Monitoring, Control, and Protection

Permanent URI for this collection

Browse

Recent Submissions

Now showing 1 - 5 of 8
  • Item
    Simultaneous Forward-Backward Prony Estimation
    ( 2022-01-04) Follum, Jim ; Tuffner, Francis ; Khan, Md. Arif ; Etingov, Pavel
    Power system dynamic stability can be evaluated through the analysis of transient oscillations that occur following significant system events. One of the earliest methods for this type of study is Prony analysis, which estimates the system's electromechanical modes. While previous studies have highlighted advantages of performing Prony analysis on data in the forward and backward directions, the proposed method does so simultaneously. As a result, signal poles corresponding to electromechanical modes can be distinguished from spurious poles more reliably. The method also produces a single mode estimate, where independent application in the forward and backward directions would produce two estimates for each mode. The method is validated using simulated and measured power system data.
  • Item
    Power Spectrum Estimation for Frequency Domain Ambient Modal Analysis
    ( 2022-01-04) Venkatasubramanian, Mani ; Thomas, Chad ; Farrokhifard, Mohammadreza Maddipour
    This paper studies the effect of Power Spectrum Density (PSD) estimation techniques on the accuracy of Fast Frequency Domain Decomposition (FFDD) modal analysis. FFDD utilizes ambient synchrophasor measurements to estimate characteristics of dominant system modes and oscillations by analyzing the PSD estimates from multiple synchrophasor measurements. In this paper, the impact of three different methods for PSD estimation on the accuracy of FFDD modal estimates is investigated: PWelch, MultiTaper Method (MTM) using Slepian Tapers, and MTM using Sine Tapers. Tests are done using synthetic and archived synchrophasor data. All three PSD methods are shown to work well for oscillation detection of sustained oscillations using FFDD. However, for ambient modal analysis, it is shown that FFDD based on MTM with Slepian Tapers has the most reliable modal estimations. FFDD using both MTM with Sine Tapers and PWelch have bias issues in estimating well-damped system modes, requiring more research for them to be suitable for FFDD.
  • Item
    Physics Informed Reinforcement Learning for Power Grid Control using Augmented Random Search
    ( 2022-01-04) Mahapatra, Kaveri ; Fan, Xiaoyuan ; Li, Xinya ; Huang, Yunzhi ; Huang, Qiuhua
    Wide adoption of deep reinforcement learning in energy system domain needs to overcome several challenges , including scalability, learning from limited samples, and high-dimensional continuous state and action spaces. In this paper, we integrated physics-based information from the generator operation state formula, also known as Swing Equation, into the reinforcement learning agent's neural network loss function, and applied an augmented random search agent to optimize the generator control under dynamic contingency. Simulation results demonstrated the reliability performance improvements in training speed, reward convergence, and future potentials in its transferability and scalability.
  • Item
    On the Verification of Deep Reinforcement Learning Solution for Intelligent Operation of Distribution Grids
    ( 2022-01-04) Hosseini, Mohammad Mehdi ; Parvania, Masood
    Capabilities of deep reinforcement learning (DRL) in obtaining fast decision policies in high dimensional and stochastic environments have led to its extensive use in operational research, including the operation of distribution grids with high penetration of distributed energy resources (DER). However, the feasibility and robustness of DRL solutions are not guaranteed for the system operator, and hence, those solutions may be of limited practical value. This paper proposes an analytical method to find feasibility ellipsoids that represent the range of multi-dimensional system states in which the DRL solution is guaranteed to be feasible. Empirical studies and stochastic sampling determine the ratio of the discovered to the actual feasible space as a function of the sample size. In addition, the performance of logarithmic, linear, and exponential penalization of infeasibility during the DRL training are studied and compared in order to reduce the number of infeasible solutions.
  • Item
    Line Faults Classification Using Machine Learning on Three Phase Voltages Extracted from Large Dataset of PMU Measurements
    ( 2022-01-04) Otudi, Hussain ; Dokic, Tatjana ; Mohamed, Taif ; Kezunovic, Mladen ; Hu, Yi ; Obradovic, Zoran
    An end-to-end supervised learning method was developed to classify transmission line faults in a two-year field-recorded dataset that includes synchronized measurements of three-phase voltages recorded by 38 phasor measurement units (PMUs) sparsely located in the US Western Grid interconnection. Statistical analysis was performed to extract features from this large dataset to train the support vector machine (SVM), random forest (RF), and extreme gradient boosting (XGBoost) classifiers. The training further leverages a simulated dataset from a synthetic grid with 12 PMUs to increase the number of types of faults infrequently seen in the field-recorded dataset. Training the classification models with the combined dataset resulted in a classification accuracy of 98.58%. This is a significant improvement over 86.87% to 87.17% accuracy obtained by relying on the field-recorded dataset alone.