DLGA-PDE: Discovery of PDEs with incomplete candidate library via combination of deep learning and genetic algorithm

•DLGA-PDE is developed that integrates deep learning and genetic algorithm for discovering PDEs.•DLGA-PDE can discover PDEs with incomplete candidate library.•DLGA-PDE is robust to noisy and limited data.•DLGA-PDE can discover PDEs when the left-hand term of PDE has multiple options.•l0-regularizati...

Full description

Saved in:
Bibliographic Details
Published in:Journal of computational physics 2020-10, Vol.418, p.109584, Article 109584
Main Authors: Xu, Hao, Chang, Haibin, Zhang, Dongxiao
Format: Article
Language:English
Subjects:
Artificial neural networks
Burgers equation
Computational physics
Crossovers
Deep learning
Deep neural network
Genetic algorithm
Genetic algorithms
Incomplete candidate library
Libraries
Machine learning
Partial differential equations
PDE discovery
Wave equations
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!