CSP models (ITD vs. cycle efficiency)

Dear Paul and the team,

I have a question regarding the initial temperature difference(ITD) in the SAM CSP cooling system. 

For example in Linear Fresnel Direct Steam model, in order to achieve a higher steam temperature at the outlet of the turbine, I increased the ITD in the cooling system (also decreased the rated cycle efficiency and increased the ambient temperature using data from the reference) and kept other parameters as default.

By doing that I was expecting a lower annual power generation since less energy is converted to electricity in the turbine. It was true if I use SAM 2018 version, but in the 2020 version(revision 3) I got an opposite result. The annual electricity generation actually increases from 240 GWh to 255 GWh after I increase the default ITD from 16 to 40 degree C. May I ask if I am correct to expect a lower elec. generation? And may I ask why the results are opposite in two different versions, but I didn't see a major change for CSP Linear Fresnel model from the SAM release note?

Much appreciated for your time and help!

Best regards,
Zhuoran

Dear Zuoran,

Here is a link to a related discussion that may be helpful:

sam.nrel.gov/forum/forum-general/1993

Some other notes:

• The design point efficiency is an input, so it only changes when you change its value.
• The initial temperature difference (ITD) correlates the design ambient temperature and cycle efficiency to a condenser temperature. (It is only available for power cycles with either air cooling or hybrid cooling.)
• At high ITD values, the model imposes a ceiling on the maximum condenser temperature, so the resulting off-design temperatures may not respond as you expect. High ITD values are not realistic for conventional plant designs, although we understand some people may be trying to manipulate its value to approximate combined heat and power applications.
• We did make some changes to the power cycle heat rejection model in versions since 2018, which is the likely cause of the difference in results you are seeing between versions.

Best regards,
Paul.