spatial forecast presentation
TRANSCRIPT
-
8/9/2019 Spatial Forecast Presentation
1/15
Company
name-1
01-04-06-
Long-Range SpatialLoad Forecasting
-
8/9/2019 Spatial Forecast Presentation
2/15
Company
name
-2
Spatial Forecasting
Locational forecasts of load growth by assigning load
to available land.
Determines spatial characteristics of long range load
growth.
Most accurate applied to undeveloped land.
Overall forecast tied to corporate econometric energy
/ demand forecast.
-
8/9/2019 Spatial Forecast Presentation
3/15
Company
name
-3
Infrastructure
Interstates / highways
Major Roadways
Railroads Waterways
Longitude / latitudecoordinate based
Can be imported from GISdatabases, or fromcommercially-availabledatabases
-
8/9/2019 Spatial Forecast Presentation
4/15
Company
name
-4
Land Use Definition
Three main classes:
Residential
Commercial
Industrial
Other classes also available
Can be based on:
Land-use databases
Satellite Imagery
Planner's knowledge and
maps
-
8/9/2019 Spatial Forecast Presentation
5/15
Company
name
-5
Land Use Definition Detail
Existing development is
designated according to
land use type.
Undeveloped areas are
designated according tozoning / development
plans.
Future Development
(Outline Only)Existing Development
(Solid)
-
8/9/2019 Spatial Forecast Presentation
6/15
Company
name
-6
Spatial Forecasting
Precedence of load growth in available areas
determined by proximity to:
Infrastructure
Interstates / Highways
Major Roadways
Railroads
Waterways
Surrounding Development
Residential
Commercial
Industrial
-
8/9/2019 Spatial Forecast Presentation
7/15
Company
name
-7
Growth of Small Areas
Governed by S-curve
characteristics
In each year, growth of
small areas constrained
by overall growth in thestudy area0%
20%
40%
60%
80%
100%
0 2 4 6 8 10 12 14 16 18 20
Years
PercentGrow
th
Start = 0, tau = 5
Start = 0, tau = 10
Start = 5, tau = 5
0)t(L = for t < t0
=
20 )tt(
max e1L)t(L
for t > t0
-
8/9/2019 Spatial Forecast Presentation
8/15
Company
name
-8
Spatial Forecast Calibration
The spatial forecast is calibrated to known weather-
adjusted load levels in Year 0 using:
Substation Area Definition
Areas served by each substation defined
Load areas calibrated to non-coincident peak load
TLM (Transformer Load Management) Data
Optional
0.5 x 0.5 mile areas, for example
Load areas calibrated to peak load in each area
-
8/9/2019 Spatial Forecast Presentation
9/15
Company
name
-9
Substation Area Definition
Substation service
areas defined based
on predominant
station serving each
TLM area.
-
8/9/2019 Spatial Forecast Presentation
10/15
Companyname
-10
TLM Load Density
TLM load density by
quarter square mile
areas, for example
TLM data allows
density calibration to
experienced load level
-
8/9/2019 Spatial Forecast Presentation
11/15
Companyname
-11
Forecast Results
Forecast provides a
yearly, stepped
evaluation of the
spatial load growth.Year 10
Year 20Year 0
-
8/9/2019 Spatial Forecast Presentation
12/15
Companyname
-12
Spatial Forecast Results
Forecast results show spatial distribution and
magnitude of load growth.
Results available in visual (map display) format and
tabular format
Results can be rolled up by:
Substation
District
Land Use Classification
-
8/9/2019 Spatial Forecast Presentation
13/15
Companyname
-13
Spatial Forecast Caveats
Most accurate when substantial vacant land isavailable for development. (Relatively inaccurate inredevelopment areas, e.g., CBDs, gentrification
areas.)
Best applied to determine long range trends. (Initialgrowth rates from short-term substation loadforecasts.)
Load growth allocations based on overall econometricforecast.
-
8/9/2019 Spatial Forecast Presentation
14/15
Companyname
-14
Spatial Forecast Accuracy
A four year old forecast was recently reviewed foraccuracy and found to provide accuracy within 4% formultiple (6-8) substation areas.
The the review considered:
Major load transfers
Differences in forecast/actual weather conditions
Deviations in large spot loads
Multiple substation areas were used since there wassignificant discretion in new load substationconnections
-
8/9/2019 Spatial Forecast Presentation
15/15
Companyname
-15
Spatial Forecast Accuracy
The four year old forecast review also considered
deviation between the econometric system-wide load
growth forecast and the actual system-wide load
growth over the prior four years.
Accounting for these differences showed the forecastwas:
2.5% accurate in all but two of the multiple substation areas
5% accurate on a per substation basis for approximately85% of the substations.