Multi_R_Designer Tutorial

---

Two case problems are presented in this section. Various proceedings relative to the analysis of models and correlations via statistical tabulations and graphical representations (Parity plots) are exhibits through these examples. Most models/correlations expose applicability restrictions, which were defined either by the author or by the Multi_R_Designer team. Experimental conditions which do not meet the model or correlation restrictions would more likely to produce erroneous predictions. These values are referred as invalid points. On the other hand, conditions which fulfill the model/correlation specifications are referred as valid points. The models/correlations characteristics are represented on three level. First, the formation of parity plots, which compare the experimental values with the predicted values by one or several models/correlations, gives a general overview on their capability to genuinely predict the studied parameter. Complementary to the parity plots, CDF diagrams (Cumulative Distribution Function) present the repartition of absolute relative errors (ARE) between model/correlation predictions and experimental values. Finally, a complete tabulations of important statistical parameters are added to the analysis.

Multi_R_Designer have calculated and tabulated, for the whole integrated databank, the prediction results for all implemented models/correlations, only if key information were available in the databank. For example, data set #1 contains measurements on dynamic liquid holdup and frictional pressure drop exclusively meaning that the gas and liquid diffusion coefficients essential for mass transfer coefficient calculations were not included for this case in the integrated databank. Therefore, the correlations/models predicting mass transfer coefficients could not be applied for data set #1 since an important information was missing.

Here is a statement overview of the case problems elaborated in this section.

 
CASE PROBLEM VI: Analysis of 3 correlations predicting frictional pressure drop
CASE PROBLEM VII: Wall heat transfer coefficient predictions for beds filled with glass beads

Multi_R_Designer Tutorial