Uncertainty in the prediction of the thermophysical behavior of new halogenated working fluids

Halogenated olefins can be an attractive alternative to working fluids that are under consideration for phase-out. In order to evaluate the potential of halogenated olefins as working fluids for organic Rankine cycle units with a realistic perspective it is necessary to predict both their primary thermophysical properties and their associated uncertainties. In this work we evaluated the uncertainty of two predictive models, based on the classical group contribution approach and on neural networks, for the critical temperature, critical pressure, acentric factor, and ideal gas heat capacity of halogenated substances. Based on this, guidelines for the most convenient model depending on the property and the fluid molecular structure were provided. The non-linear propagation of the uncertainty through the Peng-Robinson equation of state for the simulation of an organic Rankine cycle unit were also studied. The results suggest that the uncertainty of the predicted properties is highly dependent on the molecular structure for the classical model. The propagation of uncertainties in the simulation of an organic Rankine cycle unit also depends on the saturation properties of the fluid.