Model And Algorithms For Estimating Fuel Consumption And Emissions On Urban Road Network Based On Multi-Source Data

With the socio-economic development and the acceleration of urbanization process, the motor vehicle ownership has been growing rapidly, not only causing serious traffic congestion, but also generating huge traffic fuel consumption and severe environmental pollutions. Reducing the traffic fuel consumption and mitigating the resulting environmental pollutions has become one of the most urgent urban problems. To this end, a key issue is the accurate quantification of the traffic fuel consumption and emissions on urban road network. In regards to the estimation of traffic fuel consumption and emissions, ample data sources exist in various public agencies related to traffic and transportation, environmental protection, and public security. However, due to the gaps between multiple disciplines and the functional division of different public agencies, these data sources have not been integrated systematically, thus have not been utilized effectively. Therefore, this thesis aims to analyze and utilize comprehensively the multi-source data related to fuel consumption and emissions, and then establish the estimation methods of traffic fuel consumption and emissions on urban road networks.First, the thesis reviews the state-of-the-art research and development on the estimation of traffic fuel consumption and emissions on urban road network., It analyzes the similarities, differences, and limitations of the estimation methods used in different disciplines from perspectives of the need of data sources, estimation of vehicle mileage traveled, type of fuel consumption and emission parameters, and the objective of fuel consumption and emission estimations.Then, by using Beijing as the case city, the thesis analyzes the availability of existing data sources and their characteristics on the data of urban road traffic flow, speed, emission factor, emission rate, fleet composition, and vehicle information from an interdisciplinary viewpoint of traffic and environment. After the analysis of data sources, the study establishes a methodology of estimating traffic fuel consumption and emissions on road networks which integrates the use of data sources located in multi-functional areas. In the proposed methodology, fuel consumption and emissions of individual vehicle is first calculated by using the Vehicle Specific Power (VSP) distribution as the intermediate variable to link floating car data and emission rate data. After the emission estimation of the individual vehicle, fleet composition data are obtained by integrating the video-based license plate recognition data with vehicle information database. Further, the link-based fuel consumption and emissions are computed by combining the data sources including traffic flow, road length, fleet composition, and individual fuel consumption and emissions. In light of the problem regarding the incomplete data coverage of license plate recognition and traffic flow data, the thesis proposes a sampling expansion method for traffic flow and fleet composition data which is characterized by regional locations and road classes. As a result, the fuel consumption and emissions on the whole road network can be estimated based on the link level fuel consumption and emissions.Finally, the proposed method is applied to quantify the traffic fuel consumption and emissions on urban road network in Beijing. The characteristics and regularities of temporal-special distribution of fuel consumption and emissions are analyzed from the aspects of time period, regional location, and road classes. The proposed methodology of emission estimation and analysis is able to support the development of energy saving and emission reduction strategies.