Numerical and artificial neural network analysis of an axisymmetric co-flow-focusing microfluidic droplet generator using active and passive control

Droplet generation in microscale has gained enormous attention recent years especially the pharmaceutical industry due to their application targeted drug delivery into droplets. In most of these applications, monodispersity and uniformity droplets are essential. Microfluidic devices can generate at high throughput, enabling thousands droplet compound encapsulation per second. The is ensured hydrodynamically through dripping regime controlled by active passive microflow control methods. Here, we study numerically a microfluidic chip that uses non-embedded co-flow-focusing geometry, so throughput take advantage flow-focusing while co-flow geometry forecloses surfactant addition necessity. continuous dispersed phases were light mineral oil water, respectively. We investigated formation studied how changing external diameter affects transition between (which corresponds monodispersity) jetting regime. number parameters be taken account for optimization device enormous; therefore, order effect many geometrical hydrodynamical parameters, trained an artificial neural network based on our simulation data. Using this network, evaluated 3240 different cases. This approach resulted remarkable reduction computation time, from months seconds. Examining numerous cases such short period lets us choose optimum flow rate application. was able find best extend region map. Finally, harness frequency, also simulated periodically switched laser predict with same frequency as switching frequency. Therefore, altering controlling dimensions given ratio could achieved technique, even without satellite