Computational design for micro fluidic devices using lattice boltzmann and heuristic optimisation algorithms

The study on micro devices is gaining importance in various fields from biological to engineering. The dimensions of these devices range from millimetres (mm) to micrometres (μm) and they work within the laminar flow regime due to their low Reynolds number. Although diffusivity dictates the mixing in such conditions, this work is based on the simulation of two non reacting iso-thermal and incompressible fluids for both streams, so the mixing is governed only by turbulence. A numerical study using the Lattice Boltzmann method (LBM) is carried out in order to examine the mixing in one configuration. In the second part of the work an interface is developed between the LBM code and multi objective optimisation software in order to investigate new configurations which enhance the mixing. The objectives are to maximise the vorticity and minimise the pressure drop, which are conflicting between themselves. The tool that integrates the optimiser and the LBM code for simulating microreactor, can run on a multi level parallelisation, hence the time required for the whole simulation has been drastically reduced. A preliminary optimisation is performed on a microreactor with multi holed baffle plate. Despite small number of iterations, three totally different configurations have been found: greatest vorticity, smallest pressure drop and a compromise. The first two configurations satisfy the expectations, whereas the compromise solution presents an innovative configuration. In fact, it has a low pressure drop and a high vorticity, which is achievable by having high Reynolds number and big diameters of the holes. Hence, this tool proves to be robust and efficient in the multi objective optimisation of microreactor.