PV hosting capacity assessment in distribution systems considering resilience enhancement

Abstract

This paper presents an innovative strategy to assess the photovoltaic (PV)-based distributed generation (DG) hosting capacity considering the operation under normal and emergency conditions of electrical distribution systems (EDSs). In the emergency condition, the proposed strategy aims to improve the recoverability of EDSs against a set of high-impact fault scenarios. This recoverability process is achieved by the optimal coordination of topology reconfiguration, islanding operation of dispatchable DG units, and pre-positioning and displacement of mobile DG (MDG) units. This problem is formulated as a two-stage stochastic formulation, where the first one defines the DG hosting capacity and the amount of MDG units to be positioned in staging locations. Meanwhile, the second stage simulates high-impact fault events and, by applying resilience alternatives, the EDS recoverability can be improved. Inherently, the two-stage stochastic formulation is represented by a mixed-integer linear programming (MILP) model. The objective function of this MILP model maximizes the installed PV-based DG capacity while the amount of energy load shedding after fault events is minimized. To validate and show the scalability of the proposed strategy, two EDS are studied under different high-impact fault events and considering the application of multiple resilience alternatives. Results show that by estimating the capacity of PV-based DG simultaneously with the restoration process, the number of pre-positioned and dispatched MDG units can be reduced. On the other hand, when this PV capacity is determined disregarding fault scenarios, this solution could lead to unviable conditions and, thus, generation curtailment of up to 80% could be required.