Hierarchical Control for Adaptive EV Charging : A multilevel control strategy for scheduling and control of EV chargers
Abstract: Demand- side management is becoming an effective method that is being widely exploited by the balance markets, but not much of the demand flexibility is currently exploited in residential demand- side management. Electric Vehicles (EVs), with their significant presence in the transportation sector, is a critical load for residential demand- side management. The aggregator, an actor in the energy market for providing balancing services, can utilize the charging loads of EVs to balance the supply and demand of electricity. But the aggregator faces many uncertainties when coordinating the EV charging operations. These uncertainties emerge due to the EV customers’ behavior, the power grid’s fluctuating behavior, and any distributed generation units (e.g., PV units and wind turbines) owned by the aggregator. The entire concept of customer- oriented EV charging can be split into two different problem statements that need to be addressed. The first one is the overall charging cost reduction for the customer. The second problem statement is ensuring user convenience at all times while performing smart EV charging. The objective of the thesis is to analyze a two- level (Supervisory and local) hierarchical control strategy for the EV chargers. The supervisory level contains forecasters that model the uncertainties in the inputs sent to a scheduler, which is built as a part of this thesis. The local level of the hierarchical control then acts upon the schedule received from the supervisory level. The forecasters predict the daily energy requirements of the EV fleet and fleet availability during each hour. The forecasters are trained based on historical data of EV fleet charging, to model the uncertainties mentioned above. The scheduler is designed to solve a MILP based optimization problem. The objective of the scheduler is twofold, 1) to reduce the electricity bill of the real estate and 2) to ensure self- consumption maximization of locally produced energy at the real estate. The forecasted parameters ensure the daily energy of the EV fleet is satisfied by the scheduler by intelligently shifting the charging events based on fleet availability. It more importantly ensures user convenience, i.e making sure that the user has enough charge in their car when taking it out. The model proposed in this thesis maximizes the efficiency of energy distribution by smoothing the consumption peaks of the EV chargers while ensuring user satisfaction at all times.
AT THIS PAGE YOU CAN DOWNLOAD THE WHOLE ESSAY. (follow the link to the next page)