Abstract
Background: Achieving a nanofluid with optimal thermal conductivity and viscosity is one of the main problems of applications of nanofluids in industries.
Methods: There are experimental and theoretical methods to reach an applicable nanofluids with mentioned characteristics. Surely, experimental methods are not optimal in time and cost($) aspects. So, in the present study multi-objective optimization of nanofluids ND-Co3O4 is done to find the optimal solid volume fraction for having maximum thermal conductivity and minimum viscosity. The response surface methodology (RSM) is used to model target functions using empirical data. The improved non- dominated sorting method and multi-objective particle swarm optimization are used as powerful tools for optimization. In order to implement the optimization process, the obtained target function model is joined to multi-objective particle swarm algorithm and it is used in each step of the target function evaluation.
Results: The obtained results of these two algorithms are presented in the form of Pareto front. Also, a comparison between them is provided. According to the optimal results, MOPSO has a better performance that the other one.
Conclusion: It will be shown that the highest thermal conductivity and the lowest viscosity occur at the maximum temperature. By investigating obtained optimum results, the optimal point with highest thermal conductivity and lowest viscosity was found at about 60 °C and 0.1 to 0.11 of solid volume fraction.
Keywords: Non-dominated sorting optimization, response surface method, nanofluids, particle swarm optimization, thermal conductivity, pareto optimal design.
Graphical Abstract