Design and Optimization of Turbomachinery for Thermodynamic Cycles Utilizing Low-Temperature Heat Sources

Organic Rankine cycle power systems and heat pump systems are considered as attractive tech-nologies for the exploitation of excess heat into electricity and useful heat. However, the major barriers to the development of mass-produced low-temperature thermodynamic cycles are cost and efficiency. Expanders and compressors for organic Rankine cycle power systems and heat pumps are key components regarding cost and performance, and the use of reliable and accurate numerical models for these machines is currently regarded as one of the major challenges in the field.

This Ph.D. thesis investigates three turbomachinery technologies: axial and radial turbines for organic Rankine cycle power systems, and centrifugal compressors for heat pump systems. This work has two main objectives. The first one is to develop suitable mean-line models for the preliminary design and performance prediction of such machines.

The second objective is to develop and apply design optimization methodologies which allow combining the machine and the cycle designs to identify suitable solutions from the performance and technical viewpoints. First, a mean-line model for the preliminary design of axial-flow turbines is developed for single and multistage configurations operating with organic fluids.

The model is validated against single-stage and multistage turbine test cases, two of which employ organic working fluids. A methodology for the combined design optimization of both the turbine and the organic Rankine cycle system is developed and applied to two case studies representative of key applications in the Danish sector.

The results suggest that the methodology allows considering the trade-offs regarding cycle and turbine criteria, achieving more accurate results compared to the design of the system alone, and avoiding infeasible solutions in the preliminary design step. Second, a mean-line model for the preliminary design and performance of single-stage radial-inflow turbines operating at high-pressure ratio conditions is developed.

The model is calibrated with the data of six turbines operating at high-pressure ratio conditions from the literature and is validated with two additional turbines test cases, one of which employs an organic fluid. The calibration of the turbine, performed with an optimization process, allows a significant reduction in the deviation with the experimental data for both the isentropic efficiency and the mass flow rate.

Finally, a mean-line model for the preliminary design and performance estimation of centrifugal compressors is developed. The model is validated with five test cases from the open literature employing three different working fluids and is coupled to that of a heat pump system. The combined model is then optimized to design of a high-temperature heat pump system.

The combined optimization methodology allows achieving more accurate results and identifying suitable solutions regarding preliminary design and performance.