Study On The Technology Of GIS Based Road Traffic Noise Prediction

Along with the increasing Traffic mileage of highway of China, traffic noise has broughtserious influence of environmentally sensitive areas. Traffic noise prediction and evaluationhas become an important part of environmental impact assessment of highway projects. Butpre-existing noise prediction models has some defect of low precision and they cannoteffectively evaluate the spatial distribution of traffic noise.This paper focuses on traffic noise prediction theory and model，and research on theapplication of GIS，which has powerful ability of spatial analysis and data processing，ontraffic noise prediction. Then developed a GIS based road traffic prediction system. The mainresearch results and conclusions are as follows：1．Through summarizing the theory of road traffic noise prediction including noisesource，influencing factors，spreading and evaluation index，and comparing some predictionmodels at China and abroad，a conclusion that FHWA model is more scientific and reasonablebut needs to be improved is drew. According to its shortcomings，some parameters arecorrected or added：Through correcting acoustic absorption factor in accordance with noisepropagation under different conditions and considering the vegetation coverage，groundabsorb correction is deduced； From the effect of gradient of vehicles especially large ones，data in report of China Environmental Protection Agency is adopted；Based on the effect ofpavement materials and shelters，correction method of the Ministry of Communications isadopted. J-F model which is based on FHWA is put forward and its condition of applicationand precautions is researched.2．Through the detailed research on the application of GIS in highway management andtraffic noise prediction，the following results are got：(1)Considering the source，means of collection and characteristics of element of highwaydata and the needs of GIS based query，Classification system of Highway information isproposed. The information could be divided into several sets. Each set has similar categoriescharacteristics and spatial characteristics，which contains rich attribute information. Thismethod could help manage the highway information more effectively.(2) Based on the spatial distribution characteristics of road noise data and the method of the development of GIS，a method for GIS based traffic noise prediction system is putforward. Through analyzing development styles of applied GIS，Components GIS is selectedto development the road traffic noise prediction system，which makes the most of powerfulspatial database management ability of GIS components and flexibility of object-orientedlanguage. From practical applicability，advancement and difficulty of the system，and throughdeeply studying of software development tools，ArcMap is used to vectorize the spatial dataand putting attribute data in storage. Meanwhile，ArcCatalog is utilized to manage data.Development tools include VB.NET under Visual Studio2008platform， ArcEnginecomponent and Microsoft Access DBMS. This development style makes the best of theinfrastructure of ArcGIS in GIS development and reduces difficulty and costs and makes thesystem development process more convenient and efficient.3．The system general design and module design is done considering user’s requirements.Then system interface custom and integrated secondary development are carried out. Utilizingattribute and event of controls in ArcEngine，file management function，map view functionand Spatial query are achieved，which makes it convenient for users to query and managehighway facilities and noise data； As well，noise prediction calculation is achieved based onJ-F model and data access technology ADO.NET，then Contour map is drew adoptingVB.NET and surfer8.0interface，which makes noise prediction precisely and evaluate resultvisibility.Research result of this paper provides theoretical foundation for improving accuracy oftraffic noise prediction. It also offers technical reference for the application of GIS in highwaymanagement and traffic noise prediction.