Tielong Shen

This paper investigates a novel post-disaster power recovery scenario in which shelter batteries are subject to dynamic charging time constraints determined by their state-of-charge ( $SoC$ ) dynamics. In this context, electric vehicles (EVs) serve as both transportation resources and mobile energy storage units. The objective is to determine an EV delivery plan that satisfies the power demand of shelter batteries within their dynamic charging time constraints while minimizing the total fleet travel costs. In addition, road-grade effects on EV energy consumption during travel are incorporated. In the first phase, a reduced graph is extracted from the original disaster-affected network to simplify the network topology. In the second phase, the problem is formulated as a mixed-integer linear programming (MILP) model. To obtain a high-quality delivery plan, an adaptive large neighborhood search (ALNS) algorithm is developed with problem-specific destroy and repair operators. The performance of the proposed ALNS is evaluated through comprehensive numerical experiments on benchmark instances of varying sizes constructed from a public road network. The results demonstrate that the proposed ALNS outperforms state-of-the-art methods. Finally, sensitivity analyses are conducted to quantify the impacts of uncertainties, time constraints, road grades, and reduced-graph construction.

In the past decades, the energy optimization strategy for human-driver hybrid electric vehicles (HEVs) has been widely investigated to improve fuel economy. With the highly increasing penetration of connected and automated HEVs, fuel consumption reduction can be achieved further. This paper explores the potential to reduce the whole fuel consumption by jointly optimizing speed dynamic and HEV powertrain operation for a large population of connected and automated HEVs. To avoid the computation burden of the large population of HEVs, a novel decentralized control scheme is designed by employing the mean field game (MFG) theory. Thanks to the communication by vehicle to everything (V2X) that the real-time velocity distribution of the large population of HEVs is available for each HEV, the receding horizon sense is introduced to MFG to avoid the open-loop prediction error of velocity distribution within the optimization horizon. The solution is derived from a novel numerical method for the general nonlinear MFG problem. A virtual connected traffic simulation platform is built in MATLAB/Simulink, and simulations are conducted to show the effectiveness of the proposed strategy.

We consider the problem of deploying a connected car, capable of V2X communication, to propagate data to groups of evacuation shelters that have lost mobile network connectivity. The problem is a combination of two subproblems: first of finding the optimal order in which the destinations should be visited, and second of finding the optimal path in a road network that allows a connected car to deliver information to all prescribed shelters. We develop a genetic algorithm with double chromosome encoding to simultaneously solve both subproblems. One chromosome encodes the solution to the routing problem, the other to the path-planning problem. Experimental results demonstrate that our algorithm is capable of finding the optimal solution with reasonable accuracy and precision.