Wind Energy Production in Italy: A Forecasting Approach Based on Fractional Brownian Motion and Generative Adversarial Networks

被引：0

作者：

Di Persio, Luca ^{[1
]}

Fraccarolo, Nicola ^{[2
]}

Veronese, Andrea ^{[1
]}

机构：

[1] Univ Verona, Dept Comp Sci, I-37134 Verona, Italy

[2] Univ Trento, Dept Math, I-38123 Trento, Italy

来源：

MATHEMATICS | 2024年 / 12卷 / 13期

关键词：

energy forecasting; generative adversarial networks; machine learning; renewable energies; stochastic differential equations; OPTIMIZATION ALGORITHM; COMBINATION SYSTEM; SPEED; MODEL; PREDICTION; ENSEMBLE;

D O I：

10.3390/math12132105

中图分类号：

O1 [数学];

学科分类号：

0701 ; 070101 ;

摘要：

This paper focuses on developing a predictive model for wind energy production in Italy, aligning with the ambitious goals of the European Green Deal. In particular, by utilising real data from the SUD (South) Italian electricity zone over seven years, the model employs stochastic differential equations driven by (fractional) Brownian motion-based dynamic and generative adversarial networks to forecast wind energy production up to one week ahead accurately. Numerical simulations demonstrate the model's effectiveness in capturing the complexities of wind energy prediction.

引用

页数：16

共 56 条

[1] A generative model for fBm with deep ReLU neural networks [J].