Simulating ORC expanders: CFD investigation and lumped parameter model identification

The decarbonization of the energy system is a critical step toward mitigating climate change and achieving sustainable development. Among the various strategies under investigation, waste heat recovery through Organic Rankine Cycle (ORC) systems offers a promising opportunity to reduce carbon emissions by converting low-grade thermal energy into electricity. However, the inherently variable nature of waste heat sources demands ORC systems that can maintain high efficiency across a wide range of operating conditions. This paper presents a hybrid modeling approach for scroll expanders, a key component in small-scale ORC systems. A Computational Fluid Dynamics (CFD) model is first developed to simulate the performance of the expander under various boundary conditions. The results from this detailed model are then used to identify a reduced-order lumped parameter (LP) model, which is computationally efficient and suitable for integration into dynamic ORC simulations. The identification procedure combines direct geometric extraction with a least-squares minimization technique to calibrate key parameters such as leakage areas and mechanical losses. The LP model demonstrates good agreement with the CFD results in terms of shaft power and overall performance trends, despite some discrepancies in mass flow rate and discharge temperature. These differences are analyzed and attributed to modelling assumptions and simplifications. Nevertheless, the LP model proves effective for generating operating maps and guiding the redesign of the expander for a wider range of operating conditions. This method enables the development of more flexible and efficient ORC systems, supporting the broader goal of sustainable energy conversion from variable waste heat sources.