Machine Learning for Fluid Mechanics
This review chapter shows how supervised, semi-supervised, and unsupervised learning can be leveraged for fluid mechanics. The review discusses Neural Networks, RNNs, LSTMs, GANs, reinforcement learning, dimensionality reduction, clustering, and more for flow modeling and flow control applications. There is also a discussion of flow optimization in the chapter that covers using stochastic algorithms. The chapter presents a historical context to the information alongside a thorough presentation of the state-of-the-art, and comments on future issues in the field.
Type:
Scientific Paper
Area:
Machine Learning
Target Group:
Advanced
DOI:
10.1146/annurev-fluid-010719-060214
Author of the review:
Jake Walker
TU Delft
You have to login to leave a comment. If you are not registered click here