สำนักราชบัณฑิตยสภา

The Journal of the Royal Institute of Thailand Volume II - 2010 Sira Saisorna, Somchai Wongwises 77 Only methods proposed by McAdams et al. (1942) and Beattie and Whalley (1982) were found to be roughly predictable. The comparisons illustrate that the viscosity model developed by McAdams et al. (1942) gives the best prediction, with MAE of 46.9% and 67.1% of the predicted data falling within a ±40% error band. With MAE of 44%, the method of Beattie and Whalley (1982), the second best prediction, captures 29.4% of the data within a ±40% error band. As seen from the models shown in Eqs. (7) - (11), mass quality or volumetric quality is an important parameter in determining the two-phase viscosity. The models taking into account the mass quality tend to over-predict the data as illustrated in Figs. 3 and 4. As presented in Figs. 5 and 6, on the contrary, the models based on volumetric quality give under-prediction in comparison to the experimental results. Figure 2 Experimental data vs. predicted pressure drop with McAdams et al. model [11]. Fiure 3 Experimental data vs. predicted pressure drop with Lin et al. model [12]. Figure 4 Experimental data vs. predicted pressure drop with Cicchitti et al. model [13]. Figure 5 Experimental data vs. predicted pressure drop with Dukler et al. model [14].