Spatial Electric Load Analysis for Substation Siting and Load Balancing At United Power, the engineering and GIS groups were tasked with answering the following question, “Will we have the infrastructure to support future demand in 5 or 10 years?” We turned to spatial technologies to provide management with an accurate and detailed GIS-generated load density forecast. Demand and energy readings from CIS were integrated with GIS to produce base grids for summer and winter peaks. The analysts combined base load grids with 2 forecast sources to produce long-range forecast raster grids. The complex analysis process was performed with multiple Model Builder models for consistency and repeatability. ESRI’s Spatial Analyst extension performed raster analysis. By maintaining a spatial history of power consumption, accurate data is readily available for a plethora of statistical studies and testing what- if scenarios. Using GIS technologies for planning purposes increases forecast accuracy and efficiency and creates a roadmap for future land and ROW acquisitions.
44
Embed
Spatial Electric Load Analysis for Substation Siting and ... · Spatial Electric Load Analysis for Substation Siting and Load Balancing At United Power, the engineering and GIS groups
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Spatial Electric Load Analysis for Substation Siting and Load Balancing
At United Power, the engineering and GIS groups were tasked with answering the following question, “Will we have the infrastructure to support future demand in 5 or 10 years?” We turned to spatial technologies to provide management with an accurate and detailed GIS-generated load density forecast. Demand and energy readings from CIS were integrated with GIS to producebase grids for summer and winter peaks. The analysts combined base load grids with 2 forecast sources to produce long-range forecast raster grids. The complex analysis process was performed with multiple Model Builder models for consistency and repeatability.ESRI’s Spatial Analyst extension performed raster analysis. By maintaining a spatial history of power consumption, accurate data is readily available for a plethora of statistical studies and testing what-if scenarios. Using GIS technologies for planning purposes increases forecast accuracy and efficiency and creates a roadmap for future land and ROW acquisitions.
Spatial Electric Load Analysis for Substation Siting and Load Balancing
David Hollema – GIS AnalystJared Weeks – Electrical Engineer
United Power, Inc.Brighton, Colorado
ESRI EGUG 2008
Today’s agenda
▪ Who we are▪ Long range forecast goals▪ Spatial load analysis basics▪ Components of spatial load forecasting▪ Load center prediction▪ Results and looking ahead
United Power Facts
Rural electric cooperative headquartered in Brighton, CO Incorporated in October of 1938 Wires hot in 1940 to 750 customersNearly 65,000 customers today covering 900 square milesHistorically fast growing – up to 5000 new accounts per yearAmong the top 10 fastest growing coops nationwide
Long range forecast goals and tasks
▪ Spatially project coincidental peak load over thenext 10 years
▪ Define substationinfluence areas
▪ Forecast and locate future load centers for substation placement
▪ Determine substation transformerupgrades
▪ Update every 3 years (modular)
What is spatial load analysis?
▪ A process of looking at historical electric power consumption with a spatial (e.g. mapping) component– Seasonal peak focus– Demand (kW) or energy (kWh)
▪ Typically includes forecasting for substation siting▪ Change detection analysis▪ Temporal study▪ Used to optimize current electric distribution system
Monthly Demand - 1998 to 2008
0
50,000
100,000
150,000
200,000
250,000
300,000
350,000
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
Mon
thly
Coi
ncid
ent P
eak
Dem
and
(kW
)
Monthly Demand - 1998 to 2008
0
50,000
100,000
150,000
200,000
250,000
300,000
350,000
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
Mon
thly
Coi
ncid
ent P
eak
Dem
and
(kW
)
NO SPATIAL COMPONENTWinter Peak
Summer Peak
Summer Peak Demand Linear Forecast
0
100000
200000
300000
400000
500000
600000
2000 2002 2004 2006 2008 2010 2012 2014 2016 2018
Peak
Dem
and
(kW
)
5yr trendHistoric
Summer Peak Demand Linear Forecast
0
100000
200000
300000
400000
500000
600000
2000 2002 2004 2006 2008 2010 2012 2014 2016 2018
Peak
Dem
and
(kW
)
5yr trendHistoric
NO SPATIAL COMPONENT
Spatial Load Forecast Approach – 3 components
1. Base Map▪ Historical snapshot of peak seasonal load▪ Peak load density map
2. Metrostudy Data (www.metrostudy.com)
▪ Provider of housing data▪ Used to forecast residential growth
3. Point Loads▪ Internal knowledge of future large commercial loads from district reps
20022002
Meter Reading Extraction
Base map preparation
Load Table Preparation
Spatial Join
Rasterization
– “need to know where you’ve been to know where you’re going”
Must be repeatable and modular!
Base Map Preparation
Meter Reading Extraction
▪ Primary input for entire analysis
▪ Interested in 1 moment in time with coincident peak demand, settle for 30 day data (billed monthly)
▪ Oracle view used to extract CIS data to SDE instance
▪ Revolving billing cycle makes capturing monthly peak difficult
Meter Read-type Breakdown
55%35%
10%
Manual ReadCarrier Line AMRDrive-by AMR
Base Map Preparation
Base Map Preparation
SDE Oracle view into CIS databaseSELECT MAX(BI_CONSUMER.BI_ACCT) AS ACCT_NBR, MAX(BI_TYPE_SERVICE.BI_SRV_STAT_CD) AS SRV_STAT_CD,
/* BI_CONSUMER_VIEW_1.BI_ADDR_TYPE,*/MAX(BI_CONSUMER_VIEW_1.BI_SORT_NAME) AS NAME,MAX(BI_CONSUMER_VIEW_1.BI_LNAME) AS LAST_NAME, MAX(BI_CONSUMER_VIEW_1.BI_FNAME) AS FIRST_NAME, BI_SRV_LOC.BI_SRV_MAP_LOC AS SERVLOC,
/* BI_HIST_USAGE.BI_CUR_HIST_SW, */MIN(BI_HIST_USAGE.BI_RATE_SCHED),MAX(BI_HIST_USAGE.BI_PRES_READ_DT) AS READ_DATE, SUM(BI_HIST_USAGE.BI_USAGE) AS KWH,MAX(BI_HIST_USAGE.BI_BILL_DMD_HIST) AS DEMAND_KW,MAX(BI_HIST_USAGE.BI_REV_YRMO)
/* BI_HIST_USAGE.BI_PRES_MTR_RDG AS KWH */
FROM ((((([email protected] BI_CONSUMER INNER JOIN [email protected] BI_AR ON BI_CONSUMER.BI_ACCT=BI_AR.BI_ACCT) INNER JOIN [email protected] BI_CONSUMER_VIEW_1 ON BI_CONSUMER.BI_ACCT=BI_CONSUMER_VIEW_1.BI_VWN_CO_ACCT) INNER JOIN [email protected] BI_TYPE_SERVICE ON (BI_AR.BI_ACCT=BI_TYPE_SERVICE.BI_ACCT) AND (BI_AR.BI_TYPE_SRV=BI_TYPE_SERVICE.BI_TYPE_SRV)) INNER JOIN [email protected] BI_HISTORY ON ((BI_TYPE_SERVICE.BI_ACCT=BI_HISTORY.BI_ACCT) AND (BI_TYPE_SERVICE.BI_TYPE_SRV=BI_HISTORY.BI_TYPE_SRV)) AND (BI_TYPE_SERVICE.BI_SRV_LOC_NBR=BI_HISTORY.BI_SRV_LOC_NBR))INNER JOIN [email protected] BI_SRV_LOC ON BI_TYPE_SERVICE.BI_SRV_LOC_NBR=BI_SRV_LOC.BI_SRV_LOC_NBR) INNER JOIN [email protected] BI_HIST_USAGE ON ((((BI_HISTORY.BI_ACCT=BI_HIST_USAGE.BI_ACCT) AND (BI_HISTORY.BI_TYPE_SRV=BI_HIST_USAGE.BI_TYPE_SRV)) AND (BI_HISTORY.BI_SRV_LOC_NBR=BI_HIST_USAGE.BI_SRV_LOC_NBR)) AND (BI_HISTORY.BI_HIST_CD=BI_HIST_USAGE.BI_HIST_CD)) AND (BI_HISTORY.BI_BILL_DT_TM=BI_HIST_USAGE.BI_BILL_DT_TM)
WHERE BI_CONSUMER_VIEW_1.BI_ADDR_TYPE=N' ' AND BI_HIST_USAGE.BI_CUR_HIST_SW=N'Y' AND BI_HIST_USAGE.BI_REV_YRMO=200807AND BI_RATE_SCHED NOT IN ('SC0','SC1','SC5') --removes Golden Aluminum (fed off transmission) and Frederick/Evanston area primary meters--such as 3236-2652-0 and 3224-4552-0 (double-counted load)AND SUBSTR(BI_SRV_LOC.BI_SRV_MAP_LOC,5,1) !='6'AND BI_SRV_LOC.BI_SRV_MAP_LOC !='333305250' --removes Spindle Hill Energy Peak Plant fed off transmission
GROUP BY BI_SRV_LOC.BI_SRV_MAP_LOC
CIS GIS
Base Map Preparation
Base Map Preparation
Load table preparation
▪ Engineering calculation fields added▪ Some meters billed by usage (kWh), others by
usage and demand (kW)▪ Estimate coincident peak kW using kWh