Cyber-physical energy management systems

Abstract

Following the global concerns on the growing demand for energy and the increasing rate of environmental emissions, coinciding with the necessity to move toward sustainable energy systems, this doctoral dissertation focuses on presenting effective energy management systems for low-voltage systems. The applications of the proposed energy management systems include home energy management systems, energy hub operation and management, and microgrid scheduling. In this respect, first, the principles of decision-making applicable to energy systems are given; then, a self-scheduling model is developed for end users, enabling them to actively participate in the demand response program and save costs in their electricity bills through a time-of-use (TOU) tariff, achieved through a mixed-integer linear programming (MILP)-based home energy management system (HEMS). Then, the model is upgraded to a more realistic simulation platform considering the end user preferences and changing the decisions during the scheduling horizon. The problem is modeled using a model predictive control (MPC) strategy to take account of the local power generation. Further, the real-time pricing (RTP) mechanism is used in the study. Consequently, an energy trading-enabled multicarrier energy hub microgrid is proposed, and the problem is studied by employing a mixed-integer quadratically constrained programming (MIQCP)-based optimal power flow (OPF) model for a network with hubs. The proposed model is intended for handling the OPF problem with AC constraints in distribution networks to deal with the more realistic operating conditions. In addition, the energy flow model is used for the energy balance model for all energy carriers. Employing the energy flow model provides more flexibility in handling the input–output features of the hub assets. The energy hub model is presented as a standard MILP model dealing with scheduling constraints. A scenariobased stochastic optimization model is selected to address the uncertainties caused by the input parameters and the complexity of the optimization problem. Then, the problem is converted into a robust chance-constrained optimization problem to handle the model uncertainties. The forecasting errors on both the input and output sides are projected in the developed model, resulting in a more realistic operation of the energy hubs. The loadability index is introduced to evaluate the robustness of the model against system uncertainties. Lastly, this dissertation addresses the optimal operation strategies in charging and discharging of energy storage units to minimize the overall energy costs, maximize the utilization of renewable energy sources, and improve the performance of individual cells of the batteries. To this end, the thermal model of Li-ion batteries is presented as an interdependent heat generation model for Li-ion batteries, and to linearize the model, the McCormick relaxation method is employed. In order to maintain the temperature of the large-scale Li-ion batteries, an optimal operation strategy is introduced and implemented for the microgrid scheduling strategy. In this regard, the temperature control strategy is elaborated using a heating, ventilation, and air conditioning (HVAC) system.