a call for greater transparency in energy modeling and...
TRANSCRIPT
![Page 1: A Call for Greater Transparency in Energy Modeling and Analysisscienceag.cluster013.ovh.net/wp-content/uploads/2014/02/... · 2014. 6. 23. · Energy-economy optimization (EEO) models](https://reader035.vdocument.in/reader035/viewer/2022071611/614ad6bc12c9616cbc69acd7/html5/thumbnails/1.jpg)
A Call for Greater Transparency in Energy Modeling and Analysis
Joe DeCarolis
Assistant Professor
Dept of Civil, Construction, and Environmental Engineering
NC State University
[email protected]; @jfdecarolis 1
Les Houches, France 7 February 2014
![Page 2: A Call for Greater Transparency in Energy Modeling and Analysisscienceag.cluster013.ovh.net/wp-content/uploads/2014/02/... · 2014. 6. 23. · Energy-economy optimization (EEO) models](https://reader035.vdocument.in/reader035/viewer/2022071611/614ad6bc12c9616cbc69acd7/html5/thumbnails/2.jpg)
Talk outline
Introduction to energy economy optimization models
Problems with the status quo; recommendations to fix
Introduction to Tools for Energy Model Optimization and Analysis (Temoa)
2
![Page 3: A Call for Greater Transparency in Energy Modeling and Analysisscienceag.cluster013.ovh.net/wp-content/uploads/2014/02/... · 2014. 6. 23. · Energy-economy optimization (EEO) models](https://reader035.vdocument.in/reader035/viewer/2022071611/614ad6bc12c9616cbc69acd7/html5/thumbnails/3.jpg)
Driving questions
How does the world balance the costs of greenhouse gas mitigation in the near-term versus long-term?
What are the anticipated economic and environmental impacts
associated with future environmental policies and energy technology deployments?
How do decision makers craft energy planning strategies that are
robust to future uncertainties? How do decision makers incorporate broader environmental
sustainability considerations — beyond simply limits to greenhouse gas emissions — into their strategies?
3
![Page 4: A Call for Greater Transparency in Energy Modeling and Analysisscienceag.cluster013.ovh.net/wp-content/uploads/2014/02/... · 2014. 6. 23. · Energy-economy optimization (EEO) models](https://reader035.vdocument.in/reader035/viewer/2022071611/614ad6bc12c9616cbc69acd7/html5/thumbnails/4.jpg)
Energy-economy optimization (EEO) models
Large uncertainties combined with a mix of technical, economic, and moral considerations preclude definitive answers to the questions above.
Model-based analysis can deliver crucial insight that informs key decisions.
Energy-economy optimization (EEO) models refer to partial or general equilibrium models that minimize cost or maximize utility by, at least in part, optimizing the energy system over multiple decades
• Self-consistent framework for evaluation
• Explore how effects may propagate through a system
• Expansive system boundaries and multi-decadal timescales
What can we usefully conclude from modeling exercises where uncertainty is rigorously quantified?
4
![Page 5: A Call for Greater Transparency in Energy Modeling and Analysisscienceag.cluster013.ovh.net/wp-content/uploads/2014/02/... · 2014. 6. 23. · Energy-economy optimization (EEO) models](https://reader035.vdocument.in/reader035/viewer/2022071611/614ad6bc12c9616cbc69acd7/html5/thumbnails/5.jpg)
Technology explicit modeling
5
Capital Cost ($M/PJ/yr) Fixed O&M ($M/PJ) Variable O&M ($M/PJ) Capacity factor Efficiency Emissions coefficient (kton/PJ)
Objective function: minimize present cost of energy supply over a defined time horizon Decision variables: activity (PJ) and capacity (PJ/yr) for each technology
![Page 6: A Call for Greater Transparency in Energy Modeling and Analysisscienceag.cluster013.ovh.net/wp-content/uploads/2014/02/... · 2014. 6. 23. · Energy-economy optimization (EEO) models](https://reader035.vdocument.in/reader035/viewer/2022071611/614ad6bc12c9616cbc69acd7/html5/thumbnails/6.jpg)
High Visibility Model-Based Analyses
IPCC Special Report on Emissions Scenarios http://www.ipcc.ch/ipccreports/sres/emission/index.htm
IEA Energy Technology Perspectives http://www.iea.org/techno/etp/index.asp
Annual Energy Outlook http://www.eia.gov/forecasts/aeo/er/
EPA Legislative Analyses http://epa.gov/climatechange/economics/economicanalyses.html
6
![Page 7: A Call for Greater Transparency in Energy Modeling and Analysisscienceag.cluster013.ovh.net/wp-content/uploads/2014/02/... · 2014. 6. 23. · Energy-economy optimization (EEO) models](https://reader035.vdocument.in/reader035/viewer/2022071611/614ad6bc12c9616cbc69acd7/html5/thumbnails/7.jpg)
Problems with the status quo
7
Inability to validate model
results
Increasing availability of
data
Moore’s Law
+
+
Increasing model
complexity
Lack of openness
+
Inability to verify model
results
Uncertainty analysis is
difficult
![Page 8: A Call for Greater Transparency in Energy Modeling and Analysisscienceag.cluster013.ovh.net/wp-content/uploads/2014/02/... · 2014. 6. 23. · Energy-economy optimization (EEO) models](https://reader035.vdocument.in/reader035/viewer/2022071611/614ad6bc12c9616cbc69acd7/html5/thumbnails/8.jpg)
Inability to validate model results
Four conditions for validatable models according to Hodges and Dewar (1992) :
• It must be possible to observe and measure the situation being modeled.
• The situation being modeled must exhibit a constancy of structure in time.
• The situation being modeled must exhibit constancy across variations in conditions not specified in the model.
• It must be possible to collect ample data with which to make predictive tests of the model.
8
![Page 9: A Call for Greater Transparency in Energy Modeling and Analysisscienceag.cluster013.ovh.net/wp-content/uploads/2014/02/... · 2014. 6. 23. · Energy-economy optimization (EEO) models](https://reader035.vdocument.in/reader035/viewer/2022071611/614ad6bc12c9616cbc69acd7/html5/thumbnails/9.jpg)
Uses of unvalidatable models
According to Hodges and Dewar (1992):
• As a bookkeeping device, to condense masses of data or to provide a means or incentive to improve data quality;
• As an aid in selling an idea of which the model is but an illustration;
• As an aid in communication, e.g., in teaching or in operating organizations
• As a vehicle to make comparisons
• As an aid in thinking and hypothesizing
To which I add:
• As a tool to identify robust outcomes across a range of tested conditions
9
![Page 10: A Call for Greater Transparency in Energy Modeling and Analysisscienceag.cluster013.ovh.net/wp-content/uploads/2014/02/... · 2014. 6. 23. · Energy-economy optimization (EEO) models](https://reader035.vdocument.in/reader035/viewer/2022071611/614ad6bc12c9616cbc69acd7/html5/thumbnails/10.jpg)
Inability to validate leads to increasing model complexity
The conventional approach to dealing with structural uncertainty is to build larger and more complex models to account for additional dynamic processes.
Little to guide the modeler and reign in efforts that do not improve model performance
Higher complexity makes it harder to address the parameter uncertainty through sensitivity and uncertainty analysis
10
![Page 11: A Call for Greater Transparency in Energy Modeling and Analysisscienceag.cluster013.ovh.net/wp-content/uploads/2014/02/... · 2014. 6. 23. · Energy-economy optimization (EEO) models](https://reader035.vdocument.in/reader035/viewer/2022071611/614ad6bc12c9616cbc69acd7/html5/thumbnails/11.jpg)
11
Past projections are generally dismal
Source: Craig et al. (2002). “What Can History Teach Us? A Retrospective Examination of Long-Term Energy Forecasts for the United States.” Ann. Rev. Energy Environ. 27:83-118.
U.S. Atomic Energy Commission forecast from 1962
Source: Morgan G, Keith D. (2008). “Improving the way we think about projecting future energy use and emissions of carbon dioxide.” Climatic Change. 90: 189-215.
![Page 12: A Call for Greater Transparency in Energy Modeling and Analysisscienceag.cluster013.ovh.net/wp-content/uploads/2014/02/... · 2014. 6. 23. · Energy-economy optimization (EEO) models](https://reader035.vdocument.in/reader035/viewer/2022071611/614ad6bc12c9616cbc69acd7/html5/thumbnails/12.jpg)
Lack of openness
Most EEO models and datasets remain closed source. Why?
• protection of intellectual property
• fear of misuse; inability to control or limit model analyses
• implicit commitment to provide support to users
• overhead associated with maintenance
• unease about subjecting code and data to public scrutiny
12
Inability to validate model
results
Increasing availability of
data
Moore’s Law
+
+
Increasing model
complexity
Lack of openness
+
Inability to verify model
results
Uncertainty analysis is difficult
![Page 13: A Call for Greater Transparency in Energy Modeling and Analysisscienceag.cluster013.ovh.net/wp-content/uploads/2014/02/... · 2014. 6. 23. · Energy-economy optimization (EEO) models](https://reader035.vdocument.in/reader035/viewer/2022071611/614ad6bc12c9616cbc69acd7/html5/thumbnails/13.jpg)
Inability to verify model results
With a few exceptions,
energy-economy models are not open source
Descriptive detail provided in model documentation and peer-reviewed journals is insufficient to reproduce a specific set of published results
Reproducibility of results is fundamental to science
Replication and verification of large scientific models can’t be achieved without source code and input data
13
Inability to validate model
results
Increasing availability of
data
Moore’s Law
+
+
Increasing model
complexity
Lack of openness
+
Inability to verify model
results
Uncertainty analysis is difficult
![Page 14: A Call for Greater Transparency in Energy Modeling and Analysisscienceag.cluster013.ovh.net/wp-content/uploads/2014/02/... · 2014. 6. 23. · Energy-economy optimization (EEO) models](https://reader035.vdocument.in/reader035/viewer/2022071611/614ad6bc12c9616cbc69acd7/html5/thumbnails/14.jpg)
Why does replication matter?
The development of repeatable experiments laid the foundation for reason-based scientific inquiry by allowing independent verification of non-intuitive empirical results
The purpose of describing experiments was to convince skeptics that a counter-intuitive result could be obtained by following a prescribed procedure.
The utility of a repeatability criterion is common to all forms of reason-based inquiry
Particularly relevant to energy- and climate-related policy analysis that could involve the transfer of huge sums of wealth.
14
![Page 15: A Call for Greater Transparency in Energy Modeling and Analysisscienceag.cluster013.ovh.net/wp-content/uploads/2014/02/... · 2014. 6. 23. · Energy-economy optimization (EEO) models](https://reader035.vdocument.in/reader035/viewer/2022071611/614ad6bc12c9616cbc69acd7/html5/thumbnails/15.jpg)
Uncertainty analysis is difficult
A common result is false precision
E.g., EPA analysis of S.2191 (Lieberman-Warner), GDP growth
predictions to 0.01%!
Large, complex models tuned to look at a few scenarios by necessity
Scenario analysis overused
Without subjective probabilities p(X|e), scenarios of little value to decision makers
15
![Page 16: A Call for Greater Transparency in Energy Modeling and Analysisscienceag.cluster013.ovh.net/wp-content/uploads/2014/02/... · 2014. 6. 23. · Energy-economy optimization (EEO) models](https://reader035.vdocument.in/reader035/viewer/2022071611/614ad6bc12c9616cbc69acd7/html5/thumbnails/16.jpg)
Cognitive heuristics play a role and can lead to misinterpretation of results.
Availability heuristic:
Probabilities of a future event or outcome assessed on the basis of how easily an individual can remember or imagine examples
Anchoring and adjustment:
People start with an initial value or “anchor” and then modify their judgment as they consider factors relevant to the specifics often insufficient adjustment
A few highly detailed scenarios can create cognitively compelling storylines.
Problems with scenario analysis
Drawn from: Morgan G, Keith D. Improving the way we think about projecting future energy use and emissions of carbon dioxide. Climatic Change 2008; 90; 189-215.
16
![Page 17: A Call for Greater Transparency in Energy Modeling and Analysisscienceag.cluster013.ovh.net/wp-content/uploads/2014/02/... · 2014. 6. 23. · Energy-economy optimization (EEO) models](https://reader035.vdocument.in/reader035/viewer/2022071611/614ad6bc12c9616cbc69acd7/html5/thumbnails/17.jpg)
Example of the availability heuristic Drawn from Slovic et al. (1976)
Tom is of high intelligence, although lacking in true creativity. He has a need for order and clarity, and for neat and tidy systems in which every detail finds its appropriate place. His writing is rather dull and mechanical, occasionally enlivened by somewhat corny puns and by flashes of imagination of the scientific type. He has a strong drive for competence. He seems to have little feel and little sympathy for other people and does not enjoy interacting with others.
17
![Page 18: A Call for Greater Transparency in Energy Modeling and Analysisscienceag.cluster013.ovh.net/wp-content/uploads/2014/02/... · 2014. 6. 23. · Energy-economy optimization (EEO) models](https://reader035.vdocument.in/reader035/viewer/2022071611/614ad6bc12c9616cbc69acd7/html5/thumbnails/18.jpg)
Which is most probable scenario for Tom?
1. Tom will select journalism as his college major.
2. Tom will select journalism as his college major but become unhappy with his choice
3. Tom will select journalism as his college major but become unhappy with his choice and switch to engineering
18
![Page 19: A Call for Greater Transparency in Energy Modeling and Analysisscienceag.cluster013.ovh.net/wp-content/uploads/2014/02/... · 2014. 6. 23. · Energy-economy optimization (EEO) models](https://reader035.vdocument.in/reader035/viewer/2022071611/614ad6bc12c9616cbc69acd7/html5/thumbnails/19.jpg)
So what should we do?
19
Recommendations drawn from DeCarolis JF, Hunter K, Sreepathi S (2012). “The case for repeatable analysis with energy economy optimization models”, Energy Economics, 34: 1845-1853.
![Page 20: A Call for Greater Transparency in Energy Modeling and Analysisscienceag.cluster013.ovh.net/wp-content/uploads/2014/02/... · 2014. 6. 23. · Energy-economy optimization (EEO) models](https://reader035.vdocument.in/reader035/viewer/2022071611/614ad6bc12c9616cbc69acd7/html5/thumbnails/20.jpg)
Recommendation 1: Make source code publicly accessible
• Ince et al. (2012): even unambiguous descriptions of computer code are no guarantee of reproducibility.
• Use software configuration management to track and control changes to software
• Utilize revision control to log changes to the codebase
Recommendations
20 Drawn from DeCarolis et al. (2012)
![Page 21: A Call for Greater Transparency in Energy Modeling and Analysisscienceag.cluster013.ovh.net/wp-content/uploads/2014/02/... · 2014. 6. 23. · Energy-economy optimization (EEO) models](https://reader035.vdocument.in/reader035/viewer/2022071611/614ad6bc12c9616cbc69acd7/html5/thumbnails/21.jpg)
Recommendation 2: Make model data publicly accessible
• Models cannot be replicated without access to the complete input data set
• Unlike source code, model input data updated frequently
• Archive input data associated with each published analysis
• Publish data in a web-accessible, free, and archived repository
Recommendations
21 Drawn from DeCarolis et al. (2012)
![Page 22: A Call for Greater Transparency in Energy Modeling and Analysisscienceag.cluster013.ovh.net/wp-content/uploads/2014/02/... · 2014. 6. 23. · Energy-economy optimization (EEO) models](https://reader035.vdocument.in/reader035/viewer/2022071611/614ad6bc12c9616cbc69acd7/html5/thumbnails/22.jpg)
Recommendation 3: Make transparency a design goal
• Documentation required to interpret model source code and data
• Documentation can come in different forms: – A standalone, comprehensive document
– Comments embedded in the source code
– Well-designed code with descriptive variable names that provide self-evident meaning
Recommendations
22 Drawn from DeCarolis et al. (2012)
![Page 23: A Call for Greater Transparency in Energy Modeling and Analysisscienceag.cluster013.ovh.net/wp-content/uploads/2014/02/... · 2014. 6. 23. · Energy-economy optimization (EEO) models](https://reader035.vdocument.in/reader035/viewer/2022071611/614ad6bc12c9616cbc69acd7/html5/thumbnails/23.jpg)
Recommendation 4: Utilize free software tools
• Use cost-free (gratis) software to minimize the barriers to entry
• Maintain a baseline model that is executable with free tools
Recommendations
23 Drawn from DeCarolis et al. (2012)
![Page 24: A Call for Greater Transparency in Energy Modeling and Analysisscienceag.cluster013.ovh.net/wp-content/uploads/2014/02/... · 2014. 6. 23. · Energy-economy optimization (EEO) models](https://reader035.vdocument.in/reader035/viewer/2022071611/614ad6bc12c9616cbc69acd7/html5/thumbnails/24.jpg)
Recommendation 5: Develop test systems for verification exercises
• Create a set of publicly available data files that represent test systems for verification exercises
• Allows modelers to debug changes to model formulation, assess computational performance, and provide a consistent evaluation mechanism for inter-model comparison
• Precedent exists: IEEE Reliability Test System (RTS) to evaluate techniques for assessing electric load reliability
Recommendations
24 Drawn from DeCarolis et al. (2012)
![Page 25: A Call for Greater Transparency in Energy Modeling and Analysisscienceag.cluster013.ovh.net/wp-content/uploads/2014/02/... · 2014. 6. 23. · Energy-economy optimization (EEO) models](https://reader035.vdocument.in/reader035/viewer/2022071611/614ad6bc12c9616cbc69acd7/html5/thumbnails/25.jpg)
Recommendation 6: Work toward interoperability of models
• Improve inter-model comparison by ensuring data consistency across models
• Reduce duplication and error in building datasets
• Use a relational database management system (RDMS): data can be queried and mapped efficiently to a project's native input format
Recommendations
25 Drawn from DeCarolis et al. (2012)
![Page 26: A Call for Greater Transparency in Energy Modeling and Analysisscienceag.cluster013.ovh.net/wp-content/uploads/2014/02/... · 2014. 6. 23. · Energy-economy optimization (EEO) models](https://reader035.vdocument.in/reader035/viewer/2022071611/614ad6bc12c9616cbc69acd7/html5/thumbnails/26.jpg)
Recommendation 7:
Build models that are as simple as possible to answer the question at hand, and then rigorously exercise it with uncertainty analysis
Recommendations
26
![Page 27: A Call for Greater Transparency in Energy Modeling and Analysisscienceag.cluster013.ovh.net/wp-content/uploads/2014/02/... · 2014. 6. 23. · Energy-economy optimization (EEO) models](https://reader035.vdocument.in/reader035/viewer/2022071611/614ad6bc12c9616cbc69acd7/html5/thumbnails/27.jpg)
What we did.
27
![Page 28: A Call for Greater Transparency in Energy Modeling and Analysisscienceag.cluster013.ovh.net/wp-content/uploads/2014/02/... · 2014. 6. 23. · Energy-economy optimization (EEO) models](https://reader035.vdocument.in/reader035/viewer/2022071611/614ad6bc12c9616cbc69acd7/html5/thumbnails/28.jpg)
Tools for Energy Model Optimization and Analysis
Temoa also means “to seek something” in the Nahuatl (Aztec) language:
Taken from: An analytical dictionary of Nahuatl by Frances E. Karttunen
28
Temoa
![Page 29: A Call for Greater Transparency in Energy Modeling and Analysisscienceag.cluster013.ovh.net/wp-content/uploads/2014/02/... · 2014. 6. 23. · Energy-economy optimization (EEO) models](https://reader035.vdocument.in/reader035/viewer/2022071611/614ad6bc12c9616cbc69acd7/html5/thumbnails/29.jpg)
Temoa goals and approach
Goal: Create an open source, technology explicit EEO model
Our Approach: • Public accessible source code and data • No commercial software dependencies • Data and code stored in a web accessible electronic repository • A version control system • Programming environment with links to linear, mixed integer,
and non-linear solvers • Built-in capability for sensitivity and uncertainty analysis • Utilize multi-core and compute cluster environments • Input and output data managed directly with a relational DB*
29
![Page 30: A Call for Greater Transparency in Energy Modeling and Analysisscienceag.cluster013.ovh.net/wp-content/uploads/2014/02/... · 2014. 6. 23. · Energy-economy optimization (EEO) models](https://reader035.vdocument.in/reader035/viewer/2022071611/614ad6bc12c9616cbc69acd7/html5/thumbnails/30.jpg)
Technology explicit modeling
30
Capital Cost ($M/PJ/yr) Fixed O&M ($M/PJ) Variable O&M ($M/PJ) Capacity factor Efficiency Emissions coefficient (kton/PJ)
Objective function: minimize present cost of energy supply over a defined time horizon Decision variables: activity (PJ) and capacity (PJ/yr) for each technology
‘Utopia’ (18 technologies included)
![Page 31: A Call for Greater Transparency in Energy Modeling and Analysisscienceag.cluster013.ovh.net/wp-content/uploads/2014/02/... · 2014. 6. 23. · Energy-economy optimization (EEO) models](https://reader035.vdocument.in/reader035/viewer/2022071611/614ad6bc12c9616cbc69acd7/html5/thumbnails/31.jpg)
TEMOA Model Features
A technology explicit model with perfect foresight, similar to the TIMES model generator.
• Flexible time slicing by season and time-of-day
• Variable length model time periods
• Technology vintaging
• Separate technology loan periods and lifetimes
• Global and technology-specific discount rates
31
![Page 32: A Call for Greater Transparency in Energy Modeling and Analysisscienceag.cluster013.ovh.net/wp-content/uploads/2014/02/... · 2014. 6. 23. · Energy-economy optimization (EEO) models](https://reader035.vdocument.in/reader035/viewer/2022071611/614ad6bc12c9616cbc69acd7/html5/thumbnails/32.jpg)
‘Utopia’ verification exercise
32
0
50
100
150
Pe
tajo
ule
s / Y
ea
r
0
0.5
1
1.5
2
Gig
aw
atts
1990 2000 20100
20
40
60
Pe
tajo
ule
s / Y
ea
r
IMPDSL1
IMPGSL1 IMPHCO1
IMPDSL1
IMPGSL1
IMPHCO1
IMPDSL1
IMPGSL1
IMPHCO1
E01
E01
E31 E31 E31
E51 E51 E51
E70 E70 E70
E01
RHE
RHE RHE
RHORHO
RHO
RL1RL1
RL1
TXD TXD TXDTXG TXG TXG
Import Technologies
Electric Generators
Demand Devices
MARKAL: Black Temoa: White
![Page 33: A Call for Greater Transparency in Energy Modeling and Analysisscienceag.cluster013.ovh.net/wp-content/uploads/2014/02/... · 2014. 6. 23. · Energy-economy optimization (EEO) models](https://reader035.vdocument.in/reader035/viewer/2022071611/614ad6bc12c9616cbc69acd7/html5/thumbnails/33.jpg)
Approach to uncertainty analysis.
33
![Page 34: A Call for Greater Transparency in Energy Modeling and Analysisscienceag.cluster013.ovh.net/wp-content/uploads/2014/02/... · 2014. 6. 23. · Energy-economy optimization (EEO) models](https://reader035.vdocument.in/reader035/viewer/2022071611/614ad6bc12c9616cbc69acd7/html5/thumbnails/34.jpg)
Approach to uncertainty analysis
Use the following techniques in series:
Sensitivity analysis and Monte Carlo simulation
→ Determine key sensitivities
Multi-stage stochastic optimization
→ Develop a hedging strategy
Explore near-optimal, feasible region
(Modeling-to-Generate-Alternatives)
→ Test robustness of hedging strategy
34
![Page 35: A Call for Greater Transparency in Energy Modeling and Analysisscienceag.cluster013.ovh.net/wp-content/uploads/2014/02/... · 2014. 6. 23. · Energy-economy optimization (EEO) models](https://reader035.vdocument.in/reader035/viewer/2022071611/614ad6bc12c9616cbc69acd7/html5/thumbnails/35.jpg)
35
Stochastic Optimization
Decision-makers need to make choices before uncertainty is resolved → requires an “act then learn” approach
Need to make short-term choices that hedge against future risk → Sequential decision-making process that allows recourse Stochastic optimization • Build a scenario tree • Assign probabilities to future outcomes • Optimize over all possibilities
![Page 36: A Call for Greater Transparency in Energy Modeling and Analysisscienceag.cluster013.ovh.net/wp-content/uploads/2014/02/... · 2014. 6. 23. · Energy-economy optimization (EEO) models](https://reader035.vdocument.in/reader035/viewer/2022071611/614ad6bc12c9616cbc69acd7/html5/thumbnails/36.jpg)
36
Simple example of stochastic optimization
Suppose we have two technologies, A and B. Let x and y represent the installed capacity in Stages 1 and 2, respectively.
t1 t2
s1
s2
Stage 1 Decision Variables: Stage 2 Decision Variables:
,A B
x x
1 1
2 2
, ,
, ,
,
,
A s B s
A s B s
y y
y y
1p
2p
1
M in im iz e : c
N
T T
s s s
s
x p d y
S u b je c t T o :
1, . . . ,
0
0 1, . . . ,
s s s s
s
A x b
T x W y h fo r s N
x
y fo r s N
Scenario 1: s1
Scenario 2: s2
![Page 37: A Call for Greater Transparency in Energy Modeling and Analysisscienceag.cluster013.ovh.net/wp-content/uploads/2014/02/... · 2014. 6. 23. · Energy-economy optimization (EEO) models](https://reader035.vdocument.in/reader035/viewer/2022071611/614ad6bc12c9616cbc69acd7/html5/thumbnails/37.jpg)
What about structural uncertainty?
37
Consider an optimization model that only includes Objective 1 and leaves Objective 2 unmodeled. The true optimum is within the feasible, suboptimal region of the model’s solution space.
Viable alterative solutions exist within the model’s feasible region.
Example adopted from Brill et al. (1990).
Objective 1
Ob
ject
ive
2
Non-inferior frontier
![Page 38: A Call for Greater Transparency in Energy Modeling and Analysisscienceag.cluster013.ovh.net/wp-content/uploads/2014/02/... · 2014. 6. 23. · Energy-economy optimization (EEO) models](https://reader035.vdocument.in/reader035/viewer/2022071611/614ad6bc12c9616cbc69acd7/html5/thumbnails/38.jpg)
Modeling to Generate Alternatives
38
Still haven’t dealt with structural uncertainty in the model
Need a method to explore an optimization model’s feasible region → “Modeling to Generate Alternatives”†
MGA generates alternative solutions that are maximally different in decision space but perform well with respect to modeled objectives
The resultant MGA solutions provide modelers and decision-makers with a set of alternatives for further evaluation
†Brill (1979), Brill et al. (1982), Brill et al. (1990)
![Page 39: A Call for Greater Transparency in Energy Modeling and Analysisscienceag.cluster013.ovh.net/wp-content/uploads/2014/02/... · 2014. 6. 23. · Energy-economy optimization (EEO) models](https://reader035.vdocument.in/reader035/viewer/2022071611/614ad6bc12c9616cbc69acd7/html5/thumbnails/39.jpg)
Hop-Skip-Jump (HSJ) MGA
39
Steps:
1. Obtain an initial optimal solution by any method
2. Add a user-specified amount of slack to the value of the objective function
3. Encode the adjusted objection function value as an additional upper bound constraint
4. Formulate a new objective function that minimizes the decision variables that appeared in the previous solutions
5. Iterate the re-formulated optimization
6. Terminate the MGA procedure when no significant changes to decision variables are observed in the solutions
Brill et al. (1982)
![Page 40: A Call for Greater Transparency in Energy Modeling and Analysisscienceag.cluster013.ovh.net/wp-content/uploads/2014/02/... · 2014. 6. 23. · Energy-economy optimization (EEO) models](https://reader035.vdocument.in/reader035/viewer/2022071611/614ad6bc12c9616cbc69acd7/html5/thumbnails/40.jpg)
Sample Result
Drawn from DeCarolis (2011) 40
![Page 41: A Call for Greater Transparency in Energy Modeling and Analysisscienceag.cluster013.ovh.net/wp-content/uploads/2014/02/... · 2014. 6. 23. · Energy-economy optimization (EEO) models](https://reader035.vdocument.in/reader035/viewer/2022071611/614ad6bc12c9616cbc69acd7/html5/thumbnails/41.jpg)
Conclusions
Most EEO models and model-based analyses are opaque to external parties
A prerequisite for making energy scenarios more scientific is open access to model source code and data enables repeatability
Rigorous uncertainty analysis is necessary to develop useful and actionable insight for policy
Challenge for modeling community to archive model, data and results in a systematic and transparent fashion.
41
![Page 42: A Call for Greater Transparency in Energy Modeling and Analysisscienceag.cluster013.ovh.net/wp-content/uploads/2014/02/... · 2014. 6. 23. · Energy-economy optimization (EEO) models](https://reader035.vdocument.in/reader035/viewer/2022071611/614ad6bc12c9616cbc69acd7/html5/thumbnails/42.jpg)
Relevant papers (published or submitted)
DeCarolis J.F. (2011). Using modeling to generate alternatives (MGA) to expand our thinking on energy futures. Energy Economics, 33: 145-152.
Howells M., Rogner H., Strachan N., Heaps C., Huntington H., Kypreos
S., Hughes A., Silveira S., DeCarolis J., Bazillian M., Roehrl A. (2011). OSeMOSYS: The open source energy modeling system: An introduction to its ethos, structure and development. Energy Policy, 39(10), 5850-5870.
DeCarolis J.F., Hunter K., Sreepathi S. (2012). The Case for Repeatable
Analysis with Energy Economy Optimization Models. Energy Economics, 34: 1845-1853.
Hunter K., Sreepathi S., DeCarolis J.F. (2013). Tools for Energy Model
Optimization and Analysis. Energy Economics, 40: 339-349.
42
![Page 43: A Call for Greater Transparency in Energy Modeling and Analysisscienceag.cluster013.ovh.net/wp-content/uploads/2014/02/... · 2014. 6. 23. · Energy-economy optimization (EEO) models](https://reader035.vdocument.in/reader035/viewer/2022071611/614ad6bc12c9616cbc69acd7/html5/thumbnails/43.jpg)
Acknowledgments
• Kevin Hunter, MS student, Civil Engineering, NCSU
• Sarat Sreepathi, PhD student, Computer Science, NCSU
• Jean-Paul Watson and Bill Hart, Sandia National Laboratory
This work would made possible through the generous support of the National Science Foundation. CAREER: Modeling for Insights with an Open Source Energy Economy Optimization Model. Award #1055622
43
![Page 44: A Call for Greater Transparency in Energy Modeling and Analysisscienceag.cluster013.ovh.net/wp-content/uploads/2014/02/... · 2014. 6. 23. · Energy-economy optimization (EEO) models](https://reader035.vdocument.in/reader035/viewer/2022071611/614ad6bc12c9616cbc69acd7/html5/thumbnails/44.jpg)
44