Essay Three Ideas - Relationship between Distributed ...

Essay Three Ideas - Relationship between Distributed ...

Work in Progress Returns From Distributed Generation: Utilities, Consumers, and Communities Arijit Sen, PhD Candidate School of Public Policy, University of Maryland US AEE 2019 RESEARCH CONFERENCE PRESENTATION NOVEMBER 4 T H , 2019 Research Question What is the value of distributed solar generation at the consumer level and how does that compare to utility valuations of solar and the returns at the community level? This is an alternate research question designed in the context of analyzing utility profitability and distributed generation adoption

Original Research Question What changes to tariff structure, utility incentive structure, utility valuation of distributed solar, and distributed PV ownership pattern are required in order to maximize residential level adoption of distributed solar adoption without adversely impacting the utility profitability? Customer owned vs Utility promoted community solar Max Distributed PV adoption (tariff, utility valuation of DPV, DPV ownership pattern) w.r.t Utility Profitability (Utility valuation of DPV, tariff, DPV ownership pattern, incentives) = Baseline level. More rate of return for investing in DPV than building T&D resources (New York has done this) The DPV- Utility Dynamics Purported Utility Death Spiral (lot of literature, the effect is exaggerated in theory but somewhat present)

Satchwell 2015 DPV Adoption Increases Utility profitability is affected Laws 2017 OR DPV Adoption Decreases Dargouth 2016 Utility raises tariff for DPV adopters (restructuring or demand/fixed charge) Costello 2014

Utility raises tariff for all The literature does not tackle The entire system dynamic together at once Any probable ways to break the dynamics DPV Adoption Increases Utility profitability is affected OR DPV Adoption Decreases Utility raises tariff for DPV adopters (restructuring or demand/fixed charge)

Utility raises tariff for all Methods and Data Requirement Set up a residential model (easy enough at an unit level, can be reasonably easily converted to aggregate level through appropriate assumptions, e.g. a Bass Curve for adoption) Tool NRELs Dgen can do this on an aggregate right off the bat, otherwise Homer can be used and scaled up [done in literature by Janko in 2016 to analyze implications of DPV for ratepayers and utilities. Very simplistic utility model used.] Set up a reasonably complex utility model where DPV actively affects utility calculations of revenue and cost (not just fuel cost but also associated costs such as T&D), so an accurate Value of Solar for utility can be constructed Tool Ideally something like FINDER from LBNL or even a version of something that is used by regulatory commissions to evaluate VoS for various states (California, Nevada etc.) The issue is that even though a stylized utility structure can be set up (Example next slide), there is lack of hard data that fits the assumptions used for the residential model. There is utility VoS data at the state level (Example next slide), but we have to take them as is.

Making the system dynamic. A static system on both ends (residential and utility) is not that hard to set up, the challenge is to make them interact with each other and the results in the next period affect that interaction FINDER does dynamic modelling on the utility side, taking the residential side as given. Dgen does dynamic modelling on the residential side, feeding the data into ReEDS (which has a very primitive utility model) and getting that data back on the residential side. Value of Solar structure for utility Net Avoided cost components Details on calculation Energy/Fuel Fuel cost avoided due to extra RE, but added backup costs

Capacity Capacity cost avoided due to extra RE, but potential extra capacity costs for stability T&D T&D capacity/O&M cost avoided, but may require additional T&D capacity Losses Losses avoided due to DG

Ancillary Services Changes in demand of AS such as voltage regluation Environment CO2 cost avoided RPS purchase RPS purchase cost avoided Table Source: Based on E3s Avoided Cost Model that has 2016 California data as an example Additional complexity Tariff

structure The Avoided-cost style VoS model takes tariff structure as given. Tariff structure (which affects BOTH utility and residential consumers) modelling required to capture the dynamic appropriately. The basic structures are not that hard to model (Homer has several structures built in, and they can be copied over pretty easily) However it is difficult to model potential structure switching by utility to maintain profitability. Even in tools like Dgen and FINDER you have to exogenously give it a particular structure. Additional complexity incentives and ownerships Literature hasnt really looked at the relationship between incentives and DPV adoption (no detailed results out yet for New York BQDM, there exists a theoretical spreadsheet and some overarching results), or ownership structure and DPV adoption To model rate of return, we need to build a proper revenue/cost structure which has more details than the avoided cost structure

Again it can be done (Looking at regulatory filings for example), but hard numbers especially projections are difficult to come by Ownership can probably be modelled in a similar fashion. The BQDM has ownership built in (not just DPV, but DR and EE assets as well) for utilities in terms of MW. But apart from ConEd, no other reliable numbers exist. In summary for the original research question Probably too hard to answer given methodological and data challenges On the Alternate Research Question It is answerable, but is extremely derivative, and largely follows research that has been well established, and answers the question based on the results of that research rather than doing anything groundbreaking Technically its still a contribution because no one has looked at all the things together that I

want to look at. However, its fit with the overall dissertation theme (Classifying outcomes of actors in their efforts to reduce GHG emissions) is tenuous at best. Alternate Research Question How does the returns for the utility from DPV adoption compare with the returns from the consumer from DPV adoption and the societal returns from DPV adoption? Literature has tackled the last two parts in a static framework (using current data from Tracking the Sun DPV database) - Vaishnav 2017 There does exist some VoS evaluation that are static as well. The idea is to combine the two (Ill run my own evaluation for Consumer/ Societal returns using a more dynamic framework of HOMER and not just use Vaishnavs results) The comparison will give states that have returns that are ranked differently, the idea is to figure out the details behind those differences. Ideally, they should have roughly similar returns/ ranking should not be different, i.e. no one should be significantly better off from DPV than the other, OR if there is a natural order of beneficiaries then that

should be consistent. Methods Collect VoS data from state or utility-level studies. 32 such studies exist, excluding generic VoS studies and older studies for the same state. Design candidate PV systems based on appropriate load and existing tariff structure in HOMER to determine LC of PV systems and compensation paid to those systems. The optimal systems are chosen on the basis of fastest payback time. Create ranges for VoS, compensation, and LC of PV systems. For VoS use generic studies to determine possible ranges of certain subcomponents. For compensation, test various tariff structures. For LC of PV systems, vary incentives. What does this tell us in terms of comparative valuation of solar by different entities? Preliminary Results Comparative Value Anlaysis (Levelized cost $/kWh)

TX OR NY MN HI NC CA AZ 0 0.05 0.1 0.15 Value of Solar

0.2 Average Compensation 0.25 LC of PV systems w. Current Incentives 0.3 0.35 0.4 Likely factors of classification Comparison on the basis of differences between VoS and compensation; differences

between VoS and LCOE; differences between compensation and LCOE. Hawaii is an outlier because of its very expensive electricity and relatively inflexible tariffs California typically has similar values for all three because its tariff structure Texas tends to undervalue solar and has low compensation due to fairly cheap electricity but doesnt have a lot of incentives for distributed PV Customer penetration is a potential factor as well, as that would change VoS calculations and if VoS isnt updated accordingly, then increased customer penetration will see increased retail rates. This may be problematic for places with high utility costs and low rate of returns. Policy Actions Ensure that compensation and VoS are equal and are updated regularly with changes to tariff structures, penetration, utility costs and other factors Incentives are helpful to ensure that distributed generation is cost competitive, but they must be weighed against public benefits (generally environmental)

In order to be reasonably certain about the DG load and the utility valuation/ compensation scheme design a program where a certain percentage of the DG load is aggregated via community solar or similar schemes and long-term contracts for valuation and compensations exist. Future Research Complete outcomes and classifications for all possible states Improve customer side design by closely matching the utility VoS assumptions (load level, consumer mix, discount rates etc.) This is harder than it sounds due to lack of transparent assumptions for some VoS calculations Complete sensitivity analysis w.r.t. VoS, compensation and LC of PV systems Design stylized model to incorporate the effect of changing customer penetration Without a dynamic effect, probably difficult because theres no way to directly affect the VoS calculation.

Recently Viewed Presentations

  • IV. Boundaries

    IV. Boundaries

    A subsequent boundary divides an ethnically homogenous group Irredentism- irredentism tries to justify its territorial claims on the basis of (real or imagined) historic or ethnic affiliations. It is often advocated by nationalist movements and has been a feature of...
  • Characteristics of the Medieval Romance

    Characteristics of the Medieval Romance

    Perceval is a young man whose mother raised him in isolation after losing her husband and two other sons to chivalric combat. One day, Perceval meets a group of knights, who so impress him that he decides to become a...
  • An Evaluative Study of Clinical Preceptorship in the Ontario ...

    An Evaluative Study of Clinical Preceptorship in the Ontario ...

    A few preceptees described time that could have been used better or having too much time "My preceptor was probably not intentionally trying to give me a limited experience, but should have declined the opportunity to be my preceptor because...
  • Reptile Wild Discover Zone

    Reptile Wild Discover Zone

    Reptile Wild Discover Zone Objective: Common Characteristics of Reptiles (slides 4- 21) A variety of body shapes that are covered with scales and plates for protection against desiccation and injury.. Most have
  • Implementing technology for clinical assessment: opportunities and challenges

    Implementing technology for clinical assessment: opportunities and challenges

    Henry Brodaty - Centre for Healthy Brain Aging, Sydney. Deep and Frequent Phenotyping. Challenges (Valley of Death) Ensuring that the academic aspirations of all collaborative teams are translated into the research objectives.
  • EDUC 6105 Organizations, Innovation, and Change Instructional Design

    EDUC 6105 Organizations, Innovation, and Change Instructional Design

    Ikeda has founded a number of educational and cultural institutions that seek to foster the values of peace, culture and education. Over time, tensions grew between SGI President Ikeda and the Nichiren Shoshu priesthood over questions of doctrine and leadership....
  • CH 9: Stereochemistry Renee Y. Becker CHM 2210

    CH 9: Stereochemistry Renee Y. Becker CHM 2210

    * Chromatographic Resolution of Enantiomers * Prochirality A molecule that is achiral but that can become chiral by a single alteration is a prochiral molecule * Chirality in Nature and Chiral Environments Stereoisomers are readily distinguished by chiral receptors in...
  • Section 5.2 Family Problems Family Problems Objectives List

    Section 5.2 Family Problems Family Problems Objectives List

    Financial Problems Financial problems can be less stressful if family members work together to improve the situation. Slide * of 19 Slide * of 19 Slide * of 19 A separation is an arrangement in which spouses live apart and...