Reducing the Peak Power through Real-Time Scheduling Techniques in Cyber-Physical Energy Systems

This paper presents a method for applying realtime scheduling techniques to balance the power usage of electric loads in cyber-physical energy systems. The aim of the proposed approach is to achieve predictability of the activation of electric loads to guarantee an upper bound on the peak electric power consumption.The contribution of this paper encompasses several aspects. The relevance of balancing electric loads is discussed, motivating the use of real-time scheduling techniques to achieve predictability on electric-load management. We introduce the innovation of modeling the physical system as a set of periodically activated loads, that can be effectively managed by adequately adapting traditional real-time system models and scheduling algorithms, to guarantee an upper bound on the peak power consumption. For this purpose, we present a problem formulation based on linear programming, while a low-complexity heuristic is proposed to limit the complexity of the optimization process. Simulation results are presented to assess the performance of proposed methods.