A Learning Scheme for Microgrid Reconnection

Abstract

This paper introduces a potential learning scheme that can dynamically predict the stability of the reconnection of sub-networks to a main grid. As the future electrical power systems tend towards smarter and greener technology, the deployment of self sufficient networks, or microgrids, becomes more likely. Microgrids may operate on their own or synchronized with the main grid, thus control methods need to take into account islanding and reconnecting of said networks. The ability to optimally and safely reconnect a portion of the grid is not well understood and, as of now, limited to raw synchronization between interconnection points. A support vector machine (SVM) leveraging real-time data from phasor measurement units (PMUs) is proposed to predict in real time whether the reconnection of a sub-network to the main grid would lead to stability or instability. A dynamics simulator fed with pre-acquired system parameters is used to create training data for the SVM in various operating states. The classifier was tested on a variety of cases and operating points to ensure diversity. Accuracies of approximately 85% were observed throughout most conditions when making dynamic predictions of a given network.

Resilience in an Evolving Electrical Grid

“…With enough data available, one can also use these datasets as sources of features and train modern machine learning approaches for predicting and quantifying risk [169,170]. Machine learning and artificial intelligence approaches also can provide timely recommendations to the operator in charge of remedial actions [171,172].…”

On PMU Data Integrity under GPS Spoofing Attacks: A Sparse Error Correction Framework

“…This algorithm is designed to operate on networks that are observable from the PMU measurements 3 . Vanfretti et al [18] develops a state estimation technique based on PMU measurements by incorporating potential phase bias errors in PMU measurements 4 . This algorithm is designed for decentralized operation wherein it can be independently applied to correct PMU data in observable islands within an unobservable network.…”

“…Due to this enriched measurement quality, there has been a wide interest in developing approaches to leverage PMU measurements for real-time power grid monitoring, protection, and control [2], [3]. While several PMU-based approaches showed improved performance compared to the legacy approaches [4], [5], those promises can be realized only if the data integrity of PMUs can be ensured.…”

“…The authors in[16] extend their algorithm to networks unobservable by PMU measurements by augmenting both PMU and SCADA measurements. Since our focus is on PMU data correction, our comparison is only with their algorithm that uses only PMU measurements4 We set PMU 102 as the "reference bus" defined in the paper[18]. This PMU is assumed to be intact from PMU attacks…”

Learning Schemes for Power System Protection

“…We draw upon previous work that predicts the stability of network reconnection with limited PMU measurements using Support Vector Machines (SVMs) [28]. To improve upon this work we use a neural network when building a classifier.…”