Back propagation modeling of shear stress and viscosity of aqueous Ionic-MXene nanofluids

Back-propagation modeling of viscosity and shear stress Ionic-MXene nanofluid is carried out in this work. The data for 0.05, 0.1, 0.2 mass concentration (mass%) are collected from the experimental analysis. Shear as a function rate mass% MXene nanoparticles used input. Additionally, temperature % separately. Based on possible combinations, five back-propagation algorithms developed. In each algorithm, models depending upon number neurons hidden layer used. training testing all algorithm performed. Statistical analysis network output done to evaluate accuracy by finding losses terms mean squared error (MAE), root-mean-squared error, absolute deviation. Model 1 found have lower than remaining its only one. performance evaluation metrices model show that involved acceptable. satisfactory be comfortably good agreement with desired output.