Dynamic Modeling of Components on the Electrical Grid

Bailey YoungWofford College

Dr. Steven Fernandez and Dr. Olufemi Omitaomu Computational Sciences and Engineering Division

August 2009

2 Managed by UT-Battellefor the U.S. Department of Energy

Overview

• Introduction and background

• Research objectives

• Methods

• Results and validation

• Conclusion

• Future research

• Acknowledgments

3 Managed by UT-Battellefor the U.S. Department of Energy

Introduction and background

• FEMA needs a tool to display outage areas in large storms, to predict electrical customer outages

• Use electrical outage predictions to– Create better emergency responses– Protection for critical infrastructures

• VERDE : Visualizing Energy Resources Dynamically on Earth

4 Managed by UT-Battellefor the U.S. Department of Energy

VERDE system

• Simulates the electric grid

• Provides common operating picture for FEMA emergency responses.

• Uses Google earth as platform

• Provides real-time status of transmission lines

• Provides real-time weather overlay

Streaming analysis

Weather overlay

Wide-area power grid situational awareness

Impact models

5 Managed by UT-Battellefor the U.S. Department of Energy

Energy infrastructure situational awareness

• Coal delivery and rail lines

• Refinery and off-shore production platforms

• Natural gas pipelines

• Transportation and evacuation routes

• Population impacts from LandScan1 USA population database

1. Bhaduri et al., 2002; Dobson et al., 2000

6 Managed by UT-Battellefor the U.S. Department of Energy

Research objectives

• Determine electrical customers for counties of US

• Translate MATLAB code to Java

• Program to estimate electrical substation service and outage areas

• Compare with geospatial metrics, MATLAB output with Java output

• Determine reliability of Java code

7 Managed by UT-Battellefor the U.S. Department of Energy

Methods

Number of customers in 2008 per county is found by using the equations

__________________ House2000 + Firms2000

Pop2000CF =

CF = Correction Factor Pop2000 = Population in 2000Pop2008 = Population in 2008House2000 = Nighttime population in 2000Firms 2000 = Firms in 2000Customers = 2008 customers

= Customers_________CF

Pop2008

8 Managed by UT-Battellefor the U.S. Department of Energy

Methods

• Compare customer estimates from correction factor with known customer data

• Translate modified Moore-based algorithm in MATLAB to Java code

• Use program to predict electrical substation area

• Use researched geospatial metrics to compare substation’s area in MATLAB and Java

• Use electrical substation locations given by CMS energy of Michigan area

9 Managed by UT-Battellefor the U.S. Department of Energy

Methods

• Modified Moore-based algorithm

• Peak substation demand data from commercial data sets and population data from LandScan

• Substation geographic location and peak demand from Energy Visuals– Cells approximately 1km – Demand and supply per cell

Supply DataDemand DataCombined Data

Inputs

Geo-locatedpopulation data

Geo-locatedsubstation data

10 Managed by UT-Battellefor the U.S. Department of Energy

Results and validation

• Receive output from Java code

• Compare MATLAB and Java outputs using Michigan substation locations

Sample output of code implementation provided by Dr. Omitaomu

Each color represents a different substation service area

11 Managed by UT-Battellefor the U.S. Department of Energy

Results and validation

• Compare customer estimates from correction factor with known customer data – Use known customer data in seven Florida and Maryland

counties• Have non-disclosure agreements with these utility companies

• These companies only electrical provider for these counties

– Use ratio of predicted to actual customer estimates from the correction factor in these counties: 100.7% ± 13.6%

– Utility companies use this number to convert population to customers

12 Managed by UT-Battellefor the U.S. Department of Energy

Results and validation Showing Correction Factor Vs. Counties

0.0

100.0

200.0

300.0

400.0

500.0

600.0

700.0

800.0

900.0

1 to

1.4

1.4

to 1

.5

1.5

to 1

.6

1.6

to 1

.7

1.7

to 1

.8

1.8

to 1

.9

1.9

to 2

.0

2.0

to 2

.1

2.1

to 2

.2

2.2

to 2

.3

2.3

to 2

.4

2.4

to 2

.5

2.5

to 2

.6

2.6

to 2

.7

2.7

to 2

.8

2.8

to 2

.9

2.9

to 3

.0

3.0

to 3

.1

3.1

to 3

.2

3.2

to 3

.3

3.3

to 3

.4

3.4

to 3

.5

3.5

to 3

.6

Correction Factors Between

Nu

mb

er o

f C

ou

nti

es

Legend

counties_test

Correction Factor

1.256757 - 1.679397

1.679398 - 1.880786

1.880787 - 1.986974

1.986975 - 2.072005

2.072006 - 2.151424

2.151425 - 2.236053

2.236054 - 2.339809

2.339810 - 2.494391

2.494392 - 2.738462

2.738463 - 3.150787

3.150788 - 4.660821

Correction factor vs. counties

Nu

mb

er

of

Co

un

ties

Correction Factors

• Shows representation of correction factor by county.

13 Managed by UT-Battellefor the U.S. Department of Energy

Conclusions

• Correction factor data is effective in predicting customer population

• Correction factor data imported into real time VERDE data

• Reliability of Java code to be determined after verification

• Used within the VERDE system to help with emergency response and outage prediction

14 Managed by UT-Battellefor the U.S. Department of Energy

Future Research

• Predict materials possibly lost in substation outage areas– Poles– Wires

• Compare Java output with actual substation service data– Use substation locations to get service area– Compare service area with actual geographic areas provided

through a non-disclosure agreement

15 Managed by UT-Battellefor the U.S. Department of Energy

References

• Omitaomu, O.A. and Fernandez, S.J. (2009). A methodology for enhancing emergency preparedness during extreme weather conditions. Proceedings of the 3rd Annual AFIT/WSU Mid-West CBRNE Symposium, Wright-Peterson Air Force Base, September 22-23.

• Sabesan, A, Abercrombie, K. , Ganguly, A.R., Bhaduri, B., Bright, E. , and Coleman, P. (2007) Metrics for the comparative analysis of geospatial datasets with applications to high-resoluttion grid-based population data. GeoJournal.

• Bhaduri, B. Bright, E., Coleman, P., and Dobson, J. (2002): LandScan: locating people is what matters, Geoinformatics, 5(2):34-37.

16 Managed by UT-Battellefor the U.S. Department of Energy

Acknowledgements

• Department of Energy

• UT Battelle

• Oak Ridge National Laboratory

• Research Alliance in Math and Science Program and Debbie McCoy

• Dr. Steven Fernandez and Dr. Olufemi Omitaomu