A Real-Time Controlled Islanding and Restoration Scheme Based on Estimated States

Abstract

Power system operators are facing major challenges today to keep the system operating at the admissible limits. Recent blackouts demonstrated the need for a systematic study and design of a comprehensive system control strategy. Intentional Controlled Islanding (ICI) has been proposed as an effective corrective control action of final resort to save the system from a partial or a complete blackout. ICI limits the occurrence and consequences of blackouts by splitting the power system into a group of smaller, stable, and sustainable subsystems, also called islands. After a controlled system separation, power system operators should resynchronize and reconnect each island to restore the system. In this sense, real-time knowledge of the operating condition of the islands is required. In this paper, a realtime ICI and restoration scheme is proposed. The proposed scheme consists of an ICI algorithm that finds islanding solutions with minimal power-flow disruption while considering power system restoration constraints (e.g., blackstart availability, sufficient generation capacity and observability), a real-time state estimator that monitors the system before and after the islanding, and a restoration process. The proposed ICI and restoration scheme is tested using the dynamic IEEE 39-and 118-bus test systems.

Sequential Disaster Recovery Model for Distribution Systems With Co-Optimization of Maintenance and Restoration Crew Dispatch

“…Moreover, the input data errors during the online execution process can cause the delay or failure of system restoration, in such way of the load metering errors or communication failure. In this case, the online execution of the proposed scheme should be cooperative with the effective state estimation [31]. By applying these methods, the proposed scheme can be improved for implementation and execution.…”

“…Let V MC m be the set of damaged components in the cell m. Then, the cell m should only be energized after all damaged components in cell m are repaired, which means: The interdependence of RCDM coupling with MCDM. If a cell m is energized by closing a manual switch at vertex a between cell m and n, this manual switch at vertex a should be closed by RCs after all damaged components in the cell m are repaired by MCs.Hence, the operation of the manual switch needs the coordination of the MCDM to repair damaged components and the RCDM to dispatch the RCs, as modelled in Eq(31). However, the operation of the remote-controlled switch is not constrained by the dispatch of RCs, i.e., the RCDM.…”

Incorporating energy storage and demand response into intentional controlled islanding using time decomposition

“…Some works in the literature have employed mixed‐integer linear programming (MILP) methodology to obtain optimal solutions of ICI problem. In Reference 28, an MILP‐based ICI and parallel power system restoration schemes are jointly proposed by adding black start capability as well as sufficient generation capacity and observability constraints to an ICI scheduling problem. In Reference 29, two ICI models based on mixed‐integer nonlinear programming (MINLP) formulation and a computationally lighter MILP model are proposed, which are able to enhance the first swing transient stability of the islands along with satisfying classical operation constraints.…”

“…While some previous ICI works (such as References 19,23,28) have focused on minimal power flow disruption, some other ones (such as References 20,35,36) have considered minimal power imbalance as the objective to obtain optimal islanding strategies. However, as the first contribution of this article, the proposed ICI approach jointly minimizes both the power imbalance and the power flow disruption to obtain more effective islanding solutions benefiting from the advantages of the both groups.…”

Network partitioning approach for reactive power/voltage control using analytical nodes coupling expressions

“…The power-grid partitioning aims to decompose the whole network into several regions so that any disturbance in one region has a sufficiently small impact on other regions [4][5][6]. In the literature, several network partitioning methods have been proposed for various power system applications such as dynamic security assessment [7][8][9], system islanding [10][11][12] and reactive power and voltage control [13][14][15]. The existing network partitioning methods can be classified into two categories: static-and dynamic-based methods [16].…”