There is some interesting work going on upstairs here in my utility. SMUD has partnered with one regional developer to apply integrated solar roofs in a new low density development here in Sacramento. I still hold contempt for such sprawlish development...but it's green, right? Slap up some PV and you're green guilt free! But I digress.
The industry doesn't have experience with saturated solar installations; that is, a large quantity of discrete, independent systems operating in the same distributed electrical system. The size of the systems isn't the issue; it's the quantity, because each PV system tries to find its maximum power point.
If we thought of an inverter as a simple charge controller, the PV modules would operate to the utility voltage. But all PV systems have a maximum power point, at which a certain PV voltage produces the most power. This voltage might not 'mesh' with the utility voltage. What a typical inverter does is find this MPP voltage and will operate the PV modules there to extract full power regardless of the utility voltage. But by doing so, each inverter influences utility voltage.
What we think is happening is that once you saturate a network with a number of PV systems, each system now competes with one another to find its MPP. Each influences the network voltage, and the consequence is there might only be one system that can operate at MPP; the rest are operating at less than optimum.
The point of this is to say that we are going to discover a whole new set of problems if we think we can get 100% of our electricity from wind and solar as Gore suggested yesterday. In ten years, no less! The model nation of today is Denmark, which gets 20% from renewables. Do we think that our nation could possibly, even remotely, achieve even the current Denmarkian standard in ten years, with a population replete with energy squanderers and energy policies that can't possibly look out farther than then next election or economic cycle?
Subscribe to:
Post Comments (Atom)
4 comments:
Hey man,
You're off base a bit. The MPPT algorithm only affects the PV system DC voltage to extract maximum power from the modules. The inverter matches grid voltage and frequency on the AC side at all times. If these go out of a fairly narrow range, the inverter will trip off per distributed generation requirements (IEEE 1547).
It is true that if inverters are exporting and injecting current onto the distribution system, voltage along a feeder will rise instead of fall. But this has nothing to do with an interaction of MPPT algorithms, it is simply a function of having net generation on the feeder.
SMUD and NREL did a good study on this topic on the Anatolia III subdivision recently, ask around.
Cheers,
- A PV Engineer
Thank you for your post. I remember this discussion with a distribution engineer who suggested that multiple PV systems may be in competition with one another, each looking for its MPP.
However, as you point out, the MPP is on the DC side, looking for the optimum DC voltage to operate to. I believe that my system, a single array input as a single string into the inverter, is optimized as a single array (one MPP calculation), not on each individual panel.
In any event, perhaps my question is better worded by "wondering what effect (if any) a changing feeder voltage would have on inverter performance." That is -- if my array is indeed producing DC at an MPP voltage point that's independent of utility voltage, how does this influence my DC-to-AC power conversion? Or...does it?
Profitez de votre eau, profiter de votre vie. Nous offrant toutes sortes de salle de bains Robinetterie Vasque et robinets de cuisine pour vous. Consultez notre boutique en ligne et de trouver votre meilleur robinet ici: www.robinetonline.com
FindTaps - Top quality, Great value bathroom and kitchen sinks and taps from leading brands,Buy designer taps in our website,Amazing selections here, low prices and high quality!UK taps available from http://www.findtaps.co.uk
Post a Comment