Dimensionality Increment of PMU Data for Anomaly Detection in Low Observability Power Systems

2019/10/19 by Xin Shi, Shi, Xin, Robert Qiu +1
Computer Science · Engineering · #FOS: Electrical engineering #Signal Processing (eess.SP) #Systems and Control (eess.SY) #cs.SY #eess.SP #eess.SY #electronic engineering #information engineering

paper · pdf · doi:10.48550/arxiv.1910.08696

8 pages, submitted to IEEE Trans. Industrial Informatics

arxiv created 2019/10/19 · arxiv updated 2019/10/22

Abstract

Anomaly detection is an important task in power systems. To make better use of the phasor measurement unit (PMU) data collected from a low observability power system for anomaly detection, a data dimensionality increment algorithm is proposed in this paper. First, a low-dimensional spatio-temporal data matrix is formulated by using the synchrophasor measurements collected from a limited number of PMUs in a power system. Then, a data dimensionality increment algorithm based on random tensor theory (RTT) is proposed for anomaly detection. The proposed algorithm can help improve the sensitivity of random matrix theory (RMT) based and machine learning (ML) based anomaly detection approaches, and it is able to accelerate the convergence rate of model training in the ML based anomaly detection approach. Case studies on the IEEE 118-bus test system validate the effectiveness of the proposed algorithm.

Related

• Early Anomaly Detection in Power Systems Based on Random Matrix Theory
• Physics-Informed Machine Learning for Data Anomaly Detection, Classification, Localization, and Mitigation: A Review, Challenges, and Path Forward
• Real-time Anomaly Detection and Classification in Streaming PMU Data
• A Deep Learning Approach to Anomaly Sequence Detection for High-Resolution Monitoring of Power Systems
• A Robust and Explainable Data-Driven Anomaly Detection Approach For Power Electronics