Accurate probabilistic forecasting of wind power output is critical to maximizing network integration of this clean energy source. New research by Professors Arrieta-Prieto and Schell, published in the International Journal of Forecasting, developed a spatio-temporal framework that outperformed competitive models in both the univariate and multivariate contexts. It can provide a simple and reliable forecasting mechanism that efficiently handles the challenging aspects of wind power data, by (1) introducing nonlinear transformations of the data and sinusoidal basis functions to handle the eventual nonlinear and nonstationary patterns exhibited, (2) considering imputation mechanisms and temporal models that can handle discrete-continuous outcomes to overcome potential boundedness/censoring issues, (3) establishing parsimonious models that can fairly capture the features derived from high-observation frequency data and can be estimated with reasonable computational resources, and (4) using DVINE copula models to allow asymmetric spatial dependence structures among the different variables.

There is a large literature on temporal modeling of wind power forecasting, but considerably less work combining spatial dependence into the forecasting framework. Through the careful consideration of the temporal modeling component, complemented by support vector regression of the temporal model residuals, this work demonstrates that a DVINE copula model most accurately represents the residual spatial dependence. Additionally, this work proposes a complete set of validation mechanisms for multi-h-step forecasts that, when considered together, comprehensively evaluate accuracy. The model and validation mechanisms are demonstrated in two case studies, totaling ten wind farms in the Texas electric grid. The proposed method outperforms baseline and competitive models, with an average Continuous Ranked Probability Score of less than 0.15 for individual farms, and an average Energy Score of less than 0.35 for multiple farms, over the 24-hour-ahead horizon. Results show the model’s ability to replicate the power output dynamics through calibrated and sharp predictive densities.