A Partnership of UConn and Eversource
Eversource Energy Center
Center Research on Storm Outage Modeling Published in “Risk Analysis,” an International Journal
Jichao He and his UConn Department of Civil & Environmental Engineering coauthors, together with the Department of Statistics at Brigham Young University, have published a peer-reviewed article on storm outage modeling in the journal Risk Analysis. The “Nonparametric Tree-Based Predictive Modeling of Storm Outages on an Electric Distribution Network” research presents the performance of two algorithms for predicting storm outages (defined as locations requiring manual intervention for restoring power.) The research demonstrated that while one model, Quantile Regression Forest (QRF), was effective for predicting outages at high spatial resolutions (e.g. 2 kilometer grid cells and towns), the Bayesian Additive Regression Trees (BART) model more accurately aggregates predictions at coarser resolutions (e.g. service territory resolution).
To directly access the article, please click here:
http://onlinelibrary.wiley.com/doi/10.1111/risa.12652/abstract
Congratulations to Jichao and his coauthors!
Published: June 20, 2016
Categories: Recent News
Available Archives