2016•Durham e-Theses (Durham University)Open access

Adaptive three-stage controlled islanding to prevent imminent wide-area blackouts

Hongbo Shao

Open full text 3 citations

Abstract

Power blackouts are a recurring problem worldwide, and research in this area continues to focus on developing improved methods for their prediction and prevention. Controlled islanding has been proposed as a last resort action to save the network before imminent blackouts when the usual means fail in an unexpected manner. Successful controlled islanding has to deal with three important issues that are involved in the implementation of islanding: when to island, where to island and what to do after islanding is implemented in each island. This thesis presents a framework that combines all three issues to achieve successful islanding based on wide area measurement systems (WAMS). In addition, this thesis focuses on the question of when to island. This question is critical to the success of the three-stage controlled islanding scheme because the possible issues of false dismissal and false alarm have to be handled. In false dismissal, islanding is triggered too late. However, the potentially unstable system is still allowed to operate, and this unstable system, which could have survived, may cause uncontrolled cascading blackouts. In false alarm, islanding is triggered too early, and an originally stable system is forced to split into islands, resulting in unnecessary disruption and economic loss. Thus, the early recognition and identification of “the point of no return” before blackout is inevitable. The single machine equivalent (SIME) method is adopted online to predict transient stability during cascading outages that would shortly lead to blackouts, giving support in decisions about when to island in terms of transient instability. SIME also evaluates dynamic stability after islanding and ensures that the selected island candidates are stable before action is taken. Moreover, in this thesis, the power flow tracing-based method provides all possible islanding cutsets, and SIME helps to identify the one that has the best transient stability and minimal power flow disruption. If no possible island cut set exists, corrective actions through tripping critical generators or load shedding are undertaken in each island. The IEEE 10-generator, 39-busbar power system and 16-generator 68-busbar system are used to demonstrate the entire framework of the controlled islanding scheme. The performance of each methodology involved in each stage is then presented.

Open-access reader

About this research paper

What this paper is about

Power blackouts are a recurring problem worldwide, and research in this area continues to focus on developing improved methods for their prediction and prevention. Controlled islanding has been proposed as a last resort action to save the network before imminent blackouts when the usual means fail in an unexpected manner. Successful controlled islanding has to deal with three important issues that are involved in the implementation of islanding: when to island, where to island and what to do after islanding is implemented in each island. This thesis presents a framework that combines all three issues to achieve successful islanding based on wide area measurement systems (WAMS). In addition, this thesis focuses on the question of when to island. This question is critical to the success of the three-stage controlled islanding scheme because the possible issues of false dismissal and false alarm have to be handled. In false dismissal, islanding is triggered too late. However, the potentially unstable system is still allowed to operate, and this unstable system, which could have survived, may cause uncontrolled cascading blackouts. In false alarm, islanding is triggered too early, and an originally stable system is forced to split into islands, resulting in unnecessary disruption and economic loss. Thus, the early recognition and identification of “the point of no return” before blackout is inevitable. The single machine equivalent (SIME) method is adopted online to predict transient stability during cascading outages that would shortly lead to blackouts, giving support in decisions about when to island in terms of transient instability. SIME also evaluates dynamic stability after islanding and ensures that the selected island candidates are stable before action is taken. Moreover, in this thesis, the power flow tracing-based method provides all possible islanding cutsets, and SIME helps to identify the one that has the best transient stability and minimal power flow disruption. If no possible island cut set exists, corrective actions through tripping critical generators or load shedding are undertaken in each island. The IEEE 10-generator, 39-busbar power system and 16-generator 68-busbar system are used to demonstrate the entire framework of the controlled islanding scheme. The performance of each methodology involved in each stage is then presented.

Why it matters

OpenAlex reports 3 citations for this work. Citation counts describe recorded attention and do not establish research quality.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

Power blackouts are a recurring problem worldwide, and research in this area continues to focus on developing improved methods for their prediction and prevention. Controlled islanding has been proposed as a last resort action to save the network before imminent blackouts when the usual means fail in an unexpected manner. Successful controlled islanding has to deal with three important issues that are involved in the implementation of islanding: when to island, where to island and what to do after islanding is implemented in each island. This thesis presents a framework that combines all three issues to achieve successful islanding based on wide area measurement systems (WAMS). In addition, this thesis focuses on the question of when to island. This question is critical to the success of the three-stage controlled islanding scheme because the possible issues of false dismissal and false alarm have to be handled. In false dismissal, islanding is triggered too late. However, the potentially unstable system is still allowed to operate, and this unstable system, which could have survived, may cause uncontrolled cascading blackouts. In false alarm, islanding is triggered too early, and an originally stable system is forced to split into islands, resulting in unnecessary disruption and economic loss. Thus, the early recognition and identification of “the point of no return” before blackout is inevitable. The single machine equivalent (SIME) method is adopted online to predict transient stability during cascading outages that would shortly lead to blackouts, giving support in decisions about when to island in terms of transient instability. SIME also evaluates dynamic stability after islanding and ensures that the selected island candidates are stable before action is taken. Moreover, in this thesis, the power flow tracing-based method provides all possible islanding cutsets, and SIME helps to identify the one that has the best transient stability and minimal power flow disruption. If no possible island cut set exists, corrective actions through tripping critical generators or load shedding are undertaken in each island. The IEEE 10-generator, 39-busbar power system and 16-generator 68-busbar system are used to demonstrate the entire framework of the controlled islanding scheme. The performance of each methodology involved in each stage is then presented.

Key concepts: Islanding, Blackout, Reliability engineering, ALARM, Electric power system, Transient (computer programming), Engineering, Computer science

Related papers

Back to paper searchBrowse research topicsOriginal source
Adaptive three-stage controlled islanding to prevent imminent wide-area blackouts — Research Paper | ScholarLens