In the research for this paper, a GA–PNN hybrid system was used for modelling the convective heat transfer
characteristics and pressure drop of TiO2–water a nanofluid in a fully developed turbulent flow based
on an experimentally obtained train and test data set. Models were developed for the Nusselt number
and the pressure drop of the nanofluid as a function of Reynolds and Prandtl numbers, nanofluid volume
concentration and average nanoparticle diameter. The results of the proposed models were compared
with experimental data and with existing correlations. The validity of the proposed models was benchmarked
by using statistical criteria and NSGA-II was used for multi-objective optimisation for the convective
heat transfer. In the optimisation procedure model, the Nusselt number and pressure drop were
considered as the objective functions. However, when the set of decision variables was selected based
on the Pareto set, it ensures the best possible combination of objectives. The Pareto front of multi-objective
optimisation of the Nusselt number and pressure drop proposed models were also shown and discussed.
It was found that application of the multi-objective optimisation method for the turbulent
convective heat transfer characteristics and pressure drop of TiO2–water nanofluid could lead to finding
the best design points based on the importance of the objective function in the design procedure.