CREATING DYNAMIC MAPS OF NOISE THREAT USING PL-GRID INFRASTRUCTURE Maciej Szczodrak 1, Józef Kotus 1, Bożena Kostek 1, Andrzej Czyżewski 2 1 Academic Computer Center TASK, Gdansk University of Technology, Narutowicza 11/12, 80-233 Gdansk, Poland 2 Multimedia Systems Department, Gdansk University of Technology, Narutowicza 11/12, 80-233 Gdańsk, Poland grid@sound.eti.pg.gda.pl
Outline Introduction PLGrid Plus Project Methodology Reverse ingeneering Results Conclusion
Introduction - motivation Noise annoyance Noise Induced Hearing Loss
PLGRID PLUS PROJECT www.plgrid.pl
Plgrid Plus Project Domain-oriented services and resources of Polish Infrastructure for Supporting Computational Science in the European Research Space Most important task is preparation of specific computing environments so called domain grids i.e., solutions, services and extended infrastructure (including software), tailored to the needs of different groups of scientists. 13 groups of users: AstroGrid-PL, HEPGrid, Nanotechnologies, Acoustics, Life Science, Chemistry and Physics, Ecology, SynchroGrid, Energetics, Bioinformatics, Health, Materials, and Metallurgy. www.plgrid.pl
Plgrid Plus Project 576 TFlops, HDD 5.58 PB, 41 248 cores, RAM 113.26 TB www.plgrid.pl
METHODOLOGY
Methodology Noise source part Propagation part
METHODOLOGY NOISE SOURCE PART (ROAD NOISE)
Road noise source model Road as line source consists of 2 submodels: The vehicle model, describing the sound power of single moving vehicle The traffic model, combining the noise emission of numerous single vehicles into the sound power per meter length of the line source Two main sources of vehicle s noise taken into account in mathematical model: interaction between wheel and road surface vehicle s propulsion (engine, transmission, exhaust)
Road noise source model Category Name Description 1 2 Light motor vehicles Medium heavy vehicles Passenger cars, delivery vans <3.5t, SUV's, MPV's including trailers and caravans Medium heavy vehicles, delivery vans >3.5t, buses, touring cars, etc. with two axles and twin tire mounting on rear axle Vehicle category according to the EU/ECE type approval M1 and N1 M2, M3 and N2, N3 3 Heavy vehicles Heavy duty vehicles, touring cars, buses with three or more axles M2 and N2 with trailer, M3 and N3 4 Powered two-wheelers 4a motorcycles, tricycles or quads, engine <50 ccm 4b motorcycles, tricycles or quads, engine >50 ccm L3, L4, L5, L7
Road noise source model
METHODOLOGY PROPAGATION PART
Methodology on the basis of Harmonoise/Imagine projects Industrial sources Airport sources Propagation classes Railroad sources PROPAGATION L den L night Road Traffic sources P2P model Road properties Traffic data Meteorological conditions Geometry (surroundings buildings)
Propagation model Based on Point to Point Harmonosie model The acoustic ray tracing method Additional libraries: Harmonoise, CGAL, Tardem Geometrical description of sources and buildings Meteorological conditions
Ray tracing method
Parallel processing PREPARE DATA (MASTER) PARALLEL PROCESSING START Read input data Spread data to all cores All cores received task? YES Worker returned output? YES All tasks calculated? NO NO NO Send next task to worker core Wait for workers output Send next task to worker core The usage of MPI* Master core distributes tasks to the others In the beginning master sends 1 task to each core When worker core finishes computations, sends message with outcome Master receives outcome and sends next task to worker End, when all tasks are processed YES Format and save output data STOP *MPI (Message Passing Interface) - is a specification for an API that allows many computers to communicate with one another. It is used in computer clusters and supercomputers.
Service usage Input data edition Input data source A. User interaction with system Task submission Access node QCG / UNICORE Task execution PLGrid Clusters Outcome analysis and visualization
Service usage Input data edition Input data source A. User interaction with system B. Dynamic noise map creation Monitoring station Task submission Access node QCG / UNICORE Task execution PLGrid Clusters Outcome analysis and visualization Monitoring station Monitoring station Interne t System database Supercomputer HTTP Server Background map server Visualizing application System website Smartphone PC PDA Laptop Remote sensors Server part User part
NOISE MAPPING IN GRID COMPUTING EXAMPLE
Evaluation of the results Noise Prediction Model (Grid calculation) Noise map calculated using CadnaA software Difference map calculation > -6.0 d B > -4.5 d B > -3.0 d B > -1.5 d B > 1.5 d B > 3.0 d B > 4.5 d B > 6.0 d B NPM (Grid) CadnaA software Difference map
N [%] Evaluation of the results Histogram of difference of noise levels in large area (338,099 number of points) 25 20 15 10 5 0-12 -6-5 -4-3 -2-1 0 1 2 3 4 5 6 12 >12 ΔL [db]
Scenario Calculation time results Scenario grid raster [m] points cores time [s] 1 16 16 256 32 1908 2 8 8 1024 32 7452 3 4 4 3969 64 16974 4 2 2 15625 64 70895 Time [s] 0 1 2 3 4 5 6 7 8 1 2 3 4
REVERSE INGENEERING
REVERSE INGENEERING Typically, if we want to calculate the noise map for a given area, we need the input data for the noise source model. In the considered case, only road noise sources were taken into account. In consequence, we need information about: number of vehicles per hour, type of road surface, type of vehicle, vehicle speed. The employed noise monitoring station did not deliver data about the road noise parameters. The missing data are calculated on the basis of the measured noise level.
REVERSE INGENEERING The reverse engineering technique is applied for this purpose. We assume that monitoring stations measure noise the main source which derives from the nearest road. To get the input data for the noise source model we need to calculate the number of vehicles on the basis of the measured noise level. Other factors of road noise source remain constant.
Localization of measurement stations (Gdańsk)
Measurement station L free L near 3 L 6 facade
Localization of experiment measurement stations
ALGORITHM Noise source Acoustic pressure Noise monitoring station L Aeq,T Reverse function Traffic flow Noise prediction model Immision levels Dynamic noise map
RESULTS
Input data LAeq,1h [db] 75 NMS 122 NMS 139 NMS 143 NMS 169 70 65 60 55 50 45 40 35 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 Time [h] Selected noise measurement results obtained by noise monitoring stations
Reverse function L Aeq [db] 70 65 Road 186 Traffic flow 10000 1000 NMS 122 y = 9.30E-23x 1.37E+01 R² = 9.98E-01 60 55 50 100 10 45 1 40 10 100 Traffic flow 1000 0.1 40 45 50 55 60 65 L Aeq [db] 70 Reverse function calculation for road no. 186 (Noise Monitoring Station no. 122)
Traffic flow Traffic flow Road 186 Road 189 Road 422 1000 900 800 700 600 500 400 300 200 100 0 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 Time [h] Traffic flow calculated for selected roads.
Dynamic Noise Map
CONCLUSIONS
CONCLUSIONS As a result of the described work, the system for dynamic noise maps calculation employing supercomputing grid and sensor network was practically implemented, and tested. The implementation of the software for the traffic flow determination on the basis of acoustic climate measurements was performed. Use of hardware devices for traffic flow measurements, which are currently being installed in the city, would help to achieve precise source model parameters.
Acknowledgements This research has partially been supported by the European Regional Development Fund program no. POIG.02.03.00-00-096/10 as part of the PLGrid PLUS project and E! 3266 EUROENVIRON WEBAIR project