Analysis of Power Electronics-Dominated Hybrid AC/DC Grid for Data-Driven Oscillation Diagnosis

Abstract

The high penetration of renewables has imposed great oscillation risks on power grids, especially for the power electronics-dominated hybrid AC/DC grid. Measurement data-driven diagnosis of oscillations regarding where, why and how they happened is crucial to the emergency control. However, before implementing any data-driven diagnosis, a comprehensive analysis of the system oscillation is required. In this paper, a frequency-domain network model of the hybrid AC/DC grid is established, considering both ac-and dc-side dynamics. Based on such model, the frequencydomain modal analysis (FMA) method is introduced to its oscillation assessment, where information like participation factor can be extracted. Finally, the mapping relationship between FMA results and measurement data is identified, laying the ground for data-driven oscillation diagnosis. Case studies are conducted in a typical hybrid AC/DC grid to validate the accuracy of the above analysis.

Research on high proportion of clean energy grid-connected oscillation risk prediction technology based on CNN and trend feature analysis

“…The purpose of stability model analysis is to reveal the mechanism and characteristics of harmonic disturbance caused by power electronic grid-connection. Compared with conventional thermal generating units, power electronic equipment has the advantage of fast output response, but its moment of inertia is low, so it needs to be discussed in the stability of grid-connected system 18,19 . This also lays a theoretical foundation for subsequent power output density analysis and moving average method.…”