| Autor: |
Zhang Y; Department of Civil and Environmental Engineering, University of South Carolina, Columbia, South Carolina, USA., Chen Y; Department of Civil and Environmental Engineering, University of South Carolina, Columbia, South Carolina, USA., Sun R; Department of Civil and Environmental Engineering, University of South Carolina, Columbia, South Carolina, USA., Huynh N; Department of Civil and Environmental Engineering, University of South Carolina, Columbia, South Carolina, USA., Comert G; Department of Physics and Engineering, Benedict College, Columbia, South Carolina, USA. |
| Abstrakt: |
Vehicle emission analysis currently faces a trade-off between easy-to-use, low-accuracy macroscopic models, and computationally intensive, high-accuracy microscopic models. In this study, we develop a surrogate model that leverages microscopic traffic and emission simulations to predict link-level emission rates. The input variables are obtained by aggregating 1 Hz simulated vehicle trajectories into hourly traffic condition factors (e.g., link average/variation of speed, truck fleet percentage, road grade, etc.). The emission ground truth data are generated using the Motor Vehicle Emission Simulator (MOVES) opmode-based analysis module. We explore different parameter and machine learning model structures to establish the statistical relationship of the input variables and the link-level emission rates. We demonstrate the ability of our model to accurately estimate vehicle-related emissions by using the Columbia, South Carolina road network as an example. This model can serve as a high-level planning tool to assess the impacts of emissions from transportation projects. Implications : Vehicle emission analysis is facing trade-offs between easy-to-use macroscopic emission models with low accuracy and computationally intensive microscopic models with high accuracy. Existing studies attempted to cope with the trade-off by pre-selecting representative emission rates but are still subject to the risk of not considering differentiated traffic patterns by using single emission rate. To fill in the knowledge gap in the literature, we develop a surrogate approach that fully integrates driving trajectories of heterogenous traffic patterns into a link-level emissions estimation model considering road characteristics. The model can achieve high accuracy and utilize publicly available traffic data in vehicle emission prediction. We apply the proposed model in a middle size city road network and demonstrate its capability to capture and quantify the impacts of traffic patterns on link-level vehicle-related emissions. Additionally, the proposed model can serve as a sketch planning tool for researchers and transportation air quality practitioners to quickly assess bounds of emissions benefits due to traffic operational and transportation strategies. |
