Search

Some of the deadliest wildfires in the U.S., such as California’s 2018 wildfires, have been ignited by power systems. In an effort to prevent and minimize the ignition of wildfires, or control them if ignited, energy companies have developed wildfire mitigation plans. This paper provides energy companies and power system operators, engineers, researchers, and suppliers an overview of the state-of-the-art studies that address key topics in these wildfire mitigation plans and compares the wildfire mitigation plans of several energy companies. The key topics include grid design and system hardening, asset management and inspection, situational awareness and forecasting, operational response, vegetation management, public safety power shutoff, and risk-spend efficiency. This paper also presents a comparison of several energy companies’ decision-making criteria for initiating a public safety power shutoff. Finally, we discuss opportunities for future research studies that could help energy companies prevent wildfire ignitions. [ABSTRACT FROM AUTHOR]

This paper proposes a modeling approach for characterizing the probability of wildfire ignition caused by faults on power distribution systems. The proposed model serves as a starting point in research literature to illustrate, from an analytical perspective, the many factors that influence wildfire ignitions in power distribution systems. This paper presents the series of events that leads to power-system-related wildfire ignitions, and characterizes the wildfire ignition probability as a combination of the probability that a fault occurs along a power distribution line segment, and the probability that the fault results in the sustained ignition of a vegetation fuel bed surrounding the line. The proposed model integrates a variety of data including environmental conditions, power system protection settings, and power system line flows. A case study is performed on a test 33-bus distribution system using observed historical weather data from a high-threat fire district in California. The California case study is utilized to investigate the effects of three primary factors (wind speed, line congestion, and protection settings) on wildfire ignition probability. [ABSTRACT FROM AUTHOR]

This paper presents a methodological framework to evaluate the resilience, with the primary focus on performance degradation, of levee-protected electric power networks to flooding in a changing climate. To this end, a multi-disciplinary framework is established by integrating climate science, hydrology, and electric power network analysis. The framework quantifies the effect of climate change on flood hazard levels in a levee-protected area and the subsequent changes in the resilience of the electric power network. In the first step, the changes in the exposure of levee-protected regions to flooding hazard in a warming climate are calculated. Then, the probabilities of failure of power network components due to flooding under current and projected future climate are determined. Finally, the power system resilience index is used to assess the system resiliency for pre-flooding (baseline), historic flooding, and two projected future flooding scenarios. For demonstration, the proposed framework is applied to a levee-protected area in Northern California. The IEEE 118-bus standard test system is employed to represent the power network in the study area. Results reveal that climate change can considerably decrease the system resilience of the levee-protected electric power network. The findings of this study can contribute towards more resilient power network systems under a changing climate. [ABSTRACT FROM AUTHOR]

Channel allocation has a direct and profound impact on the performance of vehicle-to-everything (V2X) networks. Considering the dynamic nature of vehicular environments, it is appealing to devise a blended strategy to perform effective resource sharing. In this paper, we exploit deep learning techniques predict vehicles’ mobility patterns. Then we propose an architecture consisting of centralized decision making and distributed channel allocation to maximize the spectrum efficiency of all vehicles involved. To achieve this, we leverage two deep reinforcement learning techniques, namely deep Q-network (DQN) and advantage actor-critic (A2C) techniques. In addition, given the time varying nature of the user mobility, we further incorporate the long short-term memory (LSTM) into DQN and A2C techniques. The combined system tracks user mobility, varying demands and channel conditions and adapt resource allocation dynamically. We verify the performance of the proposed methods through extensive simulations and prove the effectiveness of the proposed LSTM-DQN and LSTM-A2C algorithms using real data obtained from California state transportation department. [ABSTRACT FROM AUTHOR]