Real-time low-frequency oscillations monitoring

Area of Science:

• Electrical Engineering
• Power Systems Analysis
• Control Theory

Background:

• Interconnected power grids face stability challenges due to low-frequency oscillations.
• These oscillations limit scalability and transmission capacity, potentially causing blackouts.
• Real-time synchrophasor data offers a solution for detecting and controlling oscillations.

Purpose of the Study:

• To propose two low-complexity algorithms for identifying and monitoring low-frequency oscillations in power systems.
• To enable fast tracking of these oscillations using real-time Phasor Measurement Unit (PMU) data.
• To enhance the reliability and scalability of power grids.

Main Methods:

• A one-time matrix pencil method applied to the power spectrum matrix of PMU data for initial detection.
• Two distinct low-complexity algorithms for fast tracking: a recursive fast data projection method and a gradient-descent based fast recursive algorithm.
• Comparison with state-of-the-art techniques including matrix pencil method, frequency domain decomposition, and TLS-ESPRIT.

Main Results:

• The proposed algorithms accurately detect and monitor low-frequency oscillations.
• Both methods demonstrate high accuracy, particularly in terms of computational complexity for large-scale systems.
• The gradient-descent based algorithm effectively tracks oscillations on a per-PMU basis.

Conclusions:

• The developed algorithms offer efficient and accurate solutions for managing low-frequency oscillations.
• These methods are suitable for real-time monitoring in large power systems with numerous PMUs.
• Implementation of these algorithms can improve power grid stability and transmission capacity.