Zhou, YuanSalahshour, SoheılDerakhshanfard, Amir HosseinSajadi, S. MohammadJasim, Dheyaa J.Nasajpour-Esfahani, NavidSalahshour, SoheilEftekhari, S. Ali2024-05-252024-05-25202312352-492810.1016/j.mtcomm.2023.1076122-s2.0-85177617658https://doi.org/10.1016/j.mtcomm.2023.107612https://hdl.handle.net/20.500.14517/1306Eftekhari, SeyedAli/0000-0002-9730-4232; Jasim, Dheyaa Jumaah/0000-0001-7259-3392; toghraie, davood/0000-0003-3344-8920In this study, the thermal conductivity (knf) of Silicon Oxide-MWCNT-Alumina/Water hybrid nanofluid (HNF) is predicted versus solid volume fraction (SVF) and temperature. For this reason, various combinations of SVF and temperature are considered from SVF= 0.1-0.5% and 20-60 (degrees C) respectively. Then, an adaptive neuro-fuzzy inference system (ANFIS) has been effectively used to model the knf of HNF as one of the effective machine learning techniques. Various shapes of membership functions are considered and the generalized bell shape membership function showed to have acceptable accuracy for knf prediction using an ANFIS-based model. Moreover, the outcomes reveal that the effect of SVF is higher than temperature influence on the knf of HNF. Specifically, when the SVF is increased from 0.1% to 0.5%, there is an approximate 25% increase in knf. Conversely, an increase in temperature leads to a smaller ratio of knf increment. When the temperature rises from 20 degrees to 60 degrees C, knf only increases by less than 10%. The highest error value is found at phi = 0.2% and T = 60 degrees C, amounting to 0.01128 W/mK.eninfo:eu-repo/semantics/closedAccessAdaptive neuro-fuzzyThermal conductivitySilicon oxideMWCNT -alumina-waterHybrid nanofluidArticleQ2Q237WOS:001123391000001