Intelligent parameters identification on numerical model of EOF-based gated injection in microfluidic channels

In this paper, a numerical simulation model of electroosmosis flow (EOF) based gated injection in microfluidic channels is established. Based on which, a control purpose to quantify and locate separated sample is to be realized, in which the key problem is to define the electric field, the injection time and the separation time. In order to give the proper control parameters, the artificial neural network (ANN) is adopted as an intelligent parameter identifier, in our design, it will give the injection time and the separation time properly if the expectation sample volume and the electric field are given. Tested by the numerical simulation model with 10 random calculation examples, the results indicate that the ANN identifier can give the corresponding control parameters correctly, and the control method on quantifying and locating the separated sample in gated injection can be successfully achieved.