A Solar Boom So Successful in Hawaii, It's Been Halted

Re: A Solar Boom So Successful in Hawaii, It's Been Halted

From the article in Scientific American linked in the Op:

More than 4 percent of households have photovoltaics. Hawaii last year led the nation in the portion of its electricity that comes from solar, with 2.6 percent.

and you are claiming 5% penetration is reaching that point now. Not even close.

If there are 40 houses connected on a neighborhood circuit with 240 VAC, 200 A service, and 5% grid-tied PV penetration, then the power line and neighborhood transformer are rated for 1.92 MW and two of those houses have grid-tied PV. If those 2 houses have big 5 kW PV systems pointed optimally and are consuming no power (which is unlikely due to clocks, phantom loads and refrigerator/freezers), then they are outputting a mere 10 kW to the grid on a day with great insolation. That circuit rated for nearly 2 MW is dealing with a power source that is .0052 (.52%) of its rated capacity. This is tiny, so tiny that the phantom loads and refrigerator/freezers in all 40 of those unbelievably empty houses consume all of the 10 kW. If a cloud comes by decreasing the PV power output to 1 kW, it is equivalent to 6 houses turning off 1,500 W electric tea kettles simultaneously like happens in England at tea time. If the grid can not deal with this level of variation, then it is caused by some other problem unrelated to the PV systems.

If all 40 houses have 5 kW grid-tied PV arrays, they would be outputting 200 kW to the grid, 10.4% of the rated power of the neighborhood grid. Now maybe the power company skimped on the specifications by overloading the neighborhood power grid with too many houses or assumed it unlikely that all houses would consume 48 kW simultaneously and therefore installed a neighborhood grid with lower capacity. Given a 2 times safety margin to deal with unusually bright insolation, they would have to have undersized the grid to 20 kW which would be so small it could not power the loads of 40 houses. 5% penetration is well within the noise of loads being switched on and off.

Even this industry sponsored study, Grid Impacts and Solutions of Renewables at High Penetration Levels (Quanta Technology, Dr. Johan Enslin, no date, appears to be 2009 or more recent, PDF file) concludes 20% is the minimum on page 3:

... 20% penetration levels of renewable resources, it requires a dramatic change in planning and operational practices.

In Distributed Smart Solar Technology and Distribution Grid Reliability (Ruba Amarin, Hussam Alatrash, Hisham Othman, Khalid Rustom; KACARE Meeting, May 9, 2012, PDF file), the authors polled 50 utility executives with the results:

At what percentage penetration will the grid face challenges or require upgrade in order to handle high-penetration PV?

% Penetration % of respondents
Greater than 40%8%
25-40%12%
15-24%34%
Less than 15%38%

I want to read a meaningful description of what the utility executives think needs to be upgraded. The phrasing "challenges or require upgrade" is very broad and goes beyond technical issues.

NREL PV Grid Integration (Benjamin Kroposki, May 25, 2010) on page 13 has a chart of the power consumption and production from a 2 kW grid-tied PV array on a house. The power consumption of the house varies by up to 6 kW while the PV system outputs a maximum of 1 kW to the grid after the power consumption settles down. Continuous load from the house is a little under 1 kW judging by the amount before sunrise. The PV system would have to be 5 kW to 6 kW to get variability on a partly cloudy day comparable to the load.

There is a chart showing 30 rated kilowatts of PV on 12 houses, and another showing no noticeable effect on the ~2.5 MW feeder and ~5 MW substation from 238 kW_{ac rated PV on 115 homes. Conclusions from page 18:}

NREL/SMUD Residential PV Research Project – after one year…
•PV penetration levels on the feeder were no greater than 13% under lightly-loaded conditions (2.0MW) on the individual circuit and about 4.0% of the total substation transformer load
•There was no excessive service or substation voltage due to reverse power flow from exporting PV systems
•It was possible to see the effects of the PV systems on the voltage at the individual homes and the distribution transformers
•PV systems have not adversely affected voltage regulation
•The validated model will be used to determine acceptable levels of PV penetration and impacts of a variety of factors on the electric power system
PV Penetration = (PV output (kW)) / (Feeder load (kW))

Since 13% penetration of PV did not adversely affect voltage regulation, 5% certainly will not do it. The Hawaiian utility's argument about grid stability justifying no more grid-tied PV systems is nonsense.