Determination of the Optimum Stability Conditions in Al2O3 Nanofluids with Artificial Neural Networks

Abstract

In this study, the optimum stability conditions in Al<inf>2</inf>O<inf>3</inf> nanofluids were determined by utilizing artificial neural networks (ANN). First of all, nanofluids used in the experimental study were prepared by synthesizing Al<inf>2</inf>O<inf>3</inf> nanoparticles and mobile brand oil as a base fluid, which is used as a heat transfer fluid in the industry. To ensure stability, the nanoparticles were synthesized in the oil by adding the specified acid and base solutions. The sedimentation method was applied to measure the stability after ultrasonic mixing stage of nanofluids which determined as Al<inf>2</inf>O<inf>3</inf> nanoparticles 1 %, 2 %, and 3 % by mass. Periodic sedimentation measurements were continued for 36 h. Optimum conditions were obtained using the successful models. Experiments were repeated for optimum conditions, and the consistency of the model and agreement with the experimental system were observed. According to the findings, the highest improvement rates in the sedimentation values of the optimum acid–base ratios obtained by modeling with ANN were 11.2 %, 32.6 %, and 34 % for acid simulations and 55.2 %, 47.3 %, and 49.2 % for base simulations, respectively. Besides, the experimental results have been successfully overlapped with a detailed simulation pattern. © 2020, Springer Science+Business Media, LLC, part of Springer Nature.