Here we present a highly customisable image-based fuzzy logic control (FLC) method for pressure-driven droplet microfluidics. The system is designed to position droplets of different sizes in microfluidic chips of varying channel size in the centre of the region of interest (ROI) using two parallel multiple input single output (MISO) FLCs. Overall, 95.1% of the droplets with an average displacement of 2.5 μm could be kept within the ROI during the pre-defined time intervals of up to 10 s. This is achieved by pre-determined pressure values that are kept constant during this time. The control principle was tested on different pressure controllers and microfluidic chips varying in material, channel layout and cross section. Droplet volumes ranged from a few hundred picolitres to a tenth of a microlitre. The droplets were composed of deionised water or contained two concentrations of S. cerevisiae. The average processing time was 12.5 seconds. This makes the method suitable for studying several hundred pre-sorted droplets from high-throughput screening (HTS) experiments.

中文翻译：

---

基于图像的模糊逻辑控制，用于压力驱动的液滴微流体，作为多模态成像显微镜的自动进样器。


在这里，我们提出了一种高度可定制的基于图像的模糊逻辑控制 （FLC） 方法，用于压力驱动的液滴微流体。该系统旨在使用两个并联多输入单输出 （MISO） FLC 将不同通道大小的微流体芯片中不同大小的液滴定位在感兴趣区域 （ROI） 的中心。总体而言，在长达 10 秒的预定义时间间隔内，平均位移为 2.5 μm 的 95.1% 的液滴可以保持在 ROI 内。这是通过在此期间保持恒定的预定压力值来实现的。控制原理在材料、通道布局和横截面不同的不同压力控制器和微流体芯片上进行了测试。液滴体积从几百皮升到十分之一微升不等。液滴由去离子水组成或含有两种浓度的酿酒酵母。平均处理时间为 12.5 秒。这使得该方法适用于研究来自高通量筛选 （HTS） 实验的数百个预分选液滴。