Chetan Mishra

This paper introduces a fully automated, unsupervised technique for detecting step changes in synchrophasor measurements, without prior knowledge or learning data features, without applying the complex task of baselining, parameter tuning. Step changes in synchrophasor measurements occur due to shunt switching, controller set point changes, and etc. With increasing grid oscillations from inverter-based resources (IBRs) step changes are vital for tracking grid response and anticipating IBR controller-related oscillations. This study identifies step changes through a nonorthogonal discrete wavelet transform that relies on smoothed gradient estimation. A non-linear filter based on a multiscale point-wise product of wavelet coefficients is proposed, which takes advantage of the broadband feature of step changes within the space of wavelet coefficients. This filter also eliminates unwanted signal elements, thus decreasing the occurrence of false positives. Finally, through the statistical characterization of multi-scale products, we use an adaptive multiscale dependent threshold for detecting steps amidst real-world measurement noise and grid oscillations. Our studies indicate the effectiveness of our method in detecting step changes in synchrophasor data, supported by showcasing real-world case studies and discussing practical applications. Our method also significantly outperforms the relevant baseline detection methods by measuring Accuracy, F1 Score, and Precision under both ambient (noise) and step change conditions.

This letter presents an analysis of a non-sinusoidal response from a transmission-level photovoltaic (PV) solar power plant triggered by a local line outage in Dominion Energy's power system. The uniqueness of this response resides in the fact that its critical component is non-sinusoidal, and unlike typical scenarios, it does not stem from conventional nonlinear power system dynamics. It is also demonstrated how careful and systematic analysis of synchrophasor measurements in the plant, through time-series analysis techniques, can help illuminate the underlying mechanism leading to the atypical response observed.

The North American Electric Reliability Corporation (NERC) is expected to mandate model validation of power plant equipment in the near future. This will create a need to validate models for a large fleet of existing and future power plants. Historically, model validation of synchronous generators, excitation system, turbine governor, and other power system equipment has been conducted in diverse platforms. As a contribution to the power system model implementation using Modelica language and validation against commercial tools this work continues to develop power system component models and enriching the Open-Instance Power System Library (OpenIPSL). As a part of the development of OpenIPSL this paper describes the development of models used by North American utilities that follow NERC modeling requirements, including models of a synchronous generator, an excitation system, a turbine and governor using Modelica language in Dymola. The component implementation process is described and the validation of the models implemented in Modelica against PSS/E using both a single machine infinite bus (SMIB) and multi-machine system models is illustrated.

In power systems, subsynchronous oscillations associated with the interaction between a mechanical rotor shaft and electrical system can lead to equipment damage if improperly mitigated. This paper describes the development of a scalable, multi-mass torsional shaft model and a synchronous machine model with DC offset torque components included using Modelica. When coupled, these models can be used to perform shaft torsional studies. Two methods of coupling the shaft with the rest of the turbine-generator system are devised and analyzed. A single-machine, infinite-bus test system using the torsional shaft model and generator model developed in this paper is proposed to observe the penetration of subsynchronous oscillations throughout an electrical system. The test system is then modified to model subsynchronous resonance leading to system instability. Analysis of the models described in this paper highlights the value of the Modelica_LinearSystems2 library in determining the torsional mode shapes and frequencies associated with a turbine-generator system model.

This paper presents the authors’ experience in using the Dynamical Mode Decomposition (DMD) for estimating electromechanical oscillation modes from real synchrophasor data from the Dominion Energy system. To gain insight into the effectiveness of this approach, the results obtained from DMD on ambient data are compared against frequency domain approaches. Furthermore, the challenges in using this method are highlighted. The comparison shows that DMD is suitable for mode estimation from ambient data and should be further investigated for real world deployment within wide-area monitoring systems.

Recent expansion of inverter-based renewable generation (RG) has introduced unprecedented stability issues. One major challenge stems from tripping RGs offline during disturbances based on predefined operation thresholds (ride through curves) on voltage magnitudes and frequency at the point of common coupling (PCC). The current practice uses a single set of thresholds for the whole year across all RGs in the system. This could result in loss of these resources during critical times, particularly endangering system voltage security, i.e a significant reduction in the load margin (LM). On the other hand, relaxing these limits could result in them staying online even during islanding scenarios, which is a safety concern to the restoration crew. Thus, an approach to find minimal relaxations to these limits to enhance voltage security is proposed here. This process first constructs an optimization model aiming to minimize adjustments on RG voltage limits to reach anticipated LM increase and then solves it with a multi-stage screening method including base power-voltage (P-V) curve acquisition, critical RGs identification, successive screenings of candidate solutions based on sensitivities, and optimal solutions validation. This largely increases the scalability of the algorithm, which is demonstrated on the IEEE 14-bus, 118-bus systems, and a 500-bus system.

The wrapping operation causes discontinuities in phase angle measurements from synchrophasor devices. While various unwrapping algorithms traditionally used for synchrophasor data have proven effective under typical operating conditions, they are likely to fail when faced with atypical data because of equipment malfunctions. For example, the underestimation of the phase angle jump when the data originates from a device with an internal clock or other timing mechanisms can introduce this kind of error. This letter proposes using an unscented Kalman filter (UKF) for unwrapping one-dimensional synchrophasor data. This technique is especially suitable for analyzing low-quality data. The effectiveness of this method is demonstrated on real phase angle measurements from devices with documented timing issues in the Dominion Energy power system.