A Bayesian Approach Based Outage Prediction in Electric Utility Systems Using Radar Measurement Data
Severe weather events such as strong thunderstorms are some of the most significant and frequent threats to the electrical grid infrastructure. Outages resulting from storms can be very costly. While some tools are available to utilities to predict storm occurrences and damage, they are typically very crude and provide little means of facilitating restoration efforts. This study developed a methodology to use historical high-resolution (both temporal and spatial) radar observations of storm characteristics and outage information to develop weather condition dependent failure rate models (FRMs) for different grid components. Such models can provide an estimation or prediction of the outage numbers in small areas of a utility’s service territory once the real-time measurement or forecasted data of weather conditions become available as the input to the models. Considering the potential value provided by real-time outages reported, a Bayesian outage prediction (BOP) algorithm is proposed to account for both strength and uncertainties of the reported outages and failure rate models. The potential benefit of this outage prediction scheme is illustrated in this study.
- Research Organization:
- Brookhaven National Laboratory (BNL), Upton, NY (United States)
- Sponsoring Organization:
- New York State Energy Research and Development Authority (NYSERDA); USDOE
- Grant/Contract Number:
- SC0012704
- OSTI ID:
- 1389221
- Report Number(s):
- BNL-114128-2017-JA
- Journal Information:
- IEEE Transactions on Smart Grid, Vol. 9, Issue 6; ISSN 1949-3053
- Publisher:
- IEEECopyright Statement
- Country of Publication:
- United States
- Language:
- English
Web of Science
Similar Records
Predicting Storm Outages Through New Representations of Weather and Vegetation
ASSIMILATION OF DOPPLER RADAR DATA INTO NUMERICAL WEATHER MODELS