High-Fidelity Modeling of Light-Duty Vehicle Emission and Fuel Economy Using Deep Neural Networks

The transportation sector contributes significantly to emissions and air pollution globally. Emission models of modern vehicles are important tools to estimate the impact of technologies or controls on vehicle emission reductions, but developing a simple and high-fidelity model is challenging due to the variety of vehicle classes, driving conditions, driver behaviors, and other physical and operational constraints. Recent literature indicates that neural network-based models may be able to address these concerns due to their high computation speed and high-accuracy of predicted emissions. In this study, we seek to expand upon this initial research by utilizing several deep neural networks (DNN) architectures such as a recurrent neural network (RNN) and a convolutional neural network (CNN). These DNN algorithms are developed specific to the vehicle-out emissions prediction application, and a comprehensive assessment of their performances is done. Sensitivity analysis is carried out for input predictor selection. Training and testing datasets are selected and a random route is selected for validation of the learning procedure. In addition, evaluation of the effect of different groups of input data for the prediction is studied. Results show that deep recurrent and convolutional neural networks have relatively better accuracy compared to other prediction models. Also, preliminary results show that the deep neural network’s performance consistently improves when given datasets with more input variables, potentially indicating improved usability for researchers compared to shallow and basic neural networks. The observations and summaries in this paper are meant to be used as a constructive source for scholars and users of vehicle emissions models. Fast computation and high-accuracy emissions models at the individual vehicle level can improve technology implementation and control in real-time. A possible extension of this work, includes study on other vehicle types and developing selected models using more datasets and integration with fuel economy models.