Abstract

Abstract
This article presents a novel mixed-integer second-order cone programming model to increase the photovoltaic (PV) hosting capacity and optimize the operation of distribution networks. The problem considers voltage and reactive (Volt/VAr) control through the optimal operation of capacitors banks, substations' on-load tap changers, voltage regulators, and network reconfiguration with radial and closed-loop operation topologies. The proposed formulation considers voltage-dependent models for loads and capacitor banks. The objective function maximizes the PV hosting capacity of the network. Numerical experiments are carried out using the 33-node and the 85-node networks. Results demonstrate the effectiveness of the proposed formulation to increase the penetration of PV sources, especially when the closed-loop operation is allowed, together with network reconfiguration and Volt/VAr control.

Abstract
This article presents a restoration approach for improving the resilience of electric distribution systems (EDSs) by taking advantage of several operational resources. In the proposed approach, the restoration process combines dynamic network reconfiguration, islanding operation of dispatchable distributed generation units, and the prepositioning and displacement of mobile emergency generation (MEG) units. The benefit of exploring a demand response (DR) program to improve the recoverability of the system is also taken into account. The proposed approach aims to separate the in-service and out-of-service parts of the system while maintaining the radiality of the grid. To assist the distribution system planner, the problem is formulated as a stochastic-scenario-based mixed-integer linear programming model, where uncertainties associated with PV-based generation and demand are captured. The objective function of the problem minimizes the amount of energy load shedding after a fault event as well as PV-based generating curtailment. To validate the proposed approach, adapted 33-bus and 83-bus EDSs are analyzed under different test conditions. Numerical results demonstrate the benefits of coordinating the dynamic network reconfiguration, the prepositioning and displacement of MEG units, and a DR program to improve the restoration process.

Abstract
The issue of a state estimation-based fault-tolerant controller for direct current (dc) microgrids (MGs) is studied in this article. It is considered that the dc MG contains nonlinear constant power load (CPL) and is subjected to actuator faults. Current sensors are not installed and the voltages of the dc MG are measured in the presence of noise and sensor faults. To estimate the system states, a novel dual-Extended Kalman filter is proposed, which simultaneously estimates the states and faults. The fault- and noise-free estimations are then deployed in a nonlinear Takagi-Sugeno fuzzy predictive controller to regulate the dc MG. The proposed method outperforms the exiting results, being robust against faults and noise. Also, the predictive scheme makes it robust against system uncertainties and forces the system states to converge the desired values, precisely. The accuracy and robustness of the developed method are evaluated and compared to advanced state-of-the-art techniques for a typical dc MG with a resistive load, CPL, and energy storage unit.

Abstract
The penetration of the distributed energy resources in the distribution networks is facilitated by the structure of the microgrids (MGs). The MG operator (MGO) can schedule the MG resources to meet the local load and participate in the wholesale markets. In this article, a new model is developed for the MGO participation in the day-ahead (DA) (energy and reserve) and the real-time (RT) energy markets under uncertainties. For this purpose, the effect of the uncertainties of demand and generation from renewable energy sources on the MGO decisions is represented in a two-stage stochastic model. The MGO bids in the DA and RT markets are modeled as the first and the second stage decisions, respectively. Moreover, the information gap decision theory method is used to model the behavior of the MGO to address the uncertainties of the RT energy market price and the probability of calling the reserve. The results show that as the RT price uncertainty radius increases, the energy sold to the RT market decreases/increases in the risk-averse/risk-taker strategy. Furthermore, to manage the uncertainty related to the probability of calling the reserve, the reserve capacity provided by the MGO in the risk-averse and the risk-taker strategies decreases and increases, respectively.

Abstract
Electric spring (ES) is a novel smart grid technology developed to facilitate the integration of renewable generation by controlling the demand of non-critical loads (NCLs). The utilization of ES to provide a single service such as voltage or frequency regulation, validated in a setup consisting of a single ES, has been extensively investigated. However, to take full advantage of this technology, it is necessary to develop control strategies to coordinate the operation of multiple distributed ESs to provide multiple services in power systems. To this end, this paper presents a rolling-optimization control strategy to coordinate the operation of multiple ESs for voltage regulation, congestion management and cost minimization of the real-time deviations from the scheduled energy exchanges with the grid in microgrids with renewable generation. The strategy is for centralized implementation, and includes a probabilistic optimal power flow-based optimization engine that finds the voltage references of ESs for each control interval taking into account generation variability and uncertainties. NCLs consist of electric water heaters, which are modeled taking into account physical constraints and the hot water demand. Simulations were carried out in two test systems with 14 and 33 buses.