A Modified Peng-Robinson Cubic Equation of State Based on Bayesian Framework
Peng-Robinson cubic equations of state (PR-EoSs) as one of the most popular two-parameter cubic equations of state (2P-EoSs) are widely used to calculate thermodynamic properties of pure substances and their mixtures. However, the prediction accuracy of 2P-EoSs varies significantly among different substances due to its intrinsic limitation. To this end, many modifications have focused on changing the dependence structure of 𝛼 function with temperature for PR-EoS to enhance prediction accuracy. In this paper, we propose a Bayesian framework to calibrate a new 𝛼 function, which is a bias-corrected parametrized model form for the PR-EoS. The developed PR-EoS with the calibrated 𝛼function is applied to evaluate the thermodynamic properties of representative substances, including oxygen, carbon dioxide, and n-decane. Results show that the new developed PR EoS significantly improves the prediction accuracy of densities for the representative substances when compared to the original PR EoS.