The Research On Comparison Between Road Traffic Noise Prediction Models And Monitoring Results

While lots of countries developed road traffic noise prediction models based on their own reality, China also released several traffic noise prediction models in the form of standard and guideline. The specification model of version1996and the specification model of version2006all have their own calculation formula on the single traffic sound pressure level but the noise guideline model of version2009. So domestic environmental impact assessment institutions are free to calculate the single traffic sound pressure level after the noise guideline model of version2009released, which results in the doubts about the noise prediction results and the feasibility and necessity of the noise pollution prevention and control measures by experts,especially the environmental protection investment become a more focused problem.Based on it, the article compares the noise prediction results with the monitoring values, gets the erros between them, and then give suggestions on the parameters selection and prediction results of several common noise prediction models.The article compares the advantages and disadvantages with the applicable conditions among the several common prediction models, then chooses eight representative roads in Guangzhou、Foshan and Zhongshan to set a series of noise monitoring sites which along the lines that are perpendicular to the roads. At the same time, the traffic flow of types of vehicle、the vehicle speed、the environment parameters around the monitoring sites and the meteorological datas are recorded.It is concluded that urban roads are more of difference from highway. Firstly, the propotion of large cars and medium-sized cars on the urban roads is almost less than10precent, when the propotion of large cars on highway in the night is bigger than that in the daytime.Secondly, the noise frequency of highway is high frequency and the noise frequency of urban roads is low frequency due to the vehicle speed.Thirdly, the traffic flow、the vehicle speed and the proportion of types of vehicle play a common role on the traffic noise results,and the traffic noise increases along with the increase of the three factors,but it does not meet the linear relationships.Then, the article analyses the errors between the monitoring values and the prediction results of five designed prediction parrerns, and finds that the errors between the monitoring values and the prediction results are more bigger when the design speeds are used in the prediction patterns; however, the errors between the monitoring values and the prediction results are more close when the actual vehicle speeds or the vehicle speeds calculated by the specification model of version2006used in the prediction patterns. Besides, when the traffic flow is considered in the distance attenuation coefficient, the prediction results are more close to the monitoring values. Cadna/A software is usually used to draw noise isoline, the attenuation trend of prediction results is very close to that of monitoring values, especially in the sites far from the road.Finally, the article suggest that the noise guideline model of version2009and the specification model of version2006can be combined to predict the traffic noise while the vehicle speed calculated by the specification model of version2006or adopted by the measured vehicle speed, the single traffic sound pressure level calculated by the specification model of version2006,with the distance attenuation coefficient considered of the traffic flow and other attenuation terms calculated by the noise guideline model of version2009.