A study on the ORC for OTEC applications

A study on the ORC for OTEC applications: Performance analysis for a changed configuration

Stelwagen, Johan (TU Delft Mechanical, Maritime and Materials Engineering)

Infante Ferreira, Carlos (mentor)

Delft University of Technology

Mechanical Engineering

2019-07-03

This thesis project aims to enable Bluerise to create a 3 [MW] plant for their OTEC system. The current model and lab setup used to create insight into the systems working mechanisms are based on a Kalina cycle configuration. This method uses a working fluid mixture of water and ammonia. After evaporation, the liquid and vapor are separated and the heated liquid is used in a recuperator to retain (some of) the heat and increase the efficiency of the system. The first 3 [MW] test plant that is planned to be built will be using an ORC configuration. This method solely uses ammonia as a working fluid and doesn’t use a recuperator to retain heat. Understanding the effects of liquid separation and re-circulation are important aspects while using this configuration. The working method of the current OTEC off-design model and lab setup will have to be altered to accommodate the ORC configuration. An extensive study on both available literature, the current OTEC off-design model and OTEC Demo lab setup lead to conclude that a gear pump will be implemented to drive the liquid re-circulation from the separator back to the evaporator. Parallel to the gear pump, a one-way valve is installed into the cycle so natural re-circulation experiments can also be conducted. From literature, a hypothesis is made on what the effect on the evaporator heat transfer rate could be by changing the re-circulation rate. The re-circulation rate must not be too high, because increased amounts of vapor bubbles increase fluid mixing and thus heat transfer. The re-circulation rate also must not be too low, because dry-out in the evaporator will occur, reducing the heat transfer rate. The OTEC off-design model that currently exists at Bluerise B.V. is used and expanded upon to accommodate the ORC configuration. The evaporator is changed in more detail, adding a heat transfer correlation and the possibility to calculate the evaporator pressure drop through two phase pressure drop correlations. After implementing the gear pump liquid re-circulation technique, the OTEC Demo can be used to create experimental data. From experiments, it is found that the re-circulation rate does not significantly change the evaporator heat transfer rate between 1.2 to 2.9 re-circulation rate. This is a remarkable result but can be explained by the low flow velocities in the evaporator, which indicate that the heat transfer process in the evaporator is mostly driven by pool boiling heat transfer mechanisms over flow boiling heat transfer mechanisms. Knowing that the re-circulation rate does not affect the evaporator heat transfer rate, liquid re-circulation can also happen naturally, by the liquid column driving force in the separator. Using this technique, a comparison with the Kalina cycle configuration using pure ammonia is made. The evaporator performance is higher in the Kalina cycle configuration, but the ORC configuration net power output is slightly higher. The main reason for this phenomenon is a lower required pumping power for the ORC configuration. The OTEC off-design model in ORC configuration is validated to the experimental data collected. The two phase heat transfer correlations used show to be very mass flux dependent, and applicable to flow boiling evaporative heat transfer. The correlations proposed by Taboas and Han et al. have the best fit to the experimental data. The correlation proposed by Taboas is used to validate the full cycle model. This two phase heat transfer correlation has a consistent negative deviation from the experimental data, causing the full cycle model to be conservative in the cycle performance it calculates. Finally, a scaling analysis is made in support of the efforts by Bluerise to create a 3 [MW] OTEC plant. Using the geometries supplied by Bluerise it is concluded that a 3 [MW] net power output could be achieved, but it heavily depends on the water side pump power needed and the turbine efficiency, which are not investigated in the current research.

OTEC
organic rankine cycle
Kalina cycle
Performance

http://resolver.tudelft.nl/uuid:32058a2f-b7d0-4a78-bbe4-67ce8f92243d

Student theses

master thesis

© 2019 Johan Stelwagen