 |
Inaugural Workshop
Breakout Sessions
This section summarizes the activities resulting from the topic-specific breakout sessions, including the goals, current related activities, next steps identified, the lead and the other participants. Additional details on topic activities can be accessed by linking on the session title below.
Session attendees:
David Chien (DOT)
Karlynn Cory (NREL)
Steve Dunn (EPA)
Allen Fawcett (EPA)
Jeff Harris (LBNL)
Susan Holte (DOE-EIA)
Michael Leifman (DOE-EERE)
Chris Namovicz (DOE-EIA)
Andrew Nicholls (PNNL)
Phil Patterson (DOE-EERE)
Walter Short (NREL)
Alan Snyder (DOT)
Maria Vargas (NETL)
Carol White (FERC)
Frances Wood (OnLocation)
Topic clarification:
- Which models? All models — including NEMS.
- There are several needs to improve these models.
Important Decision Makers Identified:
- Federal government decision makers (what benefits from federal programs?),
- State public utility commissions (costs and benefits from technologies, programs, real-time pricing),
- Manufacturers,
- Utilities (as buyers, demand-side management funders),
- Legislators (federal and state),
- Policy makers on climate change policies,
- Regional energy planning organizations,
- Program managers (federal, state, utilities),
- Industry and trade groups.
Data needs for improving technology and demand response in energy models
- Policy makers may not be asking the questions to which they really want the answers — just those they think they want answered (not necessarily the same, or complete).
- Different decision makers have different information needs (what is important to whom?).
- Many times the front-end assumptions directly impact the range of potential outcomes; can this impact be quantified so modelers focus on improving the most important ones?
- Very little recent work on price elasticities; this impacts all of these models substantially.
- Most work done in the late 1970s, except Goldman et al., demand response or DR, some CA pilots.
- David Greene (ORNL) has been keeping up transport-sector elasticity studies.
- For DR — need more work on how enabling technologies impact elasticities
- E.g., smart meters, smart thermostats, wireless/Web-based controls
- Technology changes over time
- Technology learning, economies of scale; (separate from R&D),
- Learning in technology cost (from first version to commercialization),
- What were the market dynamics and actors? (e.g., steel prices, fossil prices),
- How might things have been done differently, in retrospect — especially for policy interventions — and how does this inform future policy? (include some faster and some slower penetrating technologies)
Performing energy modeling and other analysis
- Better modeling of trade-offs between technologies and the subsequent impacts on price.
- Better "what-if" modeling is needed, to help consider policy questions
- EERE VISION model for transportation is pretty good at this.
- Scenario analysis — how can I achieve a specific goal, no matter what market does?
- What is the potential for a new, emerging technology? (Market share, impact if adapted)
- Are energy efficiency (EE), DR, and renewable energy (RE) truly modeled accurately in models? At what point should they be introduced.
- Currently, DOE-EERE does a lot of work on technology characterizations of energy efficiency and renewables; NETL has characteristics of fossil technologies.
- We need to better quantify the impact of adding/removing budget from a program.
- Buildings need better envelope modeling and cumulative impact of several applications.
- David Greene (ORNL) has been keeping up transport-sector elasticity studies; what about other sectors?
- Most models are usually weak at incorporating economic feedbacks (usually exogenous)
- Better interconnection of sectors and their impact on economics and economy.
- Models don't deal well with transients
- Demand-management, demand-response, and grid reliability value.
- Demand-side aspects of systems with minimum slack and redundancy (gas deliveries).
- Demand side in short-term models (underlying assumption that demand is inelastic in the short-term; so EE takes time to influence).
- Look more at consumer response to rate structure, not just rate levels.
- How have first costs of specific technologies changed over ~20 years, historically. Can this information be used to inform future modeling?
Collaboration process
- Stanford Modeling Forum has been active.
- There is a need for collaboration to identify cross-market impacts
- Technologies introduced in one sector may impact other sectors (e.g., fuel cells).
- Advances (or failures) in one market segment may affect progress in another.
- Increase transparency and user-friendliness: need models that decision makers and their staff can play with and understand, trust.
- Modelers need to better identify what matters to decision makers
- Which factors give decision makers the greatest leverage and what are the potential impacts of taking action on them?
- How to reduce fossil fuel use at the lowest cost?
- Look across state/regions (or even across sectors?) at "natural experiments" from which we can gain useful insights — for modeling and for policy/program decisions.
- EPA is having a forum in late 2006 on reflecting technology changes in modeling (Pete Wilcox, Alan Sanstad, others...).
Activities decided by the group:
- Technology cost and performance evolution — Improve understanding, from looking historically at actual (EE and RE?) examples, past rates of technology change (cost, performance) and the underlying dynamics and drivers, at the emergent, precommercial, and fully commercial stages. Incorporate these empirical findings into technology maturity and saturation curves in models. Also impact of global markets.
- Potential for new/emerging technologies (EE and RE only?)
- need careful terminology; people can mean very different things (risk, time-frame) when referring to energy technologies,
- by definition, because it's "new/emerging," there's uncertainty in characterizing cost and performance, how these will evolve, or what market prospects are,
- also need to consider substitutes and competitors,
- need some good horizon-scanning (including outside the U.S.).
- New empirical work on elasticities
- Distinguish short-term and long-term elasticities,
- For DR, need more work on how enabling technologies affect elasticities (e.g., smart meters, smart thermostats, wireless Web-enabled controls),
- This illustrates the point that an elasticity is an observation, not an explanation!
- For each of the above: the first priority is improved data, not improved specification in the model (but need to consider from the start how the model will use new data),
- Most work on improving technology characterization in models has been done outside the US (for climate/mitigation modeling); Stanford Modeling Forum has also been active.
- For collaboration potential, ask two questions:
- Who else does modeling? (e.g., some states, for electricity; U.S. and international groups for climate change)
- Who is a current (or potential) source of technology data? (not necessarily for modeling purposes — could be others)
|
|
