The energy requirement of the world is increasing exponentially every year. With the power demand sites concentrated in the region with the highest population density, energy production sites must be located remotely. This isolation requires lengthy and complicated grids to be designed and manufactured to transmit power over long distances. Grid failures and blackouts profoundly impact a region's social and economic conditions, and thus grids require constant monitoring. Researchers at George Washington University have designed a novel idea to incorporate analytical capabilities within the existing smart sensor technologies or as an independent system without constant monitoring.
The proposed system can detect events/anomalies accurately and quickly. Some detectable incidents are surges, voltage sags, power swings, load changes, topology changes, etc. This approach revolutionizes the existing measurement and monitoring paradigms (centralized) in power grids to a high-fidelity distributed setting for measurement (sensing) and decision-making (actuating) setting. This decentralization of decision-making power makes the grid more intelligent and self-reliant. It will shield the grid when traditional decision-making channels are disrupted due to communications failures and other vulnerabilities. The proposed system has analytical capabilities which makes use of Gabor wavelet transformation for waveform pattern recognition and uses convolutional neural network and machine learning to predict and decision making.
Applications:
Power grid online surveillance in both transmission and distribution systems
Fast and accurate post-event analysis;
Stability Monitoring;
Fault Analysis: Fault Detection, Fault Type Classification, Fault Location
Advanced Control Schemes: Wide Area Protection and Control.
Advantages:
Resistive to timing synchronization failures, communication losses (failures or attacks), data transmission delays
Self-reliant on original power waveforms and no vulnerability to PMU errors and latencies
High accuracy rate of 94.7% and a faster reporting speed at a rate higher than 60 samples per second than standard PMUs.
Real-time situational awareness with online distributed event detection and classification