Improved Very-short-term Wind Forecasting using Atmospheric Classification

Abstract

We present a regime-switching vector-autoregressive method for very-short-term wind speed forecasting at multiple locations with regimes based on large-scale meteorological phenomena. Statistical methods short-term wind forecasting out-perform numerical weather prediction for forecast horizons up to a few hours, and the spatio-temporal interdependency between geographically dispersed locations may be exploited to improve forecast skill. Here we show that conditioning spatio-temporal interdependency on `atmospheric modes' can further improve forecast performance. The modes are defined from the atmospheric classification of wind and pressure fields at the surface level, and the geopotential height field at the 500hPa level. The data fields are extracted from the MERRA-2 reanalysis dataset with an hourly temporal resolution over the UK, atmospheric patterns are classified using self-organising maps and then clustered to optimise forecast performance. In a case study based on 6 years of measurements from 23 weather stations in the UK, a set of three atmospheric modes are found to be optimal for forecast performance. The skill in the one- to six-hour-ahead forecasts is improved at all sites compared to persistence and competitive benchmarks. Across the 23 test sites, one-hour-ahead root mean squared error is reduced by between 0.3% and 4.1% compared to the best performing benchmark, and by and average of 1.6% over all sites; the six-hour-ahead accuracy is improved by an average of 3.1%.