Best way to flag times during backtracking?

[Will Hobbs]

If I wanted to flag time intervals during which a tracker is backtracking, i.e., produce an “is it backtracking” Boolean timeseries, what would be a good way to do this with pvlib?

I’m thinking about simply running pvlib.tracking.single axis twice, once with backtracking enabled and once with it disabled, and then comparing the output. That feels a bit like a “hack,” but maybe it’s not too bad. 

Bonus question: how would you flag morning vs afternoon backtracking? I was thinking of something like `np.where(solar_position.azimuth < 180
…`

Background is that I’m working on setting tracker position error thresholds/tolerances that depend on tracking mode, where error in the direction of not causing self shade is more tolerable than error in the direction of causing shade. 

Thanks!

Will

[cwh...@sandia.gov]

Running singleaxis twice would work, or you can get the ideal tracking angle from shading.projected_solar_zenith_angle (which is what singleaxis does).

Morning vs. afternoon could be relative to the tracker, if the axis is far enough from 180.

I wouldn't be opposed to adding that "is it backtracking" boolean time series to the dict that singleaxis returns. But I'd define it as True when true tracking, False otherwise.

Cliff

[Will Hobbs]

Thanks, Cliff. 

Just to be clear, what you are suggesting as a possible addition to the singleaxis function would be an “is it true tracking” (or, “not backtracking”) bool output. Is that correct? 

And do you think I should create an issue to start a discussion on that?

Will

--
You received this message because you are subscribed to the Google Groups "pvlib-python" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pvlib-python...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/pvlib-python/ea0e5e62-6cd5-4a7c-bf7d-08c625fa51c6n%40googlegroups.com.

[cwh...@sandia.gov]

>  a possible addition to the singleaxis function would be an “is it true tracking”

yes

>  do you think I should create an issue

yes

[Will Hobbs]

I created an issue: https://github.com/pvlib/pvlib-python/issues/2672. 

And, as Adam Jensen pointed out in a comment there, morning vs afternoon can be inferred from rotation angle (positive vs. negative). I think you would need to be careful about sign convention relative to the axis azimuth, e.g., 0 vs 180, but otherwise it seems like that would work well.

Will