Integrated design of ORC power plants

Abstract

In an effort to reduce CO2 emissions, the energy sector is undergoing a profound transformation, targeting decarbonization and more efficient energy use. In this context, power plants based on the Organic Rankine Cycle (ORC) concept have the potential to play a key role, both as primary power production systems and for the recovery and conversion of otherwise unused thermal energy.

ORC power plants are diverse and complex, and are characterized by the strong interdependency between the thermodynamic cycle characteristics, equipment design, and properties of the working fluid. For base-load power production applications, where the return on investment is closely correlated with power plant efficiency, it can be beneficial to adopt advanced thermodynamic cycle configurations, such as those based on the use of a zeotropic mixture as working fluid or a supercritical pressure in the heating process. These configurations aim to mitigate the heat transfer irreversibilities that arise because the working fluid undergoes phase change at constant temperature. The benefit that these configurations have on power plant efficiency is thus proven. However, this efficiency improvement comes at a cost: as the temperature differences driving the heat transfer process are reduced, the required heat transfer area-and thus, the cost of the component- increases. Consequently, a complex tradeoff exists between plant power output and plant capital expenditure. It therefore remains uncertain whether adopting such configurations, particularly the use of mixtures as working fluids, is also advantageous from an economic standpoint.

The objective of the research documented in this dissertation is to address this question. The main contributions include detailed preliminary design models of heat exchangers and turbomachinery, and the development of a new integrated ORC design framework, called WoPycle. This framework enables the simultaneous optimization of thermodynamic cycle parameters, preliminary component design, and molecular structure of the working fluid. The optimization of the molecular structure is achieved by relying on the PCP-SAFT equation of state model, equipped with group contribution methods, which enable the prediction of the fluid properties solely from the molecular groups that compose the molecule. These groups are thus treated as optimization variables, thereby allowing the identification of the optimal molecular structure of the working fluid or mixture components, as well as the discovery of novel compounds, which have not yet been considered so far due to the lack of adequate models.

WoPycle was used to perform the integrated design and optimization of ORC power plants that convert thermal energy from a low-temperature heat source into electricity, considering both multicomponent working fluids and supercritical cycle configurations. Results show that, due to the high cost of the air-cooled condenser and the distribution of exergy losses among the main components of the plant, the thermodynamic benefit of using a zeotropic mixture—which leads to a 6–7% increase in net power output—does not sufficiently reduce the Levelized Cost of Electricity (LCOE) of the plant. Therefore, for this application, a zeotropic mixture is not recommended as the working fluid, as the additional practical challenges, specifically controlling the concentration of the components during plant operation, far outweigh the advantages. On the contrary, adopting a supercritical cycle configuration appears more promising, as heat transfer irreversibilities are reduced in the Primary Heat Exchanger (PHE), which is a comparatively more affordable component. In principle, an LCOE reduction of 6-7% is achievable, provided that the working fluid flows on the shell side of the PHE, as this minimizes the cost of the component. Nevertheless, supercritical power plant operation with this arrangement is not documented in the literature. Thus, the practical feasibility remains uncertain.