Fair Communication Resource Scheduling in 6G and Digital Twin-Driven Virtual Power Plants

A virtual power plant (VPP) is a technology and management system that integrates distributed energy resources, offering significant advantages for the access and management of these distributed energy resources. With the development of distributed energy, higher requirements are being placed on the communication resource scheduling capabilities of virtual power plants. For dense source-load-storage devices, the high bandwidth, low latency, and high energy efficiency characteristics of 6G networks can meet the control requirements of virtual power plants for these devices. Digital twins are currently widely used in the industrial sector, providing positive effects such as real-time monitoring, prediction, and performance optimization. Applying digital twins to the communication resource scheduling of virtual power plants enables real-time perception of resource allocation and timely resource scheduling. This paper proposes a communication resource scheduling algorithm based on deep reinforcement learning, leveraging digital twin technology and the 6G distributed subnet architecture. This algorithm aims to efficiently utilize limited resources such as resource blocks, spectrum, and power to balance system throughput, fairness, and scheduling latency, thereby ensuring the QoS of source-load-storage device scheduling tasks. Simulation results demonstrate that the proposed resource scheduling algorithm enhances system throughput and fairness while meeting scheduling latency requirements, thereby satisfying the QoS needs of scheduling tasks.