Hierarchical Load Hindcasting Using Reanalysis Weather

By leveraging recent advances in atmospheric reanalysis it is possible to more fully characterize the effects of low frequency weather phenomena simultaneously affecting the native load and power output of weather-sensitive generators. To this end, this paper describes load “hindcasting”-a method of using reanalysis data to re-synthesize multiple decades of historical load data such that it represents a current and consistent load profile. When used together with coincident, reanalysis-derived records of weather-sensitive power output, load hindcasting enables a robust, long-term characterization of these resources that accounts for weather variability spanning decades. Drawing from the field of short-term load forecasting, hierarchical load hindcasting models are developed for summer weekday hours in New England using weather variables from the Modern Era Retrospective-Analysis for Research and Applications (MERRA) dataset developed by the National Aeronautics and Space Administration (NASA). Results demonstrate the efficacy of hindcasting realistic hourly loads using MERRA.