基于参数学习的城市道路交通噪声预测与可视化研究

发布时间：2018-04-23 12:40

本文选题：参数学习 + 交通噪声　；参考：《中国民航大学》2015年硕士论文

【摘要】：随着城市化进程的推进和道路交通网的扩张,由城市机动车带来的道路交通噪声问题日益凸显,不仅对道路周边居民的生活和工作带来影响,而且对道路交通事业的蓬勃发展造成阻碍。因此,科学、有效地预测城市道路交通噪声及其对周边区域的影响显得十分迫切。首先,本文分析了城市道路交通噪声的评价指标及其适用范围。介绍了道路交通噪声的特点、危害和影响因素,详细阐述了其常用评价指标的优缺点和适用范围,最终选取LAeq(等效连续A声级)作为本文的评价指标。此外,还阐明了评价道路交通噪声的国家标准。其次,改进FHWA(美国联邦公路局)模型。阐述应用较为广泛的预测模型,分析了模型的影响因子及其适用条件。然后结合我国城市的道路交通特点,选取FHWA模型进行噪声预测,对比分析预测结果,并根据实际监测情况对模型进行修正。再次,提出了基于参数学习的道路交通噪声预测模型。在上述改进模型的基础上,引入参数学习的概念,将机器学习方法用于模型的训练,利用机器学习的自学习性和自组织性进行模型参数的修正。经过对实验结果的分析,发现模型具有较高的准确性和实用性。最后,利用GIS(地理信息系统)相关软件绘制了南京奥体中心和五台山体育场周边道路交通噪声地图。研究了GIS及其在道路交通噪声预测领域的应用,阐述了噪声地图的绘制过程。利用上述基于参数学习的预测模型得到的噪声数据,结合区域交通地理信息基础数据,绘制出了目标区域的道路交通噪声地图,直观地展示了城市道路周边的交通噪声分布情况。
[Abstract]:With the development of urbanization and the expansion of road traffic network, road traffic noise caused by urban motor vehicles is becoming increasingly prominent, which not only has an impact on the life and work of the residents around the road. And to the vigorous development of road traffic cause obstacle. Therefore, it is urgent to effectively predict the traffic noise and its influence on the surrounding areas. First of all, this paper analyzes the evaluation index of urban road traffic noise and its scope of application. This paper introduces the characteristics, harms and influencing factors of road traffic noise, expounds in detail the advantages, disadvantages and applicable scope of its commonly used evaluation indexes, and finally selects LAeq (equivalent continuous A sound level) as the evaluation index of this paper. In addition, the national standards for evaluating road traffic noise are expounded. Secondly, improve the FHWAs model. This paper expounds the widely used prediction model, and analyzes the influence factors and applicable conditions of the model. Then according to the characteristics of urban road traffic in China, the FHWA model is selected to predict the noise, and the prediction results are compared and analyzed, and the model is modified according to the actual monitoring situation. Thirdly, a traffic noise prediction model based on parameter learning is proposed. On the basis of the above improved model, the concept of parameter learning is introduced, and the machine learning method is applied to the training of the model, and the self-learning habits and self-organization of machine learning are used to modify the parameters of the model. Through the analysis of the experimental results, it is found that the model has high accuracy and practicability. Finally, the road traffic noise map around Nanjing Olympic Sports Center and Wutai Mountain Stadium is drawn by using GIS (Geographic Information system) software. In this paper, GIS and its application in road traffic noise prediction are studied, and the process of noise map drawing is expounded. The road traffic noise map of the target area is drawn by using the noise data obtained from the above prediction model based on parameter learning and combining with the basic data of regional traffic geography information. The distribution of traffic noise around urban roads is displayed intuitively.
【学位授予单位】：中国民航大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：U491.91

【相似文献】