Thesis
Enhanced real time monitoring and decision support for critical grid situations using synchronised measurements
- Creator
- Rights statement
- Awarding institution
- University of Strathclyde
- Date of award
- 2026
- Thesis identifier
- T18050
- Person Identifier (Local)
- 202168827
- Qualification Level
- Qualification Name
- Department, School or Faculty
- Abstract
- Synchronised power system measurements, conventionally provided by phasor measurement units (PMUs), are essential to the real-time detection and monitoring of abnormal conditions. As the penetration of inverter-based resources (IBRs) increases, accurate monitoring of system dynamics, e.g., non-sinusoidal distortions, sub/super synchronous oscillations (SSOs), etc., becomes critical to ensure system security. However, PMUs rely on a fundamental sinusoidal reference, which limits their ability to accurately characterise the emerging IBR dynamics. In recent years, continuously streamed synchronised waveforms (synchro-waveforms) have enabled the access to higher-resolution voltage and current samples. While they can accurately capture such IBR-induced phenomena, the large data volumes and high streaming requirements hinder their practical wide area deployment. Consequently, effective monitoring of IBR dominated power grids remains a key challenge. This research therefore aims to investigate novel synchronised measurement methods capable of characterising IBR dynamics with manageable data volumes, and their applications towards critical grid monitoring and decision support functions to facilitate the secure integration of IBRs. In the thesis, the performance of conventional monitoring solutions in the presence of IBR-induced dynamics is first investigated via an analysis of real-world PMU measurements of SSO events in the Great Britain (GB) power system. Precursory signatures in phasor measurements are identified, and a machine learning (ML) based early warning system is developed. It is demonstrated that PMUs can be inadequate in capturing IBR-induced dynamics and more informative analyses using continuously streamed synchro-waveforms remain impractical. The research therefore proposes a novel generalised synchro-waveform measurement unit (G-SWMU) processing algorithm. The G-SWMU represents non-sinusoidal transients using a wavelet scale correlation transient filter (WSCTF) and estimates multi-frequency phasors using a novel iterative extended discrete Fourier transform based empirical wavelet transform (I-EDFT-EWT). Compatibility with existing supporting infrastructure is demonstrated by mapping the G-SWMU outputs to standard messaging formats, along with theoretical estimates of the reduced bandwidth requirements. The performance of the G-SWMU is validated using real-world event waveforms from a wide range of system conditions, enabling enhanced SSO analysis and accurate anomalous event measurement. The adaptability and scalability of the proposed G-SWMU algorithm is further demonstrated beyond IBR-specific use cases through a novel G-SWMU enabled open-set event classification algorithm. The transient distortions captured by the G-SWMU algorithm provide discriminative features from limited training data, which enables accurate classification of power system events while effectively rejecting previously unseen events. The contributions of this thesis therefore provide valuable knowledge and practical insights that support the adoption of the G-SWMU as a foundation for a new generation of synchronised measurement technologies, thereby enabling accurate monitoring and effective mitigation of critical events, which support the secure and reliable operations of IBR-dominated power systems.
- Advisor / supervisor
- Hong, Qiteng
- Resource Type
- DOI
Relações
Itens
| Miniatura | Título | Data de carga | Acesso | Ações |
|---|---|---|---|---|
|
|
PDF of thesis T18050 | 2026-07-24 | Público | Baixar |