A new approach to building and using global TIMES models

1. A new approach to Global energy systems modeling
Knowledge based investigation of energy system scenarios
Kinesis (noun): movement in a cell or organism in response to an external stimulus

2. Outline
● The need for a new approach to model development
● Knowledge-based modeling
● Model design and deployment
● Examples

3. Motivation
● The need for energy system modeling has never been greater
○ Rapid transitions across the energy system
○ Cross-sector interactions are crucial to system change
○ Urgent short-term needs in long-term context
○ Multiple intersecting uncertainties
● We need to be able to envision, explore, and communicate about
disruptive scenarios - futures that look nothing like the past
● These are precisely the capabilities of TIMES…
● So why isn’t everyone using TIMES?

4. How hard is it to use a TIMES model?
● Takes a long time to build
● Enhancements and updates need a lot of effort
● Difficult for even other TIMES modelers to understand
● Practically inaccessible to non-TIMES modelers
● In other words, effective model users are those who have built those models
themselves. There is no clear path to become a good model user, without
becoming a “TIMES expert” first.

5. Broadening TIMES use and decision relevance
● The core modeling team can no longer monopolize the process of
extracting insights from model results
○ Wider engagement brings diverse perspectives
○ Stakeholders use the insights to guide actions much more efficiently if they have
the option to engage with models directly
○ Of course, without investing the time it takes to become a “TIMES expert”

6. Rethinking model aggregation and data structures
● Many questions need global models with different regional focus
○ Global hydrogen scenarios
○ LNG supply, demand, and trade
○ Energy security in Europe-Central Asia region
● A single model with static regional aggregation is not the best way to address
these questions
● More temporal and technology flexibility would also be convenient

7. Knowledge-based energy system modeling
● Typically models are built by first deciding on a regional structure, then seeking or
aggregating/disaggregating data to fit that structure
○ Changing regional structure later can be a major undertaking
○ And similarly with structuring model data in time
● In a knowledge-based system, data is stored in a relational database at the most
granular level available
○ Country or state/province level
○ Individual plant/unit level
○ Hourly
● Database queries are designed to aggregate/disaggregate data in the desired
structure, creating inputs for a model instance
● Multiple model instances with different time/space/technology aggregations can
be developed readily from the same knowledge database

8. KINESYS - Model development and deployment
● Knowledge-based model framework built upon the former TIAM global model
○ Flexible aggregation allows rapid reconfiguration of model regions/technologies/timeslices to
suit analysis needs
○ Also supports creation of country “starter” models
● Deployed in the VEDA-Online framework
○ Engage stakeholders directly in results review
○ Participants can learn as much or as little about the underlying model details as they choose
while doing meaningful analysis
● Updates to the framework are available to improve all KINESYS models
Knowledge based investigation of energy system scenarios

9. KINESYS knowledgebase major data sources
Component Source Use
Base year characterization IEA Country Energy Balances Base year fuel production, transformation, and sectoral consumption
Oil and gas supply RYSTAD* Country-level oil and gas supply curves
Gas supply and trade Global Energy Monitor Gas pipelines and LNG liquefaction/regasification terminals – by country
Gas supply and trade GIIGNL LNG liquefaction and trade contracts by country
RE Supply REZoning (ESMAP) Solar, Wind onshore, and Wind offshore
RE Supply EIA Country level hydro potential
RE Supply EU PV GIS Solar and wind shapes
RE Supply IIASA GLOBIOM model Country level biomass potential
Transformation SPG Power Plant Database Country level power plant capacity (and known builds) by type and vintage
Transformation IRENA Country level existing RE capacity and generation
Transformation and trade World Energy Outlook 2022 New power plant technology library
Demand devices and hydrogen E3 Modeling Transportation vehicles and hydrogen technology libraries
Demand devices US EPA TIMES model Building and industry future technology libraries
Demand devices EU PV GIS Country-level HDD/CDD for device CFs and heat/cool load shapes
End use demand projections IPCC SSP scenarios Projections for population and GDP to drive end use demands

10. Knowledgebase data operations
● Aggregation
● Disaggregation based upon assigned weighting variables
● Assignment of default data
● Assignment of missing data from model regions/technologies

11. Several regional aggregations have been created
https://vedaonline.cloud/kanors/kinesys.html

12. Deploying KINESYS models in VEDA-Online
1. Build confidence by verifying calibration and comparing results to external
scenarios
2. Begin analysis by reviewing results from a suite of scenarios that stress the
model, examining fuel and technology switches across a spectrum of
abatement levels and pathways
3. Engage stakeholders in using results review to prioritize input refinements
and policy scenario development

13. Example: Gas infrastructure disruption

14. Western Europe in 2023

15. Western Europe in 2023 – Sankey Diff view

16. Example: Hydrogen trade – Imports into Japan

17. Example: Country starter models
● KINESYS was used to create country starter
models for Egypt and Indonesia within the
Net Zero World initiative
● The countries were broken out as individual
regions and could be run within the global
model or standalone with global fuel market
prices taken from the global model
● Used to develop and explore Net Zero
pathways in Indonesia and assess Egypt’s
competitiveness for hydrogen exports to
Europe under global Net Zero scenarios

18. Example: Mark-to-model valuation
● Conducted in the Net Zero
World Indonesia model
● Analysis takes the perspective of
an investor contemplating
investment in power plant,
subject to future carbon price
risk
● Investment in each project is
forced and lifetime discounted
cash flows are computed
● ROI and other investment
valuation measures can be
computed
-1.00
-0.80
-0.60
-0.40
-0.20
0.00
0.20
0.40
0.60
0.80
0.05
0.06
0.07
0.08
0.09
0.1
0.11
0.12
0.13
0.14
0.15
0.16
0.17
0.18
0.19
0.2
0.21
0.22
0.23
0.24
0.25
0.26
0.27
0.28
0.29
0.3
0.31
0.32
0.33
0.34
0.35
0.36
0.37
0.38
0.39
0.4
0.41
0.42
0.43
0.44
0.45
0.46
0.47
0.48
0.49
0.5
0.51
0.52
0.53
0.54
0.55
0.56
0.57
0.58
0.59
0.6
0.61
0.62
0.63
0.64
0.65
0.66
0.67
0.68
0.69
0.7
0.71
0.72
0.73
0.74
0.75
0.76
0.77
0.78
0.79
0.8
0.81
0.82
0.83
0.84
0.85
0.86
0.87
0.88
0.89
0.9
0.91
0.92
0.93
0.94
0.95
SCALED
NET
PRESENT
VALUE
Scaled Return on Investment versus Carbon Price Scenario
Solar Coal Gas Coal (CCS) Gas (CCS)

19. Example: Indonesia country model

20. Example: Indonesia country model

21. Final thoughts
● The combination of a knowledge based approach to model
development and deployment in VEDA-Online helps us quickly get to
the point where we are:
○ Reviewing results with model stakeholders
○ Designing and answering what-if questions with them
○ Collaboratively generating insights on questions of real interest to them