Resilient Networks

Permanent URI for this collection

Browse

Recent Submissions

Now showing 1 - 7 of 7
  • Item
    Towards the Automation of Remedial Action Schemes Design
    ( 2021-01-05) Li, Hanyue ; Shetye, Komal ; Overbye, Thomas ; Davis, Katherine ; Hossain-Mckenzie, Shamina
    The constantly evolving nature of the grid is compelling the design process of Remedial Action Schemes (RAS) to keep up with the changes. This paper proposes a flexible and computationally efficient approach to automatically determine RAS corrective actions that alleviate line overloading violations. Statistical and functional characteristics summarized from RAS implemented in real power systems are used to guide the design parameters. This paper also leverages sensitivity-based techniques to determine corrective actions for specific contingencies quickly without repeated numerical simulations.
  • Item
    Toward Preventing Cascading Blackouts: Vulnerability and Criticality Stress Metrics
    ( 2021-01-05) Hossain, Md ; Davies, Donald ; Bodson, Marc ; Merrill, Hyde
    An approach is presented for measuring the susceptibility of power systems to cascading blackouts. A failure network and an analysis method are proposed, based on line outage distribution factors. The structure and properties of this network are analyzed with metrics that quantify stress. The metrics can be computed rapidly, so that they are feasible for use in real-time operations as well as for planning and post-mortems. The studies presented are remarkable for use of real, large-scale data for the entire Western Interconnection of North America, with more than 26,000 branches. The metrics are consistent with expectations in different seasonal and loading conditions. Metrics for a pre-blackout reconstruction show that the system was highly stressed and correctly identified the most vulnerable and critical areas and branches. Lowering stress is a serious candidate for reducing risk of cascading.
  • Item
    Stochastic Optimization of Power System Dynamics for Grid Resilience
    ( 2021-01-05) Arguello, Bryan ; Stewart, Nathan ; Hoffman, Matt
    When faced with uncertainty regarding potential failure contingencies, prioritizing system resiliency through optimal control of exciter reference voltage and mechanical torque can be arduous due to the scope of potential failure contingencies. Optimal control schemes can be generated through a two-stage stochastic optimization model by anticipating a set of contingencies with associated probabilities of occurrence, followed by the optimal recourse action once the contingency has been realized. The first stage, common across all contingency scenarios, co-optimally positions the grid for the set of possible contingencies. The second stage dynamically assesses the impact of each contingency and allows for emergency control response. By unifying the optimal control scheme prior and post the failure contingency, a singular policy can be constructed to maximize system resiliency.
  • Item
    PMU-based Line Differential Protection under GPS Spoofing Attack
    ( 2021-01-05) Xie, Jiahao ; Meliopoulos, A. P. Sakis ; Cokkinides, George
    Many power grid applications rely on the time synchronized measurement data collected by phasor measurement units (PMUs) and or Merging Units (MUs)). The synchronization of data from multiple PMUs and MUs relies on the global positioning system (GPS) to provide the time reference. The civilian GPS receivers used in PMU operate according to the publicly available mechanism. Therefore, attackers can easily generate false GPS signals with low-cost portable devices. During the spoofing attacks, the signal received by the PMUs and MUs is arbitrarily modified by the attacker without being detected, leading to the malfunction of other applications in the power system including transmission line protection. In this paper, we investigate the effect of GPS spoofing attack on transmission line differential protection schemes based on PMUs, including percentage differential method and alpha-plane protection method. The mis-operation of these relays is observed during GPS spoofing attacks. In addition, we propose a quasi-dynamic state-estimation-based method to detect GPS spoofing attacks, identify the affected PMUs, correct the compromised data, and enable the continuous and reliable operation of the relays. Numerical results show that the proposed method detects the GPS spoofing successfully and the recovered measurement data eliminate the misoperation of line differential protection.
  • Item
    Characteristics and Risk of Microgrid Outages from a Complex Systems Point of View
    ( 2021-01-05) Bowker, Anna ; Newman, David ; Carreras, Benjamin ; Huang, Daisy ; Dobson, Ian ; Koplin, Clay
    Cordova is a town of approximately 2,000 people located on the southern coast of Alaska. A power grid for a town this size, with a large seasonal fishing economy, is considered a moderate to large sized microgrid in terms of power produced. Understanding the vulnerabilities and risks of failures in such a grid is important for planning and operations. Investigating these characteristics in the context of complex system dynamics is a novel approach. The analysis of Cordova’s microgrid is a case study relevant to a large class of microgrid communities. We analyze the outage data based on size, cause characteristics and load demand on the system and find long time correlations and power laws in the failure size distributions. Finally we apply a risk metric to give a single numerical value to the risk of an outage occurring during certain time periods and under certain conditions.
  • Item
    A Neural Lyapunov Approach to Transient Stability Assessment in Interconnected Microgrids
    ( 2021-01-05) Huang, Tong ; Gao, Sicun ; Long, Xun ; Xie, Le
    We propose a neural Lyapunov approach to assessing transient stability in power electronic-interfaced microgrid interconnections. The problem of transient stability assessment is cast as one of learning a neural network-structured Lyapunov function in the state space. Based on the function learned, a security region is estimated for monitoring the security of interconnected microgrids in real-time operation. The efficacy of the approach is tested and validated in a grid-connected microgrid and a three-microgrid interconnection. A comparison study suggests that the proposed method can achieve a less conservative characterization of the security region, as compared with a conventional approach.
  • Item