CSP Power cycle: Alternative Power Cycles

Hi Paul,

I understand SAM assumes a steam Rankine cycle for its power cycle computations. I would like to hear your opinion on how good/bad it would be approximating a Brayton cycle with an alternative HTF (such as air or CO2) by using the SAM model and specifying the correct HTF temperatures, corresponding cycle efficiencies and scaling the power cycle cost accordingly?

Thank you very much

Hello,

This is a good question. We do not yet have good data on how well SAM's regression model for the steam cycle works for cycles other than the Rankine cycle.

To learn more about SAM's steam cycle model, you can explore the sam_mw_pt_TYPE224.f90 source code file (the TRNSYS source code files are in the /exelib/trnsys/source of your SAM installation folder, and read the (Type 224), and refer to Chapter 3 in the physical trough model reference manual ( PDF 3.7 MB ).

The answer to your question depends in part on how well you can estimate the alternative cycle's off-design performance, and how well that performance matches that of the basis steam Rankine cycle for SAM's regression model. Note that the model inputs are the inlet HTF temperature and mass flow rate, and ambient temperature, and the outputs are the cycle power and efficiency. It also depends on the level of uncertainty that is acceptable for your analysis.

Another issue is with modeling cooling parasitic loads. If the off-design cooling loads are distributed over different temperature ranges than the Rankine cycle, the cooling parasitic loads will be somewhat different. (For an open-air Brayton cycle, there may be no cooling parasitic loads). This is less of a problem than the steam cycle modeling issues because the cooling parasitic load calculations can be post-processed (after the steam cycle calculations), and the cooling model outputs do not strongly interact with other CSP component models during the hourly TRNSYS simulations.

Best regards,
Paul.