Charging Scheduling with Computation Energy Consumption and Carbon Tax in Vehicle Networks

For electric vehicles (EVs) that are running out of power during the journey, how to select the charging station (CS) and charge at the minimum cost is an important problem. Taking the computation energy consumption and the carbon tax into consideration, in this paper, the CS and charging power selection problem is formulated into an optimization problem, where the charging expenditure and waiting time are minimized. Since the proposed problem changes dynamically with EV user behaviors, the proposed problem cannot be solved by traditional optimization methods. Then the Multi-Agent Deep Deterministic Policy Gradient (MADDPG) algorithm is used to find the near-optimal solution. Simulation results show that our proposed strategy helps reduce not only the charging costs but also the CO<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf> emissions. In addition, our strategy balances the load of each CS and reduces the peak-to-valley difference in electricity consumption.