Continuous-time Model Predictive Control for Real-time Flexibility Scheduling of Plugin Electric Vehicles

This paper proposes a continuous-time model predictive control (MPC) for co-optimizing the charging flexibility of plugin electric vehicles (PEVs) and generation schedule of generating units in real-time power systems operation. A continuous-time queuing model is developed to aggregate and cluster a large population of PEVs, which represents their aggregate flexibility to the power system operator. The proposed model integrates the most recent information about the system load and PEV arrival, and utilizes the flexibility of PEVs and generating units to supply the ramping requirements of load, while ensuring the delay-based and deadline-based service quality constraints of PEVs. The proposed model is implemented on the IEEE Reliability Test System, using PEV and load data of California. The simulation results demonstrate effectiveness of the proposed model to utilize the flexibility of PEVs in real-time power systems operation, which considerably reduces ramping requirements from conventional generating units.