PPO Based Computation Offloading and Resource Allocation Algorithm in Smart Grids

In smart grids, data communication and computation are required between electric vehicles (EVs) and charging stations (CSs) to realize charging management. In this paper, we investigate the joint optimization of computation offloading and resource allocation in edge computing (EC)-based smart grids to address the massive energy consumption caused by charging management. Firstly, a collaborative task offloading, computation resource allocation, and spectrum and power allocation problem is formulated to minimize the energy consumption in the EC-based system, by considering time-varying conditions as well as a delay constraint. Then, this NP-hard long-term optimization problem is defined as a Markov decision process, and a proximal policy optimization (PPO)-based algorithm is proposed to avoid the curse of dimensionality. Simulation results show that the PPO-based algorithm can effectively reduce the average energy consumption by 20.3% and 25.6% compared with two benchmark algorithms. Moreover, as the number of CSs increases, the advantage of the PPO-based algorithm is more obvious.