Organic Rankine Cycle (ORC) architectures
Steven Lecompte, Henk Huisseune, prof. Michel De Paepe.
This report is composed in the frame of the IWT SBO-110006 project The Next Generation Organic Rankine Cycles (www.orcnext.be), funded by the Institute for the Promotion and Innovation by Science and Technology in Flanders (IWT).
The presented work is part of WP1 ‘Cycle architectures, design methods and optimization’. In particular a literature survey is made in agreement with subtask 1.1 ‘Study of architectures’. In this report the research on Organic Rankine Cycle (ORC) architectures is discussed solely based on thermodynamic considerations. Thermo-economic factors are out of the scope of this report.
The goal of this report is to communicate recent progress in ORC architectures towards the research partners and advisory board of the ORCNext project. As such, this work should not be considered a scientific article.
The Organic Rankine Cycle (ORC) has a lot of potential to convert waste heat into electricity thanks to the selection of an appropriate working fluid. Furthermore, ORC’s are designed for unmanned operation with little maintenance. Because of these excellent characteristics several ORC waste heat recovery plants are already in operation. Although the basic ORC is gradually adopted into industry, the demand of an increased performance appears. Therefore, the next logical step is the development of new ORC architectures. In the last decade there has been a strong renaissance towards ORC research. However, relative little attention has been paid to ORC cycle architectures. A summary of the literature review is given in Appendix A together with the used boundary conditions. It is crucial to notice that because of the various boundary conditions used no concise comparison is possible.
In waste heat applications the optimization parameter is the net power output. The net power output (Wnet) is the product of the heat input (Qin) and the thermal efficiency (η_th). Therefore both Qin and ηth must be optimized in order to maximize the power output. As a result two general types of cycle modifications are proposed, those which:
Increase the thermal efficiency of the cycle (ηth):
- Regenerative cycles: closed heater (recuperator) or open heater (regenerator);
Increase the heat addition (Qin) to the cycle:
- Supercritical cycles (SC)
- Trilateral cycles (TLC)
- Zeotropic mixtures (ZM)
- Multiple pressure levels
- Organic Flash Cycle (OFC)
Still, we see that the modifications proposed will in general have an effect on both (Qin) and (ηth).
The regenerative cycles recuperates heat after the turbine in order to increase the thermal efficiency while the heat input decreases. As a result the net power output is lower or equal than that of the basic ORC. Furthermore, the complexity of the cycle increases.
When reheating, the working fluid leaving the first expander is again heated and subsequently fed to a second expander. Only a small benefit in thermal efficiency is reported while the complexity of the cycle increases.
In a supercritical cycle the working fluid is pressurized and heated to a supercritical state. No evaporation takes place. The thermal efficiency is lower than that of the basic ORC in favour of an increased heat addition to the cycle. The net result is an increase in power output over the basic ORC. Yet, the cycle layout stays the same although the components need to be designed for higher pressures. The knowledge about the stability of the working fluids and the heat exchange and pressure drop correlations is limited.
In the trilateral cycle evaporation of the working fluid is omitted in order to better match the cooling process of the waste heat carrier. This results in an increased heat input at the expense of a lower thermal efficiency. Nevertheless, the net power output is increased compared to the basic cycle. The main challenge lies in developing two-phase expanders with a high isentropic efficiency.
For the Organic Flash Cycle (OFC), the working fluid is again not evaporated. In contrast to the TLC, the working fluid is first flashed in a flash tank and only the vapour is directed to the expander. The heat input is increased but also large throttling loses emerge. Therefore, compared to the basic cycle, there is no performance benefit. However, by extending the cycle with multiple flash tanks or multiple turbines a gain in power output is possible. The complexity of the system is however a challenge.
The basic ORC with zeotropic mixtures as working fluid has the ability to increase both the heat addition in the evaporator and the heat rejection in the condenser. Again there is an increased net power output compared to the basic ORC. However there is the risk of fluid fraction. Furthermore, there is limited knowledge about heat exchange and pressure drop correlations.
In the ORC with multiple pressure levels the heat addition process is divided into several pressure levels. This effectively reduces the losses associated to finite temperature heat transfer and as a result increases the heat input to the cycle. Therefore the power output is higher than that of the basic ORC. The complexity of the system is significantly increased with several expanders, pumps and heat exchangers.
It is clear from this work that there are possibilities to increase the performance of the simple ORC. Especially the supercritical cycle, trilateral cycle and the cycle with multiple pressure levels are promising. However the current research is insufficient to make a fair thermodynamic comparison between the cycle modifications proposed.