基于多源数据的货车交通模型构建及应用研究——以广州市为例

Truck Transport Model Development and Application Based on Multi-source Data:A Case Study of Guangzhou City

通过分析既有货运交通模型的难点及建模数据源的独立性,设计了多源数据融合的数据驱动货车交通模型框架。面向长周期道路卡口监测数据,运用轨迹溯源方法,建立了基于卡口交通分析小区的车辆出行OD。针对重型货车长距离和短距离出行活动特征的差异性,基于重型货车GPS数据的时空OD出行链生成算法,建立了区分高速公路和非高速公路的重型货车出行OD矩阵。基于数据驱动的货车模型框架,建立土地利用数据与高速公路收费流水数据、重型货车GPS数据及道路卡口监测数据之间的联系,推算得到以高速公路和非高速公路、重型货车和非重型货车为分类基础的货车出行需求矩阵。结合广州实例,从货车出行总体特征、时空分布、道路货车交通及与城市空间结构的协调性等方面进行了分析。研究结果表明,基于多源数据融合的货车交通模型能够较好地适应城市现状和初近期货车交通预测分析的需要。

By analyzing the difficulties of established freight transport models and the independence of modeling data sources,a data-driven truck transport model framework with multi-source data fusion is designed.Facing the longcycle road checkpoint monitoring data,using the trajectory traceability method,the vehicle travel OD based on the checkpoint traffic analysis zone is established.For the differences in the characteristics of long-distance and short-distance travel activities of heavy trucks,based on the spatio-temporal OD trip chain generating algorithm of GPS data of heavy trucks,the OD matrix of heavy trucks'trips that distinguishes between highways and non-highways is established.Based on the data-driven truck model framework,the linkage between land use data and highway toll data,heavy truck GPS data and road checkpoint monitoring data is established,and the truck travel demand matrix based on the classification of highways and non-highways,heavy trucks and non-heavy trucks is derived.Combined with the example of Guangzhou,the overall characteristics of truck travel,spatial and temporal distribution,road truck transport and the coordination with urban spatial structure are analyzed.The results show that the truck transport model based on multi-source data fusion can be better adapted to the current situation of the city and the needs of the initial and near-term truck transport prediction and analysis.