Session: 32-14 ORC & Supersonic Turbines 2

Paper Number: 127133

127133 - Numerical Analysis of a Two-Phase Turbine: a Comparative Study Between Barotropic and Mixture Models 

Unlocking the potential of low-temperature energy resources, including geothermal energy, solar power, and waste heat from heat engines and industrial processes holds the promise of significantly advancing our efforts to attain ambitious emissions reduction targets. Rather than solely focusing on incremental enhancements of conventional processes, the adoption of innovative power system technologies that are made possible through the use of two-phase turbomachinery can enable considerably higher performance gains. One of these technologies is the partial evaporation organic Rankine cycle, which enables enhanced utilization of sensible heat sources. Additionally, two-phase turbines have the potential to improve the performance of refrigeration and liquefaction systems. By replacing throttling valves with two-phase turbines, there is an opportunity to recover expansion work and increase the cooling capacity of the system, leading to higher performance.

Understanding the fluid dynamics in two-phase turbines is crucial for these applications, especially during the flashing process, where a rapid decrease in density occurs as liquid transitions to vapor due to a reduction in pressure. Computational fluid dynamics stands as an essential tool for analyzing the flow in two-phase turbines, as it can resolve the intricate flow features inherent in such systems. While a few previous computational fluid dynamics studies investigated the performance of two-phase turbines for these applications, a common limitation is that their models were either not validated under the relevant conditions or were validated exclusively with nozzle data, without comparison to experimental data for rotating machinery.

In view of the shortcomings in the existing literature, this paper aims to identify a suitable computational fluid dynamics modeling approach for two-phase turbo-expanders and validate the numerical predictions again experimental data from a two-phase turbine. Both two-dimensional (2-D) and three-dimensional (3-D) simulations are employed in this study. Initially, a 2-D axisymmetric analysis is conducted to examine the nozzle's characteristics. Subsequently, a 3-D simulation is employed to explore the entire stage in greater detail. Our evaluation is centered around two models: the mixture model and the barotropic model. The mixture model solves the Reynolds-Averaged Navier-Stokes equations for the two-phase mixture, treating it like a single-phase fluid with averaged properties, and incorporates an additional transport equation for the volume fraction of the dispersed phase. The barotropic model, on the other hand, assumes that the fluid follows a predetermined thermodynamic process—an isentropic expansion. Based on this assumption, the fluid properties of the mixture, including density and viscosity, are expressed solely as a function of pressure. However, while the barotropic model offers computational advantages, it comes with the shortcoming of being unable to account for flow segregation.

The validity and accuracy of these two modeling approaches is assessed using existing experimental data from a two-phase turbine – a single-stage impulse turbine featuring a single convergent-divergent nozzle followed by a rotor row with 151 blades. The working fluid, a mixture of liquid water and gaseous nitrogen at room temperature, expands from 2000 kPa to atmospheric pressure within the nozzle, generating a high-velocity jet that drives the rotor. A systematic comparison is made between the barotropic and mixture models by assessing key performance metrics of the nozzle and two-phase turbine, such as mass flow rate, torque, efficiency, pressure distribution, and phase separation phenomena across a spectrum of operating conditions and liquid-to-gas ratios.

Preliminary results suggest that the performance of the nozzle is strongly influenced by the slip between the two phases, which increases with the increase of nitrogen gas in the nozzle. Both models overpredict the exit velocity, and this deviation increases with the increase in the amount of nitrogen. The overprediction of the nozzle exit velocity eventually results in an overestimation of the torque generated by the turbine. The insights gained from this research provide a solid basis for the design, operation, and optimization of two-phase turbines.

 

Presenting Author: Amit Kumar Technical University of Denmark

Presenting Author Biography: Dr. Amit Kumar is a Post-Doctoral fellow in the Department of Mechanical Engineering at the Technical University of Denmark (DTU). Previously, he was associated with the Turbomachinery Research Laboratory at the Indian Institute of Technology, Bombay, as a Post-Doctoral Fellow. He completed his Ph.D. and M.Tech. from the same institute.
He is passionate about the research in turbomachinery, and his area of expertise lies in the experimental, analytical, and numerical techniques in turbomachinery.

Authors:

Amit Kumar Technical University of Denmark
Simone Parisi Technical University of Denmark
Roberto Agromayor Technical University of Denmark
Jens Honore Walther Technical University of Denmark
Fredrik Haglind Technical University of Denmark