Probabilistic performance assessment of power distribution infrastructure under wind events

Abstract The electric power delivery infrastructure directly contributes to essential societal functions, the economy, and the general quality of life. Recent extreme wind events, such as Hurricanes Irma and Maria, and more recently Hurricane Michael resulted in power outages that affected millions of customers and led to major social and economic disruptions throughout communities. The longer the outage duration, the greater the incurred losses. Power distribution systems have proven to be highly vulnerable to such events and responsible for 90% of the outages. Distribution structures are built according to safety standards to ensure the safety of assets in operational and extreme conditions, but the dynamic nature of the wind loading is often overlooked or only considered empirically. Therefore, in this paper, such effects are included in a comprehensive risk-informed framework to assess the performance of electric power distribution components. This methodology explicitly accounts for the inevitable uncertainty in predictions of the component response. The focus is on characterizing wind events and the component-level risk analysis of physical components of the electric power system. For this purpose, a power pole–conductor system is modeled in which wind events are simulated as one-dimensional multivariable stochastic processes along the height of the poles and the conductors. Three-second peak gust wind speeds are used as a modeling reference. A probabilistic framework is developed to estimate the capacity of the distribution poles under different aging mechanisms. The response of the power distribution poles due to the simulated wind speeds is then computed using a finite element analysis. Fragility functions are generated to estimate the probability of exceedance for different damage states. A set of hazard mitigation strategies, the associated costs, and estimated benefits are implemented, and the improved fragility functions for elements are generated. A probabilistic life-cycle cost analysis is used to assess the long-term benefits of investing in different components of the power distribution system.