TY - GEN
T1 - A Noise-Tolerant Temporal-Adaptive Approach for Short-Term Voltage Stability Assessment of Power Systems
AU - Zhang, Yuchen
AU - Dong, Zhao Yang
AU - Zhang, Rui
AU - Xu, Yan
N1 - Publication details (e.g. title, author(s), publication statuses and dates) are captured on an “AS IS” and “AS AVAILABLE” basis at the time of record harvesting from the data source. Suggestions for further amendments or supplementary information can be sent to [email protected].
PY - 2018/9/18
Y1 - 2018/9/18
N2 - Short-term voltage stability (STVS) problem has become more prominent in today's power systems, and the deployment of phasor measurement units (PMU) opens the way for real-time STVS assessment. The existing approaches for real-time STVS assessment suffer from long observation windows and lack mitigation to the impact of PMU measurement noise. Considering those inadequacies, this paper proposes a noise-tolerant temporal-adaptive STVS assessment approach with the following salient features: 1) the adoption of probabilistic classification enables the early unstable event detection ability, which significantly improves the assessment speed; 2) the probabilistic classifier consists of noisy-tolerant ensemble models that are able to learn the impact of measurement noise, which improves the assessment robustness against measurement noise. The proposed approach is tested on New England 39-bus system, and its fast assessment speed and excellent robustness against measurement noise are verified by comparative studies with existing approaches. © 2018 IEEE.
AB - Short-term voltage stability (STVS) problem has become more prominent in today's power systems, and the deployment of phasor measurement units (PMU) opens the way for real-time STVS assessment. The existing approaches for real-time STVS assessment suffer from long observation windows and lack mitigation to the impact of PMU measurement noise. Considering those inadequacies, this paper proposes a noise-tolerant temporal-adaptive STVS assessment approach with the following salient features: 1) the adoption of probabilistic classification enables the early unstable event detection ability, which significantly improves the assessment speed; 2) the probabilistic classifier consists of noisy-tolerant ensemble models that are able to learn the impact of measurement noise, which improves the assessment robustness against measurement noise. The proposed approach is tested on New England 39-bus system, and its fast assessment speed and excellent robustness against measurement noise are verified by comparative studies with existing approaches. © 2018 IEEE.
KW - ensemble learning
KW - measurement noise
KW - phasor measurement unit
KW - probabilistic classification
KW - short-term voltage stability
UR - http://www.scopus.com/inward/record.url?scp=85055571357&partnerID=8YFLogxK
UR - https://www.scopus.com/record/pubmetrics.uri?eid=2-s2.0-85055571357&origin=recordpage
U2 - 10.1109/ISGT-Asia.2018.8467941
DO - 10.1109/ISGT-Asia.2018.8467941
M3 - RGC 32 - Refereed conference paper (with host publication)
SN - 9781538642917
T3 - International Conference on Innovative Smart Grid Technologies, ISGT Asia 2018
SP - 62
EP - 67
BT - International Conference on Innovative Smart Grid Technologies, ISGT Asia 2018
PB - IEEE
T2 - 2018 International Conference on Innovative Smart Grid Technologies, ISGT Asia 2018
Y2 - 22 May 2018 through 25 May 2018
ER -