PV Smoothing Violation variable seems not to consider real violations

Hi Paul, hope everything is fine.

I am writting in the forum due to a problem I have while using the PV smoothing option in a Detail PV + Storage project. My target is to check how different batteries manage to correct the ramp violations. In that way, I am monitoring the PV Smoothing Violation variable and I found that the violations indicated are measured in the variable PV Smoothing OutPower. However, PV Smoothing OutPower is not equal to System Power Generated, which I assume is the real injected power and, therefore, the indicated violations are not the real ones.

When I set a battery power very high expecting to have 0 violations, I can see that there are no violations indicated but, they actually exist in System Power Generated. Am I doing something wrong? And, if the answer is no, are there any possible solutions?

Thanks in advance.
Kind regards.
Miguel

Hi Miguel,

There is a description of the PV Smoothing battery dispatch algorithm in the Battery Dispatch topic of SAM's Help system: samrepo.nrelcloud.org/help/battery_dispatch_fom.htm .

PV smoothing outpower is the desired output power of the PV-battery system, not the actual power. The actual power is "electricity to grid," which in this case is the same as "system power generated."

If you are seeing more ramping than you expect, you may need to adjust some of the dispatch parameters, such as maximum ramp rate, correct up-ramp violations, and ramp timestep multiplier (to set the ramp rate interval).

Best regards,
Paul.

Hi Miguel,

After some further investigation, we discovered a bug with the PV Smoothing dispatch algorithm that we will fix for the next update to SAM 2020.12.02, which we hope to release by the end of this month:

github.com/NREL/ssc/issues/814

Best regards,
Paul.

Hi Paul:

I look forward to the new version.

Thank you.
Kind regards
Miguel

Hi Paul:

I have checked that the bug was a problem with a beta version not available for the users, so no problem there. I would like to attach a figure about the problem I have, it is shown how the violations measured in "System power generated" reach a fixed value even for really high battery powers. Battery enery is not a problem as it is set to a very high value. Violations measured in pv smoothing outpower actually reach zero. 

Maybe, the problem is that the feedback variable in the pv smoothing algorithim is pv smoothing outpower and not the real injected power. Big batteries are able to achieve the demanded ramp rate but their capabilites are not used due to the control algorithim. I am using a weather file with 1 minute resolution and a ramp interval of 1 minute. I just wanted to be sure if I am explaining myselft properly and if the new version will cover these issues.

Thanks in advance
Kind regards
Miguel

Hi Miguel,

We are working with the developer of the PV Smoothing battery dispatch algorithm to address this question, and will consider changing this behavior. We probably will make this change for the Fall 2022 release of SAM rather than as an update to SAM 2021.12.02.

Best regards,
Paul.