Evaluation of Thermal Conductivity of Gases at Atmospheric Pressure through a Corresponding States Method

In the present communication, we propose a corresponding
states
method for calculation/estimation of the vapor thermal conductivity
of chemical compounds (mostly organic), applying the gene expression
programming (GEP) algorithm. Around 16000 thermal conductivity data
of gases at various temperatures from 100 to 1500 K and atmospheric
pressure related to about 1600 chemical compounds (mostly organic)
are used for development and validation of the method. The quantities
used in the corresponding states correlation include temperature,
critical pressure, molecular weight, acentric factor, and normal boiling
point. More than 14000 thermal conductivity data are randomly selected
for developing (training + optimizing) the correlation, and about
1600 data are used for checking its prediction capability. The obtained
statistical parameters including average absolute relative deviation
of the results from the applied data (about 8%) show acceptable accuracy
of the presented method. The most important features of the developed
model are its simplicity, its wide range of applicability, and its
validity based on the Leverage value test.