Data-driven based eco-driving control for plug-in hybrid electric vehicles

Jie Li,Yonggang Liu,Yuanjian Zhang,Zhenzhen Lei,Zheng Chen,Guang Li

DOI: https://doi.org/10.1016/j.jpowsour.2021.229916

IF: 9.2

2021-06-01

Journal of Power Sources

Abstract:<p>With the development of connected and automated vehicles, eco-driving control is reckoned to generate unprecedented potential on energy-saving in electrified powertrain. In this paper, a data-driven based eco-driving control strategy with efficient computation capacity is proposed for plug-in hybrid electric vehicles to achieve approximate optimal energy economy. An efficient hierarchical optimal control scheme is designed to mitigate the massive computational cost during velocity optimization and powertrain control. A data-driven optimal energy consumption cost model and an optimal battery current model are respectively constructed via two neural networks and served as the critic model and the system model during velocity optimization. Furthermore, the neural network-based dynamic programming is exploited to optimize the vehicle velocity by merging the data-driven models and Bellman optimality principle. The simulation results demonstrate that the proposed method can remarkably improve fuel economy by up to 16.7% in complicated driving conditions, compared with conventional sequential optimization methods. Furthermore, the data-driven control scheme can drastically improve the computational efficiency with slight sacrifice on fuel economy, compared with the optimum benchmark.</p>

energy & fuels,materials science, multidisciplinary,electrochemistry,chemistry, physical

What problem does this paper attempt to address?

This paper attempts to solve the energy - saving problem of plug - in hybrid electric vehicles (PHEVs) under complex driving conditions. Specifically, the author proposes a data - driven eco - driving control method, aiming to achieve approximately optimal energy economy by optimizing vehicle speed and energy management. This method takes advantage of the development of connected and automated technologies. By constructing two neural network models (one for modeling the optimal energy consumption cost and the other for modeling the optimal battery current), combined with the dynamic programming algorithm, it achieves high - efficiency computing power and a significant improvement in fuel economy. The main objective of the paper is to overcome the problem of incomplete decoupling between speed planning and energy management in traditional sequential optimization methods and to reduce the computational intensity in co - optimization methods.