IEEE Power & Energy Magazine - May/June 2018 - 52

MA
NH

CT
RI

3 July 2007
00:00

2 July 2007
12:00

2 July 2007
00:00

1 July 2007
12:00

0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
1 July 2007
00:00

BTM PV Power, Total Capacity (%)

methods, rather than forecasting demand for each light fixture or appliance at the house or business level, a diversity
of coincident, individual electricity end uses are aggregated
into electricity consumption for an area, and statistical models

ME
VT

figure 9. The average state-level BTM PV power profiles,
1-2 July 2007.

trained on historical data leverage the implicit load patterns.
in turn, these models can then accurately predict load based
solely on calendar and weather inputs.

High-Quality, High-Fidelity Results
figure 8 illustrates a comparison of the town-level simulation results (left) and the town-level performance of more
than 2,000 actual pv installations (right) on (a) 1 may 2014 at
12:30 eastern prevailing time (ept) and (b) 15 october 2014
at 12:30 ept. Both data sets represent the relative amount of
solar production at the municipal level, as described previously. Despite the two sets being independently sourced, there
is broad agreement between the simulated versus measured
pv data. the high degree of spatial color coherence in the
representation of both sets demonstrates the value of using the
town-level, per-unit production data in the estimation process,
as the heat maps clearly depict their ability to encapsulate the
effects of local weather (mainly cloud cover) on the Btm pv
fleet. for example, on (a) the 1 may plots, both maps illustrate
low pv output due to mostly cloudy conditions throughout much
of the southeastern portion of the region and higher pv output in
northwestern parts of the region due to less cloud cover. similarly,
(b) the 15 october plots illustrate mostly sunny conditions in the
southeast, with the effects of greater cloud cover reflected in
various pockets of lower pv output in the remainder of the region.

table 4. The average annual capacity factors for simulated BTM PV in New England states.
(Note: the NSRDB data for Maine is missing for 2010.)

52

Year

Massachusetts

Connecticut

Maine

New Hampshire

Rhode Island

Vermont

1998

15.5%

15.6%

14.3%

14.8%

15.8%

14.6%

1999

16.5%

16.4%

15.3%

15.7%

16.6%

15.7%

2000

15.8%

16.0%

14.6%

14.9%

16.2%

14.6%

2001

16.6%

16.5%

15.5%

15.8%

16.9%

15.5%

2002

16.1%

16.1%

15.0%

15.3%

16.4%

15.1%

2003

14.9%

14.7%

14.4%

14.5%

15.1%

14.6%

2004

15.2%

15.2%

14.7%

14.7%

15.4%

14.6%

2005

15.3%

15.4%

14.2%

14.6%

15.7%

14.8%

2006

15.1%

15.2%

13.7%

14.2%

15.2%

13.7%

2007

15.6%

15.6%

14.5%

14.8%

15.9%

14.7%

2008

15.0%

15.2%

13.7%

14.2%

15.3%

14.0%

2009

14.8%

14.7%

13.8%

14.2%

15.0%

14.1%

2010

15.4%

15.4%

-

14.4%

15.7%

13.9%

2011

15.2%

15.2%

14.1%

14.4%

15.6%

14.3%

2012

15.8%

15.9%

14.8%

15.1%

16.1%

15.2%

2013

15.8%

15.8%

14.4%

14.9%

16.0%

14.5%

2014

15.4%

15.5%

14.3%

14.6%

15.8%

14.4%

Average

15.5%

15.5%

14.5%

14.8%

15.8%

14.6%

ieee power & energy magazine

may/june 2018
Table of Contents for the Digital Edition of IEEE Power & Energy Magazine - May/June 2018
