作者: Hashem Nowruzi , Hassan Ghassemi , Mahdi Yousefifard
DOI: 10.1016/J.PADIFF.2020.100004
关键词:
摘要: Abstract Curved ducts with non-circular cross-sectional geometry have significant applications in different industries. Hydrodynamic stability these curved is an interesting issue field of fluid mechanics. In the present study, linear hydrodynamics flow rectangular duct semi-analytically investigated. Then, hydrodynamic estimated via using artificial neural networks (ANNs). To this accomplishment, critical Dean number (Dn c ) under various aspect ratios and curvature ratios. Based on semi-analytical results, Dn increased by ratio enhancement. addition, irregular variation trend found enhancement ratio. Moreover, maxima mean square error minima correlation coefficient for intended ANN are obtained 0.00144 0.98621, respectively. Finally, predictive equation suggested to estimate weights bias designed ANN.