Related Experiment Video
Updated: Jun 12, 2026

08:48
Demonstration of Spin-Multiplexed and Direction-Multiplexed All-Dielectric Visible Metaholograms
Published on: September 25, 2020
Programmable Hydrodynamic Invisibility Enabled by Machine-Learning-Guided Metamaterials
Lili Zhang1, Yiyang Zhang1, Jinrong Liu2
1Department of Physics, State Key Laboratory of Surface Physics, and Key Laboratory of Micro and Nano Photonic Structures (Ministry of Education), Fudan University, Shanghai, China.
Advanced Materials (Deerfield Beach, Fla.)
|June 11, 2026
Summary
Researchers developed programmable hydrodynamic metamaterials for fluid transport control in porous media. This machine-learning-guided approach achieves robust hydrodynamic invisibility, adapting to varying background conditions for advanced applications.
Area of Science:
- Fluid Dynamics
- Materials Science
- Metamaterials
Background:
- Fluid transport in porous media is crucial for natural processes and technology.
- Hydrodynamic invisibility cloaks manipulate flow without external disturbance but are typically static.
- Existing devices fail with variable background permeability.
Purpose of the Study:
- To develop a programmable hydrodynamic invisibility system adaptable to varying background permeabilities.
- To demonstrate a machine-learning-guided metamaterial strategy for robust flow manipulation.
- To enable hydrodynamic camouflage by matching external flow fields.
Main Methods:
- Utilized a machine-learning-guided inverse-design framework.
- Developed a tunable-permeability metamaterial shell for cloaking.
- Employed experimental validation across diverse background permeabilities.
Main Results:
- Achieved high-fidelity hydrodynamic cloaking across high, medium, and low background permeabilities.
- Demonstrated programmable control over fluid transport.
- Showcased the ability to match prescribed exterior reference flows for camouflage.
Conclusions:
- Programmable hydrodynamic metamaterials offer a scalable, general strategy for robust, on-demand fluid transport manipulation.
- This approach overcomes limitations of static invisibility devices.
- Potential applications span separation science, microfluidics, and biomechanics.

