Adaptive Two-Stage Bayesian Model Averaging for Estimating the Impact of Hazards on Power System Service
A wide variety of weather conditions, from windstorms to prolonged heat events, can have substantial impacts on power systems, posing many risks and inconveniences due to power outages. Being able to accurately estimate the probability distribution of the number of customers without power by using data about the power utility system and environmental and weather conditions has the potential to help utilities restore power more quickly and efficiently. We propose a new algorithm based on Bayesian model averaging (BMA) in order to form an ensemble model predicting daily distributions of customers interruptions. The proposed algorithm has three main characteristics: (i) the base learners can be any type of probabilistic prediction model and no limiting assumptions are made about them, (ii) weights of the base learners in the BMA are a multinomial logistic function of the data and not necessarily the same for different records, and (iii) the base learners and their corresponding weights are updated as new data are collated. Using a large, real dataset of daily customers interruptions, we demonstrate that our approach offers more accurate probabilistic predictions than traditional approaches. It also provides more insights into the data generating process, and so, results in better support for utility restoration planning. Although our work is motivated by the power system application, our methodology and insights can be extended to other predictive modeling problems in which there are model uncertainty and data is collated gradually
Year of publication: |
[2021]
|
---|---|
Authors: | Kabir, Elnaz ; Guikema, Seth ; Quiring, Steven |
Publisher: |
[S.l.] : SSRN |
Description of contents: | Abstract [papers.ssrn.com] |
Saved in:
Extent: | 1 Online-Ressource |
---|---|
Type of publication: | Book / Working Paper |
Language: | English |
Notes: | Nach Informationen von SSRN wurde die ursprüngliche Fassung des Dokuments December 7, 2020 erstellt Volltext nicht verfügbar |
Source: | ECONIS - Online Catalogue of the ZBW |
Persistent link: https://www.econbiz.de/10013247467
Saved in favorites
Similar items by person
-
Importance of soil and elevation characteristics for modeling hurricane-induced power outages
Quiring, Steven, (2011)
-
Forecasting hurricane-induced power outage durations
Nateghi, Roshanak, (2014)
-
Simulation of tropical cyclone impacts to the U.S. power system under climate change scenarios
Staid, Andrea, (2014)
- More ...