Gain scheduling PID control for directed self-assembly of colloidal particles in microfluidic devices

Directed self-assembly provides a promising route to fabricate small-scale structures for various engineering applications. The use of the external fields of variable strength and frequency allows for active controls to be implemented. For assembling specific structural features, precise control over a local particle density is often needed, which can be achieved using feedback control. However, the strong nonlinear behavior of directed self-assembly complicates such control. In this work, a gain-scheduling feedback control strategy for directed self-assembly of colloidal particles is presented. The process gain is described by an empirical steady-state model, which is used for scheduling a proportional-integral feedback controller. The automated controller is implemented experimentally in a microfluidic device for directed self-assembly of colloidal particles. The gain-scheduled controller shows a significantly improved dynamic performance compared with a conventional proportional-integral controller over a broad range of operating parameters.