You are on page 1of 19

Synopsis Presentation

Traffic Noise Modeling using soft-computing techniques

By : Puneet Dhiman

Guide : Dr. S.K. Mangal

Delhi City at Glance


Geographical Area : 1484 sq km Current Population : 14 million Expected Population by year 2021 : 40 % Population density : 9340 people per sq. km. Population density of New Delhi : 4909 people per sq km. Density of population at All-India level : 324 person per sq. km. In 2001 Industries : Thermal Power Plant & Small scale Industries On an average over 1,000 vehicles are added to the Delhi city roads everyday
4 . 8 7

1 3 . 7 1 4

) s n o i l l i M n i ( h t w o r G e l c i h e V

5 . 0 0

) s n o i l l i M n i ( h t w o r .G n p o P

4 . 5 0 4 . 0 0 3 . 5 0 3 . 0 0 2 . 5 0 2 . 0 0 1 . 5 0 1 . 0 0 0 . 5 0 0 . 0 0 0 . 0 1 0 . 0 4 0 . 1 8 0 . 3 4 1 . 8 1 3 . 4 6

4 . 1 6

1 2 9 . 4 1 0 8 6 4 2 0 2 . 6 1 . 7 4 6 . 2

1 9 5 1

1 9 6 1

1 9 7 1

1 9 8 1

1 9 9 1

1 9 5 1

1 9 6 1

1 9 7 1

1 9 8 1

1 9 9 1

2 0 0 1

2 0 0 4

2 0 0 6

2 0 0 1

Y e a r

Y e a r

Vehicular growth

Population growth

CPCB India Noise Limits


Area Code
(A) (B) (C) (D)

Category of Area/Zone
Industrial area Commercial area Residential area Silence Zone

Limits in dB(A), Leq Day Time 75 65

Limits in dB(A), Leq Night Time 70 55

55
50

45
40

Day time shall mean from 6.00 a.m. to 10.00 p.m. Night time shall mean from 10.00 p.m. to 6.00 a.m Silence zone is an area comprising not less than 100 metres around hospitals, educational institutions, courts, religious places or any other area which is declared as such by the competent authority
Source: CPCB website - http://www.cpcb.nic.in

Noise levels
SPL sound pressure level is a logarithmic measure of the effective sound pressure of a sound relative to a reference value. It is measured in decibels (dB) above a standard reference level. The standard reference sound pressure in air or other gases is 20 Pa, which is usually considered the threshold of human hearing (at 1 kHz).

Equivalent sound pressure level(Leq) Leq represents the equivalent energy sound level of a steady state and invariable sound. It includes both intensity and length of all sounds occurring during a given period.

Continue...
L10
The sound level is exceeded 10 percent of the time. This is a measure of the louder sound levels during the measurement period. Example: During a 1-hour measurement, an L10 of 85 dBA means the sound level was at or above 85 dBA for 6 minutes.

L90

The sound level is exceeded 90 percent of the time. This is a measure of the nominal background level. Example: During a 1-hour measurement, an L90 of 50 dB(A) means the sound level was at or above 50 dBA for 54 minutes.

Weighted Sound Pressure Levels


A-weighting is applied to instrumentmeasured sound levels in effort to account for the relative loudness perceived by the human ear, as the ear is less sensitive to low audio frequencies. It is employed by arithmetically adding a table of values, listed by octave or third-octave bands, to the measured sound pressure levels in dB. The resulting octave band measurements are usually added (logarithmic method) to provide a single A-weighted value describing the sound.

A graph of the A-, B-, C- and D-weightings across the frequency range 10 Hz 20 kHz

Noise annoyance due to various noise sources reported in socio-acoustic survey in Delhi city

N Garg, et al., Passive noise control measures for traffic noise abatement in Delhi, India, Journal of Scientific and Industrial Research, Vol. 71, March 2012, pp. 226-234.

Previous Work
Paper
H. N. Rajakumara & R. M. Mahalinge Gowda Road Traffic Noise Prediction Model under Interrupted Traffic Flow Condition

Year

Work Done
1)Acceleration Lane Model 2)Deceleration Lane Model

Conclusion

2008

1) Correlation coefficient 0.844 2) Correlation coefficient A multiple regression analysis with a 0.889 stepwise approach is used for both This Study provide models different model for different lanes of road.

Sheetal Agarwal, Bajrang L Swami, Akhilendra 2009 Bhushan Gupta Development of a prediction model interrupted traffic conditions: A case for Jaipur city noise under flow study

Adjustment is applied to the FHWA model so that it can fit in Indian road traffic conditions In FHWA Ah= tendency to blow horn is added

R2 value of 0.4321 for the conventional FHWA model and 0.8337 for the modified FHWA model obtained in the study

Continue...
Paper
K. Rawat, V. K. Katiyar, Pratibha Mathematical Modeling of Environmental Noise Impact

Year

Work Done
Basic equation of CRTN model (Lam and Tam 1998) is used and modified by recalculate the coefficient of equation and the constant term using the survey data.

Conclusion
New estimated L dB( A) values gives a R 2 value of 0.9165 against previous value of 0.8733 without modification

2009

V. Tripathi, A. Mittal and P. Ruwali Efficient Road Traffic Noise Model for Generating Noise Levels in Indian Scenario

2012

Prepared a Road Traffic Model(RTM) by modifying an existing model based on Iran roads and comparison of results with CALIXTO model RTM model = 27.931 + + CALIXTO model =10log(Q(1+n*VP/100))

Calixto model R2 0.84290 Efficiency 99.571 Road traffic model R2 0.85778 Efficiency 99.63

Continue...
Paper
Kalai Selvi, R. And Ramachandraiah, A.

Year

Work Done
Horn correction parameter is used

Conclusion

A Model for traffic noise prediction in heterogeneous traffic conditions

2012

Horn noise events occurring around 16 Effect of road width and multiple reflection per minute will raise is accounted the Leq by 12dB (A). Leq=Lm(15) +Chorn +CREF R2 = 0.777 A multiple regression analysis is used to (measured VS form correlation. Predicted correlation)

PROF. VILAS K PATIL, PROF.(DR) PRASHANT P.NAGARALE


Modeling of Traffic Noise in selected area of Nashik City

2013

R2 Value Varies 0.505 To Linear Regression approach is used to From calculate relation between traffic volume 0.623. This show the corelation between and noise levels traffic volume and noise levels.

Limitations Of Previous Work


Previous work done on traffic noise in abroad is not completely applicable due to Heterogeneous Traffic conditions in India. There is no generalized model created yet which can vary according to traffic conditions. Previous models cannot predict the sophisticated Non-linear nature of noise generation.

Benefits Of Proposed Work

Traffic noise prediction and monitoring can be done and its impact can be studied.
The model can be very helpful particularly in cases where there is no definite relationship existing between input and output variables. The Leq is having non-linear relationship with input variables. It can serve as alternative to the analytical models proposed in previous studies.

Proposed work Plan


Experimental observation for noise monitoring.


Analysis of data gathered. Review of ANN architecture. Development of ANN model. Training of ANN model.

Testing and validation of the model.

Select Application Select Neural Network Model

Problem Analysis

Choose Source Of Information

Select and Feedback Data (Pattern Files) And Pre-Process Data. Select The No. Of Hidden Neurons

Prepare Training Sets And Create Neural Network Files

Configure Neural Network Model

Training,Validation and Testing with 80%, 5% And 15% of The Total Data
yes

Problem In Training

yes

Problem In Generalizing

Trained And Tested Successfully Leq Predicted Matched With Experimental data
no yes

Improve Performance

Run And Validate Trained Network


no yes

Stop

Input Parameters

Traffic Flow Traffic Speed

Percentage of heavy vehicles


Distance of measuring device from traffic Road width

Reflection from the building

Output Parameter

Sound Pressure Level

Instrumentation & Software Facilities to be Used

Noise Monitoring Analyzer, B&K 2250 Noise Monitoring Terminal Softwares B&K 2650 for transfer of data to time domain and frequency domain Bushnell Laser speed gun for measuring the average speed of vehicles SPSS version 17 for regression analysis Matlab for ANN modelling

References

H. N. Rajakumara & R. M. Mahalinge Gowda, Road Traffic Noise Prediction Model under Interrupted Traffic Flow Condition,Environ Model Assess (2009) 14:251257 Sheetal Agarwal, Bajrang L Swami, Akhilendra Bhushan Gupta, Development of a noise prediction model under interrupted traffic flow conditions: A case study for Jaipur city, Environ Monit Assess (2011) 172:113120 K. Rawat, V. K. Katiyar, Pratibha, Mathematical Modeling of Environmental Noise Impact, Indian Journal of Biomechanics: Special Issue (NCBM 7-8 March 2009) V. Tripathi, A. Mittal and P. Ruwali, Efficient Road Traffic Noise Model for Generating Noise Levels in Indian Scenario, International Journal of Computer Applications (0975 8887) Volume 38 No.4, January 2012

References

Kalai Selvi, R. And Ramachandraiah, A., A Model for traffic noise prediction in heterogeneous traffic conditions, International Journal of Current Research Vol. 4, Issue, 04, pp.180-184, April, 2012 Prof. Vilas k. Patil, Prof.(DR) Prashant P. Nagarale, Modeling of Traffic Noise in selected area of Nashik City, International Journal of Advanced Technology in Civil Engineering, ISSN: 2231 5721, Volume-2, Issue-1, 2013 Naven Garg, Anil Kumar, Sagar Maji, Significance and implication of airborne sound insulation criteria in building element for traffic noise abatement, Applied Acoustics 74(2013) 1429-1435 CPCB, The Noise Pollution (Regulation and Control) Rules, (2000)

References

Kalai Selvi, R. And Ramachandraiah, A., A Model for traffic noise prediction in heterogeneous traffic conditions, International Journal of Current Research Vol. 4, Issue, 04, pp.180-184, April, 2012 Prof. Vilas k. Patil, Prof.(DR) Prashant P. Nagarale, Modeling of Traffic Noise in selected area of Nashik City, International Journal of Advanced Technology in Civil Engineering, ISSN: 2231 5721, Volume-2, Issue-1, 2013 Naven Garg, Anil Kumar, Sagar Maji, Significance and implication of airborne sound insulation criteria in building element for traffic noise abatement, Applied Acoustics 74(2013) 1429-1435 CPCB, The Noise Pollution (Regulation and Control) Rules, (2000)