Iterative Learning Control of Two-Phase Laminar Flow Interface in Y-Shaped Microfluidic Channel

In Y-shaped microfluidic chip, the laminar flow refers to a phenomenon that two fluids introduced from two inlets flow side by side without turbulence and form two stable streams in outlet with a common interface. The interface position of laminar flow has significant influence in some experiment analysis of life science, such as molecule diffusion and solvent extraction, where there are still a series of problems associated with the manipulation of interface position. In this brief, an iterative learning control (ILC) scheme is proposed for precise control of the laminar flow position. ILC can improve the current input signal iteratively based on the experimental results achieved in the previous trials and eventually produce the desired interface position in output channel. To verify the effectiveness of the proposed ILC scheme, we design and fabricate the Y-shaped microfluidic chips. Furthermore, two different scenarios are considered, where the results show that an appropriate input signal achieving the desired output can be promptly obtained via ILC. The three main advantages of the proposed control scheme lie in: 1) the simple structure and the feedforward characteristic of the control scheme make it implementable in an easy way; 2) it is a partially model-free method, and hence, no accurate model of laminar flow is required and system uncertainties can be dealt with rigorously when designing the controller; and 3) compared with the well-adopted traversal methods in life science research, the idea of ILC reduces the number of experimental trials remarkably.