Electric vehicles (EVs) are likely to have a continued presence in the light-vehicle market in the next few decades. As a result, EV charging will put an extra burden on the distribution grid and adjustments need to be made in some cases. On the other hand, EVs have the potential to support the grid as well. This paper presents a single-phase bidirectional charger topology which pairs up a photovoltaic (PV) source with an EV charger resulting in production cost reduction. The presented topology is then used for vehicle-to-grid (V2G) services. The main focus of this paper is on power quality services which only slightly discharge the battery. Among these services, it studies the possibility of local reactive injection of EVs connected to the grid through a single-phase charger to compensate for voltage drops caused by motor startup or inductive loads. It also studies the possibility of active power injection of EVs for short time periods during PV transients in cloudy weather to keep the system stable. It also studies the potential of EVs to help during low voltage ride-through of the PV sources. The studies are performed using Simulink simulations and a real-time implementation in Real Time Digital Simulator (RTDS). The results demonstrate the effectiveness of power quality V2G services with small wear on the EV battery.

This paper presents a framework for cyber attack impact analysis of a smart grid. We focus on the model synthesis stage in which both cyber and physical grid entity relationships are modeled as directed graphs. Each node of the graph has associated state information that is governed by dynamical system equations that model the physics of the interaction (for electrical grid components) or functionality (for cyber grid elements). We illustrate how cause-effect relationships can be conveniently expressed for both analysis and extension to large-scale smart grid systems.

All-electric ship power systems have finite generation and include a large portion of dynamic loads and nonlinear loads relative to the total power capacity. Therefore, the load demand and power generation of the system should be matched in real-time. In this paper, a novel multi-agent system-based real-time load management technique is proposed to optimally determine the switch status of loads in DC zones while satisfying the operating constraints of the system in real-time. The multi-agent system cooperative control protocol is developed based on a proposed reduced-order agent model and artificial potential function of the multi-agent system, which aims to maximize the energized loads in the all-electric ship power system. The cooperative controller aims to cooperatively achieve the group objectives that are difficult to reach by centralized controller. Further, simulation results verify the viability and performance of the proposed technique in PSCAD/EMTDC software.

Reactive power control is a fundamental issue in microgrids, especially during islanded mode operation with no support from the main grid. Lack of infinite bus, tightly coupled generation and consumption, and existence of nondispatchable intermittent renewable power sources reinforce the need for a new VVC scheme. This paper presents a new model predictive control (MPC)-based dynamic voltage and var control (VVC) scheme, which includes the dynamics of the microgrid in the VVC formulation. The MPC-based controller uses a simplified voltage prediction model to predict the voltage behavior of the system for a time horizon ahead. The advantage of this method is that it can avoid unstable voltage conditions in microgrids by prediction of the instability ahead of time. This method can also avoid voltage drops or swells in any of the phases of the system since the model can predict the voltage of each phase separately. Also, the presented method can be implemented online so it can efficiently use the time-variant reactive capabilities of the distributed generators to compensate for reactive power needs of the system. This controller is tested for different operating conditions of the microgrid and the simulation results confirm that the MPC controller successfully keeps the system stable and achieves a smooth voltage profile.

This paper presents a methodology for generating restoration sequences in distribution systems and microgrids with emphasis on deriving optimal switching times for remote control switches (RCS). Microgrid architecture considered are assumed to be operating in islanded single master mode. Control methods for handling the unbalance inherent in distribution systems using single-phase controlled PQ inverters were developed and incorporated in the proposed method. The problem of considering simultaneously both energy optimization and dynamic stability of the system, which is lacking in existing methods, was addressed by studying the transient behavior of microgrids based on extended EMTP simulations and then approximate and incorporate these behaviors into the optimization formulation for restoration. The approach is based on multi-objective optimization of final restored load and total restoration time. The methodology was studied on modified IEEE 123 node test feeder. An EMTP simulation in PSCAD of the formed microgrids was used to validate the accuracy of the dynamic and steady-state characteristics of the resulting systems. By considering the transient behavior of DGs during restoration, a better restoration sequence that does not trigger protection relays could be realized.

Isolated microgrid power systems are prone to large frequency and voltage deviations, since they have limited generation and finite inertia and may include pulse power loads and large proportion of dynamic loads. Thus, they require effective power management methods to operate optimally, while satisfying operating and security constraints. In this paper, a novel integrated security-constrained model-based power management approach is proposed for isolated microgrids in all-electric ships during the normal/alert operating state. The new power management method was formulated as a multi-objective optimal control problem. A method based on the Nondominated Sorting Genetic Algorithm II was developed to solve the problem. The developed integrated security-constrained power management approach was applied to a notional all-electric ship computer model. The simulation results, conducted in PSCAD, showed the effectiveness of the new power management method.