Powering Up With Space-Time Wind Forecasting

Abstract

The technology to harvest electricity from wind energy is now advanced enough to make entire cities powered by it a reality. High-quality short-term forecasts of wind speed are vital to making this a more reliable energy source. Gneiting et al. (2006) have introduced a model for the average wind speed two hours ahead based on both spatial and temporal information. The forecasts produced by this model are accurate, and subject to accuracy, the predictive distribution is sharp, i.e., highly concentrated around its center. However, this model is split into nonunique regimes based on the wind direction at an off-site location. This paper both generalizes and improves upon this model by treating wind direction as a circular variable and including it in the model. It is robust in many experiments, such as predicting at new locations. We compare this with the more common approach of modeling wind speeds and directions in the Cartesian space and use a skew-t distribution for the errors. The quality of the predictions from all of these models can be more realistically assessed with a loss measure that depends upon the power curve relating wind speed to power output. This proposed loss measure yields more insight into the true value of each model's predictions.

Probabilistic forecasting of the wave energy flux

“…This is also the case with wind model dynamics, as proposed by Gneiting et al [11], Hering and Genton [12] and Pinson [10] among others. In essence, the model includes a physics component based on the ECMWF model forecasts (θ 1xt+k|t ), and an autoregressive component (…”

A Valid Matérn Class of Cross-Covariance Functions for Multivariate Random Fields With Any Number of Components

“…Accurate forecasts of wind speed are of crucial importance in many applications such as harvesting electricity from wind energy (Genton and Hering, 2007;Hering and Genton, 2010) and severe weather warnings for the general public. For the three variables at the 100 locations, we consider a mean zero Gaussian random field…”

Using Conditional Kernel Density Estimation for Wind Power Density Forecasting

“…They propose spatial and temporal regime switching models using a vector autoregressive idea and a truncated Gaussian distribution for two-hour ahead prediction. For the same data, Hering and Genton (2010) consider a univariate wind speed model incorporating wind direction in trigonometric forms using the truncated Gaussian distribution, and also a bivariate model for wind speed and direction, in Cartesian coordinate form, using a skewed t distribution.…”

“…Advantages of using time series models are: (i) they are no less accurate than atmospheric modeling for short-term predictions (Focken et al 2002); (ii) they can produce forecasts from any time origin and for any lead time, which contrasts with atmospheric model ensemble predictions (see, for example, Taylor et al, 2009);(iii) acquiring forecasts from an atmospheric model can be costly; and (iv) predictions from such models are often not available for the wind farm location of interest. Following the work of Cripps and Dunsmuir (2003) and Hering and Genton (2010), we use a bivariate VARMA-GARCH time series model, which enhances wind speed prediction through the joint modeling of wind speed and direction. We convert the resulting wind speed density predictions into wind power density forecasts using Monte Carlo simulation and conditional kernel density (CKD) estimation (see Parzen 1962;Rosenblatt 1969;Hyndman et al 1996), which enables a nonparametric modeling of the conditional density of wind power.…”