Analysis of thermodynamic models for simulation and optimisation of organic Rankine cycles

Abstract

Equations of state (EOSs) form the base of every thermodynamic model used in the design of industrial processes, but little work has been done to evaluate these in the context of such models. This work evaluates 13 EOSs for their accuracy, computational time and robustness when used in an in-house optimisation program that finds the maximum power output of an organic Rankine cycle. The EOSs represent popular choices in the industry, such as the simple cubic EOSs, and more complex EOSs such as the ones based on corresponding state principles (CSP). These results were compared with results from using the Groupe Européen de Recherches Gazières (GERG) EOS, whose error is within experimental uncertainty. It appears that the corresponding state EOSs find a solution to the optimisation problem notably faster than GERG without significant loss of accuracy. A corresponding state method which used the Peng–Robinson EOS to calculate the shape factors and a highly accurate EOS for propane as the reference EOS, was shown to have a total deviation of just 0.6% as compared to GERG while also being 10 times as fast. The CSP implementation was also more robust, being able to converge successfully more often.