Author Topic: Programming MPPT power curves  (Read 2486 times)

0 Members and 1 Guest are viewing this topic.

kitestrings

  • Hero Member
  • *****
  • Posts: 1376
Programming MPPT power curves
« on: December 01, 2014, 04:20:55 PM »
I programmed slightly different set of power curves for our turbine over the weekend.  Primarily I'm attempting to do two things:

1)   Flatten the curve slightly from cut-in to maybe 12 mph
2)   Raise the overall voltage range and speed of the turbine so as to better match the available power.


I've done this based in large part on observation and from data collected over the last ~year.  Unfortunately I still don't have a good means of collecting wind speed, so it is a bit less precise:


In light load conditions I'd mentioned the audible “awakening” at cut-in.  My thinking is this might help.  There is very little power at cut-in, if we hit it to sharply it spends a lot of time cycling above and below cut-in.  The other fixed point of course is that we have to transition some of the load to the second controller at roughly half of the fully loaded condition (~2.5 kW).  What I tried to do is slightly compress the voltage steps in the upper rpm range.  Overall I've raised the upper voltage limit from about 140V to 160V, or about 2.7x the voltage at cut-in.

I welcome thoughts, suggestions.  Happy Holidays,

~kitestrings

SparWeb

  • Global Moderator
  • Super Hero Member Plus
  • *****
  • Posts: 5452
  • Country: ca
    • Wind Turbine Project Field Notes
Re: Programming MPPT power curves
« Reply #1 on: December 01, 2014, 10:30:54 PM »
So am I looking at two different curves for each controller?   ... and the one you've tweaked is the red line, right?
What happens at the point where the curve reverses?
If the upper graph is related to the lower graph, then I'd say there is some force causing the turbine to avoid the middle power range.


Holidays...  Lucky you.  The edict came down from the eager program manager here, that all staff are "encouraged" to keep their projects moving between christmas and new year's day.  Yeah thanks.
No one believes the theory except the one who developed it. Everyone believes the experiment except the one who ran it.
System spec: 135w BP multicrystalline panels, Xantrex C40, DIY 10ft (3m) diameter wind turbine, Tri-Star TS60, 800AH x 24V AGM Battery, Xantrex SW4024
www.sparweb.ca

kitestrings

  • Hero Member
  • *****
  • Posts: 1376
Re: Programming MPPT power curves
« Reply #2 on: December 02, 2014, 12:18:43 PM »
Quote
So am I looking at two different curves for each controller?   ... and the one you've tweaked is the red line, right?

No.  Yes.

The first chart is the power curve for both controllers.  The little divot in the middle is where the baton gets passed, and the second controller, call it “B”, starts to pick up and contribute.  Basically the last set point of “A” attempts to push the voltage sharply steeper (with no additional current output).  The first step after zero on “B” is set to the same ending value on “A”.  Controller “A” stays loaded to the I_limit programmed, and “B” picks up anything above that level, the first step (V) on “B” is set intentionally low to allow it to wake up and be ready in time.

Quote
I'd say there is some force causing the turbine to avoid the middle power range.

The second chart is, again, the combined output (A + B Logs) of the turbine vs. peak voltage (Vin) for the day.  I suspect that the gap in the mid-spectrum of the chart is attributable to the fact that the turbine is a bit like a solar panel in the shade, if you will.  It is still mounted on a short (25') stub tower.  The output of the turbine is hampered by building and trees, particularly when the leaves were still on the trees here.  So, we're seeing a lot of low wind data points and a spattering of seasonal (late fall to early summer) higher wind events.


SparWeb

  • Global Moderator
  • Super Hero Member Plus
  • *****
  • Posts: 5452
  • Country: ca
    • Wind Turbine Project Field Notes
Re: Programming MPPT power curves
« Reply #3 on: December 02, 2014, 02:31:13 PM »
I get it now; it's the hand-off point between the controllers. 

Maybe if you looked for a windspeed histogram for your area, you might find that the supposedly-smooth Weibull distribution is a little skewed.  That's how it is in my area, and it would also help explain the absence of "middle" wind speeds in your second chart.
No one believes the theory except the one who developed it. Everyone believes the experiment except the one who ran it.
System spec: 135w BP multicrystalline panels, Xantrex C40, DIY 10ft (3m) diameter wind turbine, Tri-Star TS60, 800AH x 24V AGM Battery, Xantrex SW4024
www.sparweb.ca