Power system oscillations that can trigger blackouts show detectable warning signs minutes to hours in advance, and this project will build a prototype system to catch them in real time. As the UK adds more wind, solar, and HVDC connections to decarbonise the grid, these inverter-based resources can interact unpredictably with each other and with the existing network, creating growing oscillations that degrade power quality, disrupt industrial loads, and—if unchecked—escalate into widespread blackouts. Current monitoring methods require complex modelling or new hardware. Researchers at Strathclyde have already shown, using real voltage measurements, that every oscillation event they analysed produced measurable precursors—changes in frequency spectrum patterns and damping ratios—well before instability struck. This PhD will combine those data-driven signals with engineering domain knowledge to create a practical early warning system that uses data already flowing from existing grid sensors. No new infrastructure is needed. If successful, grid operators could receive alerts minutes to hours ahead of an oscillation event, allowing them to take corrective action before power quality degrades or supply is lost. The approach directly addresses a growing threat to energy security as the grid transforms.
View original technical description
This project will address the critical and growing challenge of oscillatory instability in decarbonised power systems. It aims to develop a prototype real-time early warning system to detect oscillations risks, monitor their progression, and inform mitigating actions to prevent such instabilities in advance. This will be achieved by characterising precursory signs extracted from real-time measurements, tracking their evolvements, and identifying potential oscillation sources using a domain knowledge-informed approach. Efforts to decarbonise power systems require extensive integration of Inverter-Based Resources (IBRs), e.g. wind, solar and HVDC. IBRs can interact negatively with the grid and each other, causing oscillatory instability. These oscillations can compromise power quality, disrupt critical loads and hinder renewable integration. If left unmanaged, they can escalate to widespread instability and blackouts, posing significant threats to energy security. Recent studies conducted at Strathclyde using real-world voltage measurements found that all analysed oscillation events exhibited precursory signs minutes to hours before the events occurred. These precursors can be characterised by certain features, e.g. frequency spectrum patterns, damping ratios, etc. If properly processed and analysed, they could serve to provide early warning of oscillations, offering a promising and powerful tool for monitoring and mitigating oscillation risks. The PhD will demonstrate that early precursory signs of oscillations can be detected in real-time minutes to hours before they occur, enabling effective early warning and mitigation of oscillation risks. By combining data-driven methods with domain knowledge, the approach eliminates the need for complex modelling and additional infrastructure. With the data required readily available from existing systems, the proposed approach represents a practical and promising solution for addressing the growing oscillation challenge.
Plain English summaries and category classifications on this site are generated by AI and may not perfectly reflect the original research.
Is something wrong? Let us know