A themodynamic model for analysis and selection of a working fluid for an organic rankine cycle with an open feeding fluid heater

The purpose of this study is to develop multiple linear regressions able to express the cycle thermal efficiency (CTE) from selected operating parameters for all organic fluid categories within several thermodynamic cycles. The present study also purposes the fluid selection considering the CTE results obtained through the developed model. The results provided by the statistical data allow to affirm the validity of the developed method, with high adjusted R-square values and low estimated deviation values. Wet fluids presented CTE average deviations equal to 0.38% for subcritical cycle and 0.45% for supercritical cycle. Dry fluids presented CTE average deviation of 0.83% in subcritical cycle and 0.61% for the supercritical cycle. Isentropic category presented 1.04% of CTE average deviation for the supercritical cycle and 0.68% for subcritical and superheated cycles. The calculation of the CTE absolute mean error provided values even smaller than the CTE estimated deviation provided earlier by the statistical analyze. Thus, a general absolute mean error equal to 0.6414% confirms the multiple linear regression method accuracy. The validation for the fluid selection method regards to a comparison between the best CTE sequence produced by the thermodynamic method and the best CTE sequence produced by the multiple linear regression method.