Multi-Objective Optimization of HIV Virus Separation from the Blood Sample Using CFD and NSGA II Algorithm

In this study, the Multi-Objective Optimization (MOO) of HIV virus separation from blood sample in a Lab on a Chip (LOC) is investigated using Computational Fluid Dynamics (CFD) and NSGA algorithm. The separation device consists of two horizontal microchannel separated by a porous layer. The flow is controlled by an infinitesimal channel section connected to one of the microchannels. First, using CFD approach, the fluid flow is studied in over 150 separation devices with different geometrical parameters. All performance parameters like separation efficiency and pressure drop are calculated. Then, already computed numerical results are targeted by multi-objective genetic algorithm (NSGA II). In optimization process, eight different geometrical and process parameters are considered as optimization variables. Maximum separation efficiency and minimum pressure drop caused by separation are considered as two conflicting optimization objectives. Pareto front is presented to assist design of the efficient separators. Optimization yielded the optimum configuration of the geometrical and process parameters for the highest efficiency and lowest pressure drop in HIV virus separation process.