The Effect of Lead-Time Weather Forecast Uncertainty on Outage Prediction Modeling

Feifei Yang,Emmanouil N Anagnostou,Diego Cerrai

doi:10.3390/forecast3030031

Abstract

Weather-related power outages affect millions of utility customers every year. Predicting storm outages with lead times of up to five days could help utilities to allocate crews and resources and devise cost-effective restoration plans that meet the strict time and efficiency requirements imposed by regulatory authorities. In this study, we construct a numerical experiment to evaluate how weather parameter uncertainty, based on weather forecasts with one to five days of lead time, propagates into outage prediction error. We apply a machine-learning-based outage prediction model on storm-caused outage events that occurred between 2016 and 2019 in the northeastern United States. The model predictions, fed by weather analysis and other environmental parameters including land cover, tree canopy, vegetation characteristics, and utility infrastructure variables exhibited a mean absolute percentage error of 38%, Nash–Sutcliffe efficiency of 0.54, and normalized centered root mean square error of 68%. Our numerical experiment demonstrated that uncertainties of precipitation and wind-gust variables play a significant role in the outage prediction uncertainty while sustained wind and temperature parameters play a less important role. We showed that, while the overall weather forecast uncertainty increases gradually with lead time, the corresponding outage prediction uncertainty exhibited a lower dependence on lead times up to 3 days and a stepwise increase in the four- and five-day lead times.

Highlights

Extreme weather is a serious problem for electric distribution utilities, damaging power grid components and causing outages that can result in significant economic disruption and inconvenience for millions of customers [1,2]
We investigated historical outage events associated with 273 extratropical storms that took place between April 2016 and April 2018 and yielded 252,666 observations, integrating weather variables with information on utility infrastructure, land cover, vegetation, tree canopy, and utility-reported power outages for each storm event and modeling the power outages to a resolution of 1/32 degrees, covering the region
This section discusses the performance of the regional outage prediction model (OPM)

Summary

Introduction

Extreme weather is a serious problem for electric distribution utilities, damaging power grid components and causing outages that can result in significant economic disruption and inconvenience for millions of customers [1,2]. Nateghi et al continued this research and used the same data but used a random forest algorithm to develop an hurricane-inducedoutage prediction model, which showed an RMSE of 0.76 and 5.91 by grid level [15] They only used five hurricanes from a power distribution system in the Gulf Coast region of the United States. While the application of the different machine-learning models described in the literature has improved the ability to predict outages in the electric distribution network, we still need to understand how the accuracy of outage forecasts varies at different lead times.

Methods

Results

Conclusion