Btp report final_lalit

CPP 301 Core Project Part I

DESIGNING OF A NOISE BARRIER

Submitted by

Lalit Aggarwal &Gayathri Lakshmi Kulukuru

Supervisor

Dr. Navin Kumar

School of Mechanical Materials & Energy Engineering

INDIAN INSTITUTE OF TECHNOLOGY ROPAR

Nov 2012

~ 1 ~

CERTIFICATE

This report is submitted by Mr.Lalit Aggarwal &Ms.Gayathri Lakshmi Kulukuru detailing

the work done during the 1st semester, 2011-2012. The report written and all material taken

from other sources(books, manuals, journals,etc.) have been fully acknowledged.

Lalit Aggarwal Gayathri Lakshmi Kulukuru

P2009ME1085 P2009ME1062

SMMEE, IIT Ropar SMMEE, IIT Ropar

Date: 09-Nov-2012 Date: 09-Nov-2012

Mr. Lalit Aggarwal &Ms.Gayathri Lakshmi Kulukuru have worked under my supervision

during this semester. I have read this report; it meets the expectations and it accurately

reflects the work done by the students.

Dr. Navin Kumar

(Supervisor)

Date: 08-Nov-2012

~ 2 ~

ACKNOWLEDGMENTS

The authors acknowledge the support of the Director, Defense and Research Organization

(DRDO) for giving the opportunity to work on the project. The authors acknowledge the

guidance acquired from the project work of Mr. SahilBagat performed in the previous year.

The authors acknowledge the support of their Project advisor Dr. Naveen Kumar for their

continuous guidance and support throughout the semester.

Lalit Aggarwal Gayathri Lakshmi Kulukuru

P2009ME1085 P2009ME1062

SMMEE, IIT Ropar SMMEE, IIT Ropar

~ 3 ~

ABSTRACT

The project which was given to us addressed the investigation of the technical, aesthetic, and economic feasibility of deploying special noise barrier application into baffle range of DRDO.

So we have to suggest a most suitable design of noise and maximum sound attenuation can be achieved under having some constraints like cost, weight and aesthetic requirements.

Sound attenuation in noise barrier depends on various parameters like (height, type of material, top surface modification, receiver or source position and several other parameters, barrier shapes).

There is a threefold process in which we are going to do this project.

1. Literature Survey and Theory.2. Modeling and calculating Results.3. Make prototypes and Verifying Results.

In Literature Survey in which there a cumulative study about the past design, all the research done in the field and all of the different possibility is there in designing. In order to understand the physics behind it thorough study about the theory of acoustics design is also required.

In Modeling, Using Sysnoise software the problem will be modeled and result will be obtained and compared with the theoretical results. Simple barrier can be solved analytically using empirical relations but advanced barrier needs BEM or Some kind of simulation software (SYSNOISE). The simulation will help in estimating the behavior of the noise barrier under different parameters e.g.- changes in the barrier parameter, change in material and other parameters.

In making Prototype and verifying results, there is series of experiments have to be performed to see the practical behavior of the parameters involved in the formulation of the problem which is performed using microphone and other devices.

Then all the results will be analyzed and final design will be made.

In this report we continued our work after internship work from literature survey to modeling in matlab and taking physical readings from existing wall and on wooden barrier. Then we compared this reading with problem formulated in Matlab and with software output. Software we used was Oliver Lab Terrain.

~ 4 ~

TABLE OF CONTENTS

ABSTRACT..............................................................................................................................4

Problem statement and motivation...........................................................................................7

INTRODUCTION......................................................................................................................7

REVIEW OF PAST WORK......................................................................................................8

Conclusion from literature...............................................................................................12

Work done in this semester....................................................................................................13

Experiment work.................................................................................................................13

MATLAB Programming.......................................................................................................15

MODELLING IN OLIVE LAB TERRAIN..............................................................................18

Comparison of results with experimental results................................................................22

Future Work...........................................................................................................................23

References.........................................................................................................................23

~ 5 ~

LIST OF TABLES & FIGURES

Table 1 Literature related to insertion losses in a barrier sorted based on the parameters being varied..............................................................................................................................9

Figure 1 Noise barriers with various shape edges and surface conditions 10

Figure 2 Schematic of the experiment performed.................................................................15Figure 3 Experimental Setup for the soft ground..................................................................15Figure 4 Experiment setup for hard ground..........................................................................16Figure 5 Definitions of symbols used to determine Fresnel Number ‘N’...............................17Figure 6 Paths considered in Lam Method...........................................................................18Figure 7 Flowchart of the matlab code..................................................................................18Figure 8 Matlab GUI..............................................................................................................19Figure 9 Comparison of insertion losses with change in height at different frequencies......20Figure 10Comparison of insertion losses with frequency variation at different heights.........21Figure 11 Variation of thickness.............................................................................................22Figure 12 a) single panel barrier b) double panel barrier...........................................22Figure 13 IL variation in single and double panels (frequency=1000hz)................................23Figure 14 IL variation in single and double panels (frequency=2000hz)................................23Figure 15Comparision of matlab and software results with experimental results at 500 Hertz...............................................................................................................................................24Figure 16 Comparisonof MATLAB and software results with Experimental results at 1000 Hertz.......................................................................................................................................24

~ 6 ~

PROBLEM STATEMENT AND MOTIVATION

Specific Aim: To design an effective and cost-efficient noise barrier for the baffle range

which can reduce the unwanted noise up to the required level.

Usually firing ranges where soldiers are trained for the shooting at different conditions are

made out of the city so that there was no disturbance to the local residents. But some of the ranges are

within the residential areas where common people feel a lot of noise. So, to get the control over this

type of noise a noise barrier need to be designed around the range to reduce the noise up to a desirable

level.

INTRODUCTION

Increasing noise pollution will lead to an ever increasing need to control noise of all forms. Noise barriers are the most common solution for controlling noise from surroundings and several methods have been developed for improving their efficiency without increasing their height.

Examples of already deployed noise barriers in India:

1. Noise Barriers in BKC, Mumbai (2010)

2. IIT Powai noise barrier (2012)

3. Commonwealth Games noise barrier (2010)

4. Sound barriers at Suman Nagar, Navghar flyovers (2012)

The vast majority of these have been vertical, reflective wall made of concrete, wood or steel. The standard top for these walls is a “knife-edge”, providing a single diffraction edge with a reflective diffraction zone.

Clearly, there are many other options for noise barrier shapes than vertical reflective walls with knife-edge diffraction zone. In addition to it, there are option to make barriers absorptive, to displace the diffraction zone through the use of a slanted section on top, or to provide for a double- diffraction zone through the use of a T-top section and other modified tops of the walls.

~ 7 ~

~ 8 ~

REVIEW OF PAST WORK

During summer internship most of the literature work was performed and all the work is presented in the tabular form in table 1 and table 2. During this time also complete study was performed and the entire phenomenon was studied.

Different types of barrier that were studied are:

- Shaped Barrier- Conventional Barrier- T-shaped barrier- Multiple Edge Barrier- Arrow shape Barrier

Figure 1 illustrates various noise barriers obtained by varying the shape edges and surface conditions.

FIGURE 1: NOISE BARRIERS WITH VARIOUS SHAPE EDGES AND SURFACE CONDITIONS

~ 9 ~

TABLE 1 LITERATURE RELATED TO INSERTION LOSSES IN A BARRIER SORTED BASED ON THE PARAMETERS BEING VARIED

Different barrierParameter

Simple T-top y-top Cylindrical top

Wedge shaped

Multiple edge

Angled edge

Barrier height D. C. HOTHERSALL, S. N. CHANDLER-WILDE AND M. N. HAJMIRZAE(1990)

Takashi Ishizuka*, Kyoji Fujiwara (2003)

David J. Oldham , Christopher A. Egan (2011)

P.J. Thorsson (2003)

Change in source position


D J Oldham and C A Egan(2009)

Receiver height

D. C. HOTHERSALL, S. N. CHANDLER-WILDE AND M. N. HAJMIRZAE(1990)



Change in receiver position

A. Muradali and K. R. Fyfe (1994)


David J. Oldham , Christopher A. Egan (2011)D. C. Hothersall, D. H. Crombie & S. N. Chandler-Wilde

C.A. Egan, V Chilekwa and D. J. Oldham(2006)

D. C. HOTHERSALL, S. N. CHANDLER-WILDE AND M. N.



~ 10 ~

HAJMIRZAE(1990)

Width of top David J. Oldham , Christopher A. Egan (2011)

D. C. Hothersall, D. H. Crombie & S. N. Chandler-Wilde


D. H. Crombie, D. C. Hothersall& S. N. Chandler-Wilde (1994)C.A. Egan, V Chilekwa and D. J. Oldham(2006)

Cap depth D. C. Hothersall, D. H. Crombie & S. N. Chandler-WildeP.J. Thorsson (2003)

TomonaoOkuboa) and Kyoji Fujiwara

D. H. Crombie, D. C. Hothersall& S. N. Chandler-Wilde (1994)

Angle variationD. C. HOTHERSALL, S. N. CHANDLER-WILDE AND M. N. HAJMIRZAE(1990)

D J Oldham and C A Egan (2009)

Comparison with T-top


D. C. HOTHERSALL, S. N. CHANDLER-WILDE AND M. N. HAJMIRZAE (1990)

D. C. Hothersall, D. H. Crombie & S. N. Chandler-Wilde(1990)



D. C. HOTHERSALL, S. N. CHANDLER-WILDE AND M. N. HAJMIRZAE (1990)



barrier surfacevariation

David J. Oldham , Christopher A. Egan (2011)D. C.


P.J. Thorsson

D. C. HOTHERSALL, S. N. CHANDLER-

D. H. Crombie, D. C. Hothersall& S. N.

D J Oldham and C A Egan (2009)

~ 11 ~

HOTHERSALL, S. N. CHANDLER-WILDE AND M. N. HAJMIRZAE(1990)

(2003) WILDE AND M. N. HAJMIRZAE(1990)

Chandler-Wilde(1994)

top surface absorption variation





M.R. Monazzam, Y.W. Lam(2006)

D. C. Hothersall, D. H. Crombie & S. N. Chandler-Wilde (1990)


D. H. Crombie, D. C. Hothersall& S. N. Chandler-Wilde(1994)



Ground surface variation

Muradali and K. R. Fyfe (1994)



P.J. Thorsson (2003)

Perforated sheet on diffuser

MahdiyehNaderzadeh a,⇑, Mohammad Reza Monazzam b, ParvinNassiri b, SamanehMomenBellahFard (2011)

QRD on top M.R. Monazzam,

~ 12 ~

Y.W. Lam(2006)MahdiyehNaderzadeh a,⇑, Mohammad Reza Monazzam b, ParvinNassiri b, SamanehMomenBellahFard(2011)

Soft TomonaoOkuboa) and Kyoji Fujiwara

Reflective top D. H. Crombie, D. C. Hothersall& S. N. Chandler-Wilde(1994)


CONCLUSION FROM LITERATURE

Going through all the research papers and by seeing their comparative study, it was vaguely suggested that T-type barrier having absorptive coating suits best to get maximum sound abatement.

Since for a single barrier height/cost ration peaks at 3meter so height of T-top barrier should be taken 3meter and width of the top can be taken as 1 meter so that aesthetically is looks good and having maximum sound abatement.

Till internship the project was not having any practical touch and in this research paper experiments were conducted and results was compared to get the real and practical touch

~ 13 ~

WORK DONE IN THIS SEMESTER

EXPERIMENT WORK

SCHEMATIC

FIGURE 2: SCHEMATIC OF THE EXPERIMENT PERFORMED

Figure 2 shows the basic schematics of the experiments that were performed and it shows various parameter that is involved during calculations.

EXPERIMENT ON SOFT GROUND FOR AN EXISTING WALL

~ 14 ~

FIGURE 3 : EXPERIMENTAL SETUP FOR THE SOFT GROUND

Figure 3 shows the experiment that was performed on an existing wall at the fuel-zap in IIT Ropar to get the basic insight of the insertion loss values obtained. The material of the wall is concrete with a height of 170 cm and thickness of 22 cm. The parameters that are varied in this experiment are frequency (varied between 100-1000Hz at an interval of 100 Hz) and receiver distance (varied between 3-15m from the wall) while keeping all other parameters constant. The source height and receiver height are kept in the shadow region with values of 50 cm and 100 cm respectively.

EXPERIMENT ON HARD GROUND

FIGURE 4EXPERIMENT SETUP FOR HARD GROUND

The experiment was performed on a finite barrier on hard ground to enable comparisons as most of the analytical solutions are for hard ground. Figure 4 shows the experimental setup of the experiment. The material of the barrier is wood with a height of 90 cm, width of 108 cm and thickness of 2 cm. The parameters that are varied in this experiment are frequency (varied between 100-1000Hz at an interval of 100 Hz) and receiver distance (varied between

~ 15 ~

0-16m from the barrier) while keeping all other parameters constant. The source height and receiver height are kept in the shadow region with values of 50 cm and 83 cm respectively.

~ 16 ~

MATLAB PROGRAMMING

THEORY: LAM’S METHOD USING MAEKAWA’S CURVE

Maekawa introduced an empirically based diffraction model that provides the insertion loss due to a thin-walled barrier in terms of the Fresnel number.Maekawa then suggested that the insertion loss for a finite-length barrier could be determined by multiple application of this curve to the diffraction paths around the barrier and then summing the energy contributions of these paths.

Maekawa’s curve can be represented by the following two equations:

M={20 log2 π √ N

2

tanh π √ N2

,∧N<1

10 log (20 N ) ,∧N ≥ 1

where N is Fresnel Number given by

N=2λ

( A+B−d )

Where (A+B-d) is the path difference and λ is the wavelength. The symbols are defined as shown in the Figure 5.

FIGURE 5 DEFINITIONS OF SYMBOLS USED TO DETERMINE FRESNEL NUMBER ‘N’

Lam improved on Maekawa’s method by summing complex pressures, instead of energies, of each diffraction path around the barrier. This was done to incorporate the phase interaction

~ 17 ~

and interference between the paths, the absence of which, Lam suggested, was the cause of the poor agreement between Maekawa’s method and experimental results.A semi-infinite barrier is equivalent to a 2D geometry. Diffraction Paths considered for a semi-infinite thin barrier are as shown in Figure 6.

FIGURE 6 PATHS CONSIDERED IN LAM METHOD

The barrier insertion loss is given by:

IL=M 1+10 log

1+2d0

dr

cos π N r+( d0

dr)

2

∑i=1

4

∑m=1

410−M i/20

10−M 1 /20

10−M m /20

10−M 1/20 cos [π ( N i−Nm ) ]

Mi represents the insertion loss value from Maekawa’s curve for the ithpath. The subscript ‘o’ refers to the direct path (from the source to receiver) and the subscript ‘r’ refers to the ground reflected path (from the source image to receiver).

The Lam method fell short when the receivers were in the proximity of the line-of-sight, and when parallel geometries in 2D were considered. This is due to the fact that this method does not predict a unique phase shift at the diffraction edge for each path.

GUI IMPLEMENTATION

~ 18 ~

FIGURE 7: FLOWCHART OF THE MATLAB CODE

A MATLAB GUI has been made for calculating the insertion loss. It includes analysis for an input data.

A GUI (graphical user interface) allows users to perform tasks interactively through controls such as buttons and sliders. Within MATLAB®, GUI tools enable you to perform tasks such as creating and customizing plots, fitting curves and surfaces, and analyzing and filtering signals.

Figure 8 shows the typical GUI that was modeled in the Matlab using Lam’s equation. Parameters that were involved in GUI are barrier height, Source and receiver height, source and receiver distance, and frequency of the sound.

It also shows the graphical variation of the variation in one parameter by taking 5 other parameter constant and vary 6th one.

~ 19 ~

FIGURE 8 MATLAB GUI

MODELLING IN OLIVE LAB TERRAIN

ABOUT THE SOFTWARE

THEORY

The acoustic calculations are made by the software based on Hadden& Pierce Diffraction 3D model implemented with finite impedances faces using Salomons semi-analytical method including ground effects. Multiple barrier diffraction is calculated in a recursive way at any diffraction order.Ground effect is included using the One Parameter Theory of Chessell based on Delany and Bazley.

LIMITATIONS

The thickness of the barrier cannot be reduced to a value less than 3cm. The numerical values of the readings are available only at octave and 1/3rd octave frequencies. Any other parameter except frequency cannot be varied in the same model.

~ 20 ~

ANALYSIS MADE

VARIATION OF HEIGHT

PARAMETERS

Barrier Material: mineral wool

Barrier type: Thin Barrier

Receiver Height= 50 cm

Source Height= 50cm

Source distance=5m

0.5 1 1.5 2 2.5 3 3.5 4 4.50

5

10

15

20

25

30

35

1000 Hz

1995 Hz

3981 Hz

Barrier height(in m)

Inse

rtio

n L

oss(

in d

B)

FIGURE 9: COMPARISON OF INSERTION LOSSES WITH CHANGE IN HEIGHT AT DIFFERENT FREQUENCIES

~ 21 ~

2 2.2 2.4 2.6 2.8 3 3.2 3.4 3.6 3.8

-5

0

5

10

15

20

25

30

35

h=1m

h=2m

h=3m

h=4m

log(frequency(in Hz))

Inse

rtio

n L

oss(

in d

B)

FIGURE 10: COMPARISON OF INSERTION LOSSES WITH FREQUENCY VARIATION AT DIFFERENT HEIGHTS

It can be seen that the insertion loss increases with increase of barrier height. From the figure 7, it can be seen that the variation of the insertion loss values is not much when it is changed from a height of 3m to 4m.

VARIATION OF THICKNESS

MASS LAW

When sound is incident upon a wall or partition, some of it will be reflected and some will be transmitted through the wall. The transmission loss obtained can be determined using mass law at a particular frequency.

TL=20∗log (m)+20∗log( f /100)−7

Where m=mass density and

f=frequency

PARAMETERS

Barrier Material: mineral wool

Flow resistivity=20000 Pas/m2

Barrier Height = 2m

Receiver Height= 1 m

Source Height = 1 m

Source distance=2m

Frequency=1500 Hz

~ 22 ~

0 10 20 30 40 50 60

-5

0

5

10

15

20

25

30

thickness_3cmthickness_3.5cmthickness_4cmthickness_4.5cm

Receiver distance (in m)

Inse

rtio

n L

oss(

in d

B)

FIGURE 11 VARIATION OF THICKNESS

From figure 8 it can be clearly seen that the value of insertion loss doesn’t vary much with the receiver distance for thickness values greater than the optimum value calculated from the mass law (≈2.5cm).

SINGLE VS DOUBLE PANELS

FIGURE 12 A) SINGLE PANEL BARRIER B) DOUBLE PANEL BARRIER

PARAMETERS

Barrier Material: mineral woolFlow resistivity 20000 Pas/m2Barrier Height 2mBarrier Thickness 20cm single

10 cm doubleReceiver Height 1 mSource Height 1 mSource distance 2m

~ 23 ~

1 2 3 4 5 6 7 8 9 10 110

5

10

15

20

25

30

35

40

45

single panel

double panel_gap20cm

double panel_gap 30cm

double panel_gap50cm

Receiver distance(in m)

Inse

rtio

n L

oss(

in d

B)

Frequency = 1000 Hz

FIGURE 13 IL VARIATION IN SINGLE AND DOUBLE PANELS (FREQUENCY=1000HZ)

1 2 3 4 5 6 7 8 9 10 110

5

10

15

20

25

30

35

40

45

single paneldouble panel gap 20 cmdouble panel_gap 30cmdouble panel_gap50cm

Receiver distance (in m)

Inse

rtio

n L

oss(

in d

B)

Frequency = 2000 Hz

FIGURE 14 IL VARIATION IN SINGLE AND DOUBLE PANELS (FREQUENCY=2000HZ)

From the figures 13 and 14 it can be clearly seen that double panel barriers are more effective in sound reduction compared to single panel barrier and the insertion loss increases with increase in gap between the two panels.

~ 24 ~

COMPARISON OF RESULTS WITH EXPERIMENTAL RESULTS

PARAMETERS

Barrier Material: woodBarrier Height=90 cmBarrier Width=105 cmBarrier Thickness= 2cmReceiver Height= 80 cmSource Height= 50 cm

FIGURE 15 COMPARISION OF MATLAB AND SOFTWARE RESULTS WITH EXPERIMENTAL RESULTS AT 500 HERTZ

FIGURE 16COMPARISONOF MATLAB AND SOFTWARE RESULTSWITH EXPERIMENTAL RESULTSAT 1000 HERTZ

From the figures 15 and 16 it can be seen that the results from the software are in coherence with those of the experimental results within acceptable error limits. It can also be seen there is a considerable variation in the results obtained from MATLAB program, the reasons of which can be attributed to the assumptions made in the theoretical model, where it considers

~ 25 ~

the barrier is semi-infinite with negligible thickness and which doesn’t include the effect of the material of the barrier.

~ 26 ~

FUTURE WORK

Some part of the project is completed in B.Tech-1 project which was in this semester and this project will be continued in coming semester also as B.Tech-2 project.

The things which are planned for coming semester are:

1. In current work, Material selection for the barrier was not suggested so work will be done in this context in the upcoming semester.

2. In current work, output from the software of single and double panels was compared but there physical modelling was not done. It will be includedin the further studies.

3. The software which is presently used has certain limitations due to which one cannot vary the shape of the barrier. The new software SYSNOISE, which is BEM/FEM software, was purchased for getting more accurate results and help in physical modelling of the system.

4. In current study the MATLAB formulations are donefor a semi-infinite, thin barrier. In 2nd part the more focus will be made in including parameters like thickness, finiteness, etc.

5. In our next work formation of the optimization problem will be included and will be solved using different optimization techniques.

REFERENCES

Web site reference

1. http://sciencedirect.com/2. http://www.acoustax.com/noise-barrier-specs.php3. http://www.acousticalsurfaces.com/wall_barrier/wall_barrier.htm4. http://www.nrc-cnrc.gc.ca/eng/ibp/irc/bsi/85-sound-tranmission.html5. http://articles.timesofindia.indiatimes.com/2012-05-17/mumbai/31747986_1_noise-

barrier-noise-levels-sumaira-abdulali6. http://www.nrc-cnrc.gc.ca/eng/ibp/irc/bsi/85-sound-tranmission.html7. http://www.otlterrain.com/

Research Papers

[1] R.O.Feher, proc.Ann.Nat. Noise Abatement Symp.,1951,p-98[2] A.Muradali and K.R Fyfe,A study of 2d and 3d barrier insertion loss using

improved diffraction based methods, applied acoustics,vol.53,no-1-3,pp 49-75,1998

~ 27 ~

[3] D. C. HOTHERSALL, S. N. CHANDLER-WILDE AND M. N. HAJMIRZAE, Efficiency of Single Noise barrier, Journal of Sound and Vibration (1991) 146(2), 303-322.

[4] David J. Oldham , Christopher A. Egan.. A parametric investigation of the performance of T-profiled highway noise barriers.. Applied Acoustics 72 (2011) 803–813

[5] C.A. Egan, V Chilekwa and D. J. Oldham, Top edge treatment to enhance the performance of a noise,Acoustics Research Unit, University of LiverpoolLiverpool, L69 3BX, United Kingdom

[6] Watts, G.R., Barrier design to reduce road traffic noise. Proceedings of the Institution of Civil Engineers, 2002. 53(2): p. 79- 86.

[7] Takashi Ishizuka, Kyoji Fujiwara,Performance of noise barriers with various edge shapes and acoustical conditions.. Applied Acoustics 65 (2004) 125–141.

[8] MahdiyehNaderzadeh, Mohammad Reza Monazzam, ParvinNassiri, SamanehMomenBellahFard, Application of perforated sheets to improve the efficiency of reactive profilednoise barriers, Applied Acoustics 72 (2011) 393–398.

[9] A. Muradali and K. R. Fyfe, A Study of 2D and 3D Barrier Insertion Loss using Improved Diffraction-based Methods, Applied Acoustics, Vol. 53, No. I-3, pp. 49-15, 1998

[10] Maekawa, Z., Noise reduction by screens. Applied Acoustics, 1968, 1, 157-173.

[11] Pontus J. Thorsson,Combined effects of admittance optimisationon both barrier and ground, Applied Acoustics 64 (2003) 693–711.

[12] D J Oldham and C A Egan, The development of a practical top edge device for a noise barrier, 16th international congress on noise and vibration July 2009.

~ 28 ~

Btp report final_lalit

Documents

project work of

internship work

experiment work

future work

advanced barrier

wooden barrier

barrier shapes

barrier parameter