Statistical learning for wind power : a modeling and stability study towards forecasting

TL;DR

This paper demonstrates that machine learning models, especially CART-Bagging, outperform traditional parametric models in wind power prediction, and explores the impact of data quality and predictor selection on forecasting accuracy.

Contribution

It introduces the application of advanced machine learning algorithms to wind power modeling and proposes improvements in predictor selection methods for better forecasting.

Findings

CART-Bagging outperforms parametric models in stability and accuracy.

Deteriorated wind data significantly affect model performance.

Refined predictor selection enhances forecasting results.

Abstract

We focus on wind power modeling using machine learning techniques. We show on real data provided by the wind energy company Ma{\"i}a Eolis, that parametric models, even following closely the physical equation relating wind production to wind speed are outperformed by intelligent learning algorithms. In particular, the CART-Bagging algorithm gives very stable and promising results. Besides, as a step towards forecast, we quantify the impact of using deteriorated wind measures on the performances. We show also on this application that the default methodology to select a subset of predictors provided in the standard random forest package can be refined, especially when there exists among the predictors one variable which has a major impact.