Solar irradiance forecasting based on direct explainable neural network

[Display omitted] •Direct explainable neural network is proposed for solar irradiance forecasting.•Ridge activation function is designed to extract the nonlinear features of DXNN.•A new training process, including pretraining for error estimation, is developed.•The input-output relationship of DXNN...

Ausführliche Beschreibung

Gespeichert in:

Bibliographische Detailangaben
Veröffentlicht in:	Energy conversion and management 2020-12, Vol.226, p.113487, Article 113487
Hauptverfasser:	Wang, Huaizhi, Cai, Ren, Zhou, Bin, Aziz, Saddam, Qin, Bin, Voropai, Nikolai, Gan, Lingxiao, Barakhtenko, Evgeny
Format:	Artikel
Sprache:	eng
Schlagworte:	Algorithms Back propagation networks Data-driven model Economic forecasting Electric power Explainable neural network Feature extraction Forecasting Irradiance Machine learning Mapping Mathematical models Neural networks Parameter estimation Predictions Solar energy Solar irradiance forecasting Solar power Support vector machines
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!

Schreiben Sie den ersten Kommentar!