Neural network-enabled, all-electronic control of non-Newtonian fluid flow

Area of Science:

• Fluid Dynamics
• Microfluidics
• Control Systems Engineering

Background:

• Accurate control of non-Newtonian fluid flow in microchannels is vital for advanced manufacturing and healthcare.
• Current methods often lack the necessary sensitivity and real-time responsiveness for effective flow adjustment.
• Challenges include precise measurement and dynamic control of complex fluid behaviors at the microscale.

Purpose of the Study:

• To develop a highly sensitive, real-time, all-electronic system for controlling non-Newtonian fluid flow in microscale channels.
• To address limitations in existing technologies regarding measurement accuracy and control responsiveness.
• To provide a versatile and integrated solution for microfluidic flow management.

Main Methods:

• Integration of a contactless, cuff-like flow sensor with a neural-network based control algorithm.
• Development of a closed-loop system for dynamic adjustment of fluid flow waveforms.
• Utilized microscale outlets for precise fluid delivery (0.76 μl/min).

Main Results:

• Demonstrated closed-loop, real-time control of various non-Newtonian fluid flow waveforms.
• Achieved high sensitivity in flow measurement and control.
• The system proved effective for microscale fluid manipulation.

Conclusions:

• The developed all-electronic system offers a simple, miniaturized, and high-performance solution for non-Newtonian fluid flow control.
• This technology is readily integrable into existing microfluidic setups.
• The system advances capabilities in microscale fluid management for diverse applications.