Characterization of discrete fracture networks with deep-learning based hydrogeophysical inversion

•A deep-learning based parameterization method is utilized to map high-dimensional fracture parameters space into low-dimensional latent variables.•A joint inversion framework (CVAE-ESMDA) is developed to characterize the fracture networks.•Multiple data types, including pressure and self-potential...

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	Journal of hydrology (Amsterdam) Jg. 631; S. 130819
Hauptverfasser:	Deng, Yaping, Kang, Xueyuan, Ma, Haichun, Qian, Jiazhong, Ma, Lei, Luo, Qiankun
Format:	Journal Article
Sprache:	Englisch
Veröffentlicht:	Elsevier B.V 01.03.2024
Schlagworte:	Convolutional variational autoencoder Data assimilation data collection Deep learning Fracture network groundwater flow Hydrogeophysical inversion solutes uncertainty Deep learning Hydrogeophysical inversion Fracture network Convolutional variational autoencoder Data assimilation
ISSN:	0022-1694
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!