Monitoring, Control, and Protection

Permanent URI for this collectionhttps://hdl.handle.net/10125/112469

Browse

Now showing 1 - 8 of 8

Granular Spatio-Temporal Grid-based Machine Learning for Forced Outage Prediction in Electrical Networks
(2026-01-06) Saranovic, Daniel; Obradovic, Zoran; Baembitov, Rashid; Kezunovic, Mladen
We developed a Granular Grid-based Outage Prediction Model, GG-OPM, that utilizes outage dependencies between electric grid substations and related feeders, benefiting from more granular weather data to improve outage prediction in the distribution network. The model incorporates a spatial and temporal module and tunes the hyperparameters to allow selecting components that optimize the task performance. Sixteen levels of spatio-temporal granularities were tested, using a large real-life dataset from a major Texas utility, which consisted of six years of historical outage events. The results obtained indicate an improvement of up to 50 times in F1 score over an earlier developed baseline for the most challenging prediction task, and an overall improvement across all levels of temporal and spatial granularity, providing solid grounds for utilizing the method in real-life outage prediction scenarios.
A Causality-Oriented Neural Network Framework for Transient Stability Prediction in Power Systems
(2026-01-06) Esmaeili-Nezhad, Amir; Cheshomi, Reza; Khorsand, Mojdeh
Transient stability prediction is crucial for dynamic security assessment in power systems. Conventional correlation-based Machine Learning (ML) approaches, while computationally efficient, often lack interpretability and fail to capture the true causal relationships among system variables. This paper presents a causality-oriented Artificial Neural Network (ANN) framework designed to address these limitations. First, a hybrid causal discovery method is proposed to identify physically meaningful causal relationships among system variables, enabling a principled selection of features that directly impact transient stability. Second, these causal insights are systematically integrated into ANN training through optimized feature selection and weighting schemes. The case study results verify that the proposed framework achieves superior generalization, reduces overfitting, and improves both predictive accuracy and interpretability across key evaluation metrics. By integrating causal discovery with neural network learning, the proposed approach offers a more robust solution for transient stability assessment and advances causality-aware ML applications in power system security.
Structure Preserving Dynamic Graphs for Power Systems
(2026-01-06) Ogbonna, Gerald; Anderson, C. Lindsay
Large-scale integration of renewable energy resources presents the challenge of coordinating the output of numerous small generators and loads. This coordination problem typically involves solving a large centralized optimization problem. The growing number of decision variables associated with renewable resources increases the computational complexity of this coordination problem. Several approaches address this issue by decomposing the centralized problem into smaller, more computationally tractable subproblems. In this work, we extend the Kron-reduced dynamic graph to a structure-preserving model using the framework of non-uniform Kuramoto oscillators. We demonstrate how slow coherency can be used to identify groups of dynamically coherent nodes on the IEEE 14-bus test system, which can then serve as a basis for decomposing the centralized coordination problem.
The Role of Flexible Connection in Accelerating Load Interconnection in Distribution Networks
(2026-01-06) Gu, Nan; Chen, Ge; Qin, Junjie
This paper investigates the role of flexible connection in accelerating the interconnection of large loads amid rising electricity demand from data centers and electrification. Flexible connection allows new loads to defer or curtail consumption during rare, grid-constrained periods, enabling faster access without major infrastructure upgrades. To quantify how flexible connection unlocks load hosting capacity, we formulate a flexibility-aware hosting capacity analysis problem that explicitly limits the number of utility-controlled interventions per year, ensuring infrequent disruption. Efficient solution methods are developed for this nonconvex problem and applied to real load data and test feeders. Empirical results reveal that modest flexibility, i.e., few interventions with small curtailments or delays, can unlock substantial hosting capacity. Theoretical analysis further explains and generalizes these findings, highlighting the broad potential of flexible connection.
Inferring PV Plant Reactive Power Control Mode from Ambient PMU Data
(2026-01-06) Mishra, Chetan; Vanfretti, Luigi; Pudasaini, Bikal; Delaree Jr., Jaime; Jones, Kevin
In the real-world, photovoltaic (PV) power plant controllers (PPCs) often experience internal changes (manually and/or automatically) in the form of control mode switching. Utilities or system operators are not notified when such changes occur, do not have remote access to the PPCs and may not even have knowledge of the different control modes under which a specific PPC may operate in. Since this information is not available, it is difficult to identify the critical operating mode that causes stability problems when conducting data-driven investigations. This paper proposes an approach to deduce the reactive power control mode of solar PV plants from ambient synchrophasor data by exploiting system identification techniques and statistical tests. The problem is posed as a test for conditional independence between multiple time series at the point of interconnection (POI). The effectiveness of this approach is demonstrated on synthetic as well as real world synchrophasor data from Dominion Energy’s power system.
Physical Mode Detection for Ambient and Ringdown Oscillation Modal Estimation Methods
(2026-01-06) Farrokhifard, Mohammadreza Maddipour; Arash Sarmadi, Seyed; Venkatasubramanian, Mani
This paper proposes a universal Physical Mode Detection (PMD) method for distinguishing valid system oscillation modes from spurious modal artifacts in time-domain modal estimation algorithms. This method utilizes the frequency and mode shape of each oscillation modal estimate, along with the frequency-domain information of selected measurements. The proposed PMD technique can be applied effectively to both ringdown and ambient modal analysis methods. The proposed PMD technique is tested on the modal analysis of synthetic measurements from Kundur test system simulations and archived real measurements from the western American power system. Prony and Matrix Pencil methods are utilized for illustrating PMD modal analysis of ringdown events, and the Covariance-based Stochastic Subspace Identification (SSI-Cov) method is employed for PMD analysis of ambient modal estimates. It is demonstrated that the proposed PMD method can distinguish between real estimates and spurious ones and is also compatible with various modal analysis techniques.
Introduction to the Minitrack on Monitoring, Control, and Protection
(2026-01-06) Follum, Jim; Venkatasubramanian, Mani
An Amplitude-Based Threshold Design Process for Reliable RMS-Energy Oscillation Detectors
(2026-01-06) Follum, Jim; Biswas, Shuchismita; Black, Clifton; Breuhl, Michael
Oscillation detection is one of the key reliability-enhancing capabilities enabled by phasor measurement unit (PMU) technology. Several grid operators have deployed commercial platforms that implement oscillation detection by monitoring the RMS-energy of measured signals. Oscillations are detected when the signal's energy crosses a predetermined threshold. The current process of setting these thresholds requires the analysis of at least three months of historical data, which is time consuming and expensive. The thresholding process seeks to avoid false alarms, but it does not evaluate the likelihood that a given oscillation amplitude will trigger detection. As a result, thresholds often require re-tuning to avoid nuisance alarms. This paper proposes a method for designing RMS-energy thresholds that directly integrates oscillation amplitudes selected by the practitioner. The method improves upon previously published work by providing the practitioner with 1) guidance on the range of amplitudes that will lead to a successful design, and 2) a complete summary of the threshold's expected performance. The method is explained and tested using field-measured PMU data to demonstrate its practical viability.

Browse

Recent Submissions