A PMU-data-driven disruptive event classification in distribution systems

Niazazari, Iman; Livani, Hanif

doi:10.1016/j.epsr.2017.12.021

Cited by 46 publications

(18 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In [11], an event classifier is designed to classify partially labeled events that are extracted from micro-PMU data. Finally, in [12], event classifiers are developed for malfunctioned capacitor bank switching and malfunctioned voltage regulator events. The transient signatures of these malfunctions are obtained from simulations and not from real-world micro-PMUs data streams.…”

Section: A the Challenges And The Related Literaturementioning

confidence: 99%

A Machine Learning Approach to Event Analysis in Distribution Feeders Using Distribution Synchrophasors

Shahsavari

Farajollahi

Stewart

et al. 2019

2019 International Conference on Smart Grid Synchronized Measurements and Analytics (SGSMA)

View full text Add to dashboard Cite

This paper proposes a machine learning (ML) approach to detect, identify, and analyze the events that occur on distribution networks using data streams from real-world distribution-level phasor measurement units (PMUs). First, we develop two statistical event detection methods. One is based on testing absolute values around median and the other one is based on testing residuals on a non-linear estimation. Both methods use moving windows as well as dynamic window size. This allows us to detect events of different types and durations. Next, we use field expert knowledge to assign labels to the detected events and subsequently develop a multi-class support vector machine classifier to classify power quality events. Finally, we apply the above developed techniques to detect, identify, and analyze the events in a micro-PMU data stream from a real-world test site in Riverside, CA. We particularly study the oscillation events that occur somewhere across the distribution feeder itself, where their impacts are observed remotely by the available micro-PMUs.

show abstract

Section: A the Challenges And The Related Literaturementioning

confidence: 99%

A Machine Learning Approach to Event Analysis in Distribution Feeders Using Distribution Synchrophasors

Shahsavari

Farajollahi

Stewart

et al. 2019

2019 International Conference on Smart Grid Synchronized Measurements and Analytics (SGSMA)

View full text Add to dashboard Cite

show abstract

“…In [25], an autoencoder based neural network is proposed for the classication of the abnormal events in the distribution system,A wide variety of methods are available to train neural networks for anomaly classification, such as the scaled complex conjugate algorithm [21], improved generalized adaptive resonance theory [26], Marquardt Levenberg [27] and learning vector quantization combined with genetic algorithm [28]. Nevertheless, there are promising constrained optimization approaches in the mathematics literature that have not been used to train neural networks for classification applications.…”

Section: B Related Workmentioning

confidence: 99%

Event Classification in Distribution Networks Using a Quotient Gradient System

Khodabandehlou

Niazazari

Livani

et al. 2019

2019 North American Power Symposium (NAPS)

Self Cite

View full text Add to dashboard Cite

“…Stability criterion in Kumar et al, security criterion based on N − k redundancy in Nikkhah et al, observability during cascaded outages, and observability under various contingencies have been the objective function of several OPPs. Moreover, some advantages of PMUs in controlling the system in contingencies are investigated in the literature …”

Section: Introductionmentioning

confidence: 99%

How optimal PMU placement can mitigate cascading outages blackouts?

Karimi

Ebrahimi

Tavakoli

2019

Int Trans Electr Energ Syst

View full text Add to dashboard Cite

Summary The conventional optimal PMU placement (OPP) methods minimize the number of PMUs along with guarantying full observability of the system. However, various OPP schemes with the same number of PMUs obtained for a power system have different performances in emergency conditions. Cascading outages as the major cause of recent large blackouts cancel out full observability of the system and decrease situational awareness. In this paper, the OPA model previously developed to simulate power system blackouts is modified in order to include observability analysis through the progress of the cascading outages. The modified process is performed in turn for a given set of OPP schemes. To rank these schemes, a criterion is proposed as a monetary value using produced data by the modified OPA model. Scalability of the proposed procedure is investigated in two test systems. Simulations show that although considering N‐1 criterion in finding optimal PMU schemes decreases the consequences of transmission line cascading outages, it is not financially reasonable and does not lead to better performance of the OPP scheme. In other words, it is better to use OPP schemes in normal condition.

show abstract

A PMU-data-driven disruptive event classification in distribution systems

Cited by 46 publications

References 25 publications

A Machine Learning Approach to Event Analysis in Distribution Feeders Using Distribution Synchrophasors

A Machine Learning Approach to Event Analysis in Distribution Feeders Using Distribution Synchrophasors

Event Classification in Distribution Networks Using a Quotient Gradient System

How optimal PMU placement can mitigate cascading outages blackouts?

Contact Info

Product

Resources

About