due to the operation of the small CT (Coding Tower) installed around the residential area.

In addition, the program is composed of BASIC, and the algorithm is developed in an easy-to-understand manner, so that the user can easily modify and supplement the program on the spot.

In this method, the noise power level of the noise source is calculated using the noise magnitude measured at an arbitrary point during CT operation, or the power level of the noise source is calculated from the power of the driving motor for driving the fan, Considering the reflective wall, floor, barrier, and direction (using the direction factor DI of the CT because the directionality is different for each CT) and the air condition of the transmission path (standard: room temperature 20 ℃, humidity 70%) And calculates the noise magnitude at a certain distance from the CT.

Here, the information on the surrounding environment reflects the reflection characteristics of the noise in the residential area by performing the sampling measurement in a typical settlement area.

The reflection wall and the ground consider a reduction in the best path from the virtual noise source to the receiver to know the noise magnitude by creating a virtual source at the reflection point.

When there is a barrier, the reduction amount is determined by considering only the noise transmitted through the barrier.

In order to increase the reliability of the simulation results for the noise measurement results, it is necessary to predict the radiation noise level of the CT in advance when the CT is installed in a position having a similar surrounding environment. At this time, the propagation of the sound in the half space Frequency band.

SPL = PWL-68 + 17.51 * exp (-R / 6.302) + 47.01 * exp (-R / 211.9) + DI + △ L- △ A- △ A air (1)

Here, PWL: power level of the noise source [dBA]

R: Distance from cooling tower [m]

DI: Correction value by Solid Angle (value by ground, 3 when 2π) [dBA]

ΔL: Correction value due to vertical rigid reflecting wall [dBA]

A: Correction value by barrier [dBA]

△ A _air : Decrease by air [dBA]

In addition, since the noise emission characteristics of the top and bottom of the fan stack of the CT are different, calculate the noise emission angle by using the equation (2).

The ratio (-) of the SPL and the PSL by angle at the horizontal point with the cooling tower

Y = 1.0034 + 0.0014 x? + 2.7726 x 10^-5× θ²-7.8135 x 10^-7× θ³+ 4.7198 × 10^-9× θ⁴(2)

Here,?: Noise radiation angle [deg]

A) PWL measurement of noise source

1) When there is any measurement data from cooling tower

PWL = SPL + 20 log (R) (3)

Here, SPL is the noise size [dBA] measured at an arbitrary distance R,

2) When the same CT is installed in multiple units, the following equation (4) is used to predict the size of the noise source by the number of CTs and the fan.

PWL = 10 log (hp x N) + 10 log (ss) + 60.2 (4)

Here, hp: power of the driving motor of the fan (HP)

N: Number of CTs in operation [large]

ps: static pressure [cm]

In the equation (4), the correction value of 60.2 dBA is calculated by the trial error method. Therefore, the value of 26 dBA is smaller than that of the conventional large capacity fan drive. In the case of this experiment, it is possible to readjust the correction value by the sample test under different measurement conditions.

B) Reduction by air

The decrease due to air is expressed by the following equation (5).

Aatm, band = (R / 1000) [0.2 + 3.6 (Fband / 1KHZ) = 0.36 (pband / 1HKZ)

Where R: distance from noise source to listener [m]

Fband: Octave Band Center Frequency [HZ]

(C) Consider the surrounding reflective wall

The correction of the increase in the noise magnitude due to the surrounding reflection wall is not considered when the imaginary distance due to the reflection wall reaches 10 times or more of the distance directly from the noise source.

Log (ri / rs]] ²-3.84 [log [ri / rs]] ³ (6)

Where ri is the virtual distance [m]

rs: distance to reach directly [m]

The noise radiation prediction method of the present invention as described above will be described with reference to the flowchart of FIG.

First, the cabinet top specification: motor power for fan driving, static pressure, blowing amount, installation layout: distance from the cooling tower to the reflecting wall, environment of the installation space: temperature and humidity are inputted (S31).

Assuming that the cooling tower is a point source, assuming that the cooling tower is a point source, it is assumed that the cooling tower structure is large but emits from one of its major noise sources (S32).

In addition, the directional characteristics of the noise emitted from the cooling tower are considered (S33).

This experiment considers the following characteristics in the direction of rotation from the center of the cooling tower.

0.979797 0.959595 0.93939 0.919191

Then apply the following equation from the cooling tower in the vertical direction.

FACTOR_TH = 1.0033804 + 0.0014361637 * TH_DEGEREE + 0.00002772574 * TH_DEGREE∧2 - 0.00000078135169 * TH_DEGREE∧3 + 4.7198426E_0.9 * TH_DEGREE∧4

Where TH_DEGREE: angle in the vertical direction of the cooling tower

FACTOR_TH: Directional characteristic value according to angle

Then, the noise power level of the point noise source is calculated from the input value (S34), which is calculated according to the above equation (4)

In step S35, the noise level decrease due to the increase in distance from the noise source is calculated according to Equation (1). In step S36, the noise level decrease according to the environmental condition is calculated according to Equation (5).

Then, an echo for the surrounding building according to the installation layout is calculated (S37), and this is in accordance with the above expression (6).

Then, the noise level at the position of the listener is calculated (S38).

Since this is a three-dimensional shape, it is calculated by applying the above-described step S35 to calculate noise reduction by distance according to the direction.

And outputs the noise level to the graphic screen and file at the listener position.

As described above, according to the present invention, it is possible to predict the noise according to the surrounding environment of the cooling tower. Therefore, it is possible to effectively sound the sound when the cooling tower is soundproofed. By designing the cooling fan to have the minimum noise generating condition, .

Claims (31)

A second step of obtaining a fan design specification based on the variable input and the initial value, a third step of determining whether the values obtained in the second step converge, And a fourth step of obtaining a fan design specification capable of reducing noise generation compared to the fan design specification obtained in the second step using the fuzzy theory when converging in the third step. How to Design a Cooling Fan. 2. The method of claim 1, wherein the first step comprises a fuzzy set according to the Hall coefficient, the chamber angle, the aerodynamic load factor, the solidity and hub ratio, and the lift coefficient at the tip / hub, The step of obtaining the lift coefficient (11), the step of inputting the air flow rate, the pressure rise and the rotation speed, and the entrance condition, the polytropic efficiency, the work money factor, and the type determination and the lift coefficient in the tip / (S12). ≪ Desc / Clms Page number 20 > 3. The method of claim 2, wherein the lift coefficient is not greater than 1 for an overall radius and increases from tip to hub. The method according to claim 1, wherein the second step includes: determining a tip radius (S13) with energy transfer estimated from the Euler equation; obtaining a velocity triangle at an average radius according to the linear velocity component determination (S14) (S15) determining a velocity triangle along the radius, determining a solidity from a lift coefficient at the tip / hub (S16), determining a flow angle and a wing angle (S17), and determining A step S18 of determining the shape of the blade section and a step S19 of calculating various losses and efficiencies. 5. The method of claim 4, wherein the solidity is determined by aerodynamic load factor or lift factor. 6. The method of claim 5, wherein the aerodynamic load factor is in the range of 0.7-1. 5. The method of claim 4, wherein blade profile loss, secondary flow loss, end well loss, and tip clearance loss are obtained in step S19. The method according to claim 1, wherein, when convergence is determined in the third process, the process returns to the second process when the convergence is not converged. The method according to claim 1, wherein the fourth step comprises the steps of: obtaining a number of halos, a chamber angle, an airpheric load coefficient, and a solidity as a result of the step (S12-S20) (S23) of determining the union of memberships, and (S24) determining an increment based on the area centering method (S24). 10. The method of claim 9, wherein the Haller number is 0.75 or more in the case of a Haller number at the hub. 10. The method of claim 9, wherein the chamber angle is less than 20 degrees for a chamber angle at the hub. 10. The method of claim 9, wherein if the aerodynamic load factor at the hub for solidsity calculation is greater than 1, the lift coefficient at the hub is increased. 10. The method of claim 9, wherein if the aerodynamic load factor in the hub for solidsity calculation is less than one, the lift coefficient in the hub is reduced. 10. The method of claim 9, wherein the lift coefficient at the tip is increased if the aerodynamic load factor at the tip for solids calculation is less than 0.7. 10. The method of claim 9, wherein the lift coefficient at the tip is reduced if the aerodynamic load factor at the tip for solids calculation is greater than 0.7. 10. The method of claim 9, wherein the Haller number is smaller than 0.75 when the Haller number is a Haller number. 10. The method of claim 9, wherein the Haller's number is greater than 0.75 when the number of Haller's is greater than 0.75. 10. The method of claim 9, wherein if the chamber angle is greater than 20 degrees for a chamber angle in the hub, the hub ratio is reduced. 10. The method of claim 9, wherein the chamber angle is greater than 20 degrees when the chamber angle is at the hub, thereby increasing the hub ratio. The cooling fan design method according to claim 9, wherein the solidity is in the range of 0.6 to 1.5. 10. The method of claim 9, wherein if the solidity in the hub is greater than 1.5, the lift coefficient in the hub is reduced. 10. The method of claim 9, wherein if the solubility in the hub is less than 1.5, the lift coefficient in the hub is reduced. 10. The method of claim 9, wherein if the solidity at the tip is greater than 0.6, the lift coefficient at the tip is reduced. 10. The method of claim 9, wherein if the solidity at the tip is less than 0.6, the lift coefficient at the tip is increased. The method of claim 1, further comprising: determining convergence after the fourth step to determine a new hub ratio and a lift coefficient in the tip / hub when non-convergence and returning to step S12; Wherein the cooling fan further comprises: A second step of inputting the environment of the cooling tower specification, the installation layout and the installation space; a second step of considering a directional characteristic according to a point source assuming that the cooling tower is a point source and a noise power level of the point source; A fourth step of calculating a noise level reduction according to an environmental condition, a fourth step of calculating a direction depending on an installation layout, a fifth step of calculating a noise level at a listener position, Wherein the step of estimating the noise radiation of the cooling fan comprises the steps of: 27. The method of claim 26, wherein the noise power level in the third step is calculated as SPL + 20 log (R) if there is any measurement data from the cooling tower. Here, SPL is the noise magnitude [dBA] measured at a certain distance R away. 27. The method of claim 26, wherein the noise power level in the third step is calculated as 10 log (hp x N) + 10 log (Ps) + 60.2 when a plurality of identical CTs are installed. Way. Here, hp is the power (HP) of the drive motor of the fan, N is the number of movable parts of the CT [unit], and ps is the static pressure [cm]. 26. The method of claim 26, wherein the noise level reduction with increasing distance in the third step is PWL-68 + 17.51 * exp (-R / 6.302) + 47.01 * exp (-R / 211.9) + DI + - DELTA A < a > _air . Where d is the power level of the noise source, d is the distance from the cooling tower, m is the solid angle, and d is the solid angle, Correction value [dBA] by the reflecting wall, A: Correction value by the barrier [dBA], and ΔA _air : Decrease by air [dBA]. 27. The method of claim 26, wherein the noise level reduction according to the environmental condition in the fourth step is calculated as: (R / 1000) [0.2 + 3.6 (fband / 1KHz) = 0.36 (fband / 1KHz) Method of predicting noise radiation of a fan. Where R is the distance [m] from the noise source to the listener, and fband is the octave band center frequency [Hz]. The method of claim 26, wherein the echo according to the installation layout in the fourth step is a sound level reduction is _{3.00 - 9.29log 10 [ri / rs} ] + 10.13 [log 10 [ri / rs]] 2 -3.84 [log [ri / rs]] < 3. < / RTI > Where ri is the virtual distance [m] due to the reflecting wall, and rs is the distance [m] directly reached. ※ Note: It is disclosed by the contents of the first application.