€¦ · web view2017. 2. 21. · much capacity of integrated distributed energy resources may be...

California Distribution Resources Plan Integrated Capacity Analysis Working Group

DRAFT Final Report – Due March 1, 2017

NOTE: This is a first draft of the ICA WG report. Comments in tracked changes can be sent to [email protected].

Table of Contents

ContentsTable of Contents....................................................................................................................................1

Acronyms.................................................................................................................................................1

Executive Summary.................................................................................................................................1

Background..............................................................................................................................................1

Scope and Organization of the Report.....................................................................................................5

Discussion Topics, Points of Consensus, Proposed Alternatives if Any, Recommendations, Next Steps. 6

Use Cases.............................................................................................................................................6

ACR Requirements – ICA Modeling and Input Assumptions..............................................................12

Short-term activities of the ICA WG...................................................................................................16

Long-term refinement activities with potential for inclusion into initial ICA methodology...............25

Next Steps for the ICA WG.................................................................................................................26

Appendix...............................................................................................................................................28

AcronymsTBD

Executive SummaryThis is the Final Report of the Integrated Capacity Analysis Working Group to the California Public Utilities Commission. The Report summarizes the development of the Integrated Capacity Analysis to date and discusses timeline considerations for implementation across entire IOU service territories, recommendations on how to improve the methodology going forward and how the results may be used to inform decision-making on the part of the Commission, utilities, providers of distributed energy resources, and customers.

BackgroundAdopted in 2014, Section 769 of the California Public Utilities Code requires the IOUs to prepare a distribution resources plan which identifies optimal locations for the deployment of distributed energy resources. In August 2014, the California Public Utilities Commission (CPUC) began implementation of this requirement through Rulemaking (R.) 14-08-013, the Distribution Resources Plan (DRP) proceeding. A Ruling from the Assigned Commissioner in November 2014 introduced the Integration Capacity Analysis (ICA) as a tool which would support the determination of optimal locations by specifying how

1

mailto:[email protected]




much capacity of integrated distributed energy resources may be available on the distribution network at the circuit level.1 Pursuant Commission direction, California’s Investor Owned Utilities (IOUs) filed Applications2 developing their Distribution Resource Plans, including a proposal to complete a Demonstration of their proposed ICA (“Demo A”). Stakeholder holder input on the IOU proposals was gathered, leading to further guidance issued in May 2016. That guidance authorized a demonstration project of the ICA, requiring the Investor Owned Utilities (IOUs) to meet the following nine functional requirements:

1. Quantify the Capability of the Distribution System to Host DER 2. Common Methodology Across All IOUs 3. Analyze Different Types of DERs 4. Line Section or Nodal Level on the Primary Distribution System 5. Thermal Ratings, Protection Limits, Power Quality (including Voltage), and Safety Standards 6. Publish the Results via Online Maps7. Use Time Series Models 8. Avoid Heuristic approaches, where possible9. Perform the complete ICA analysis for all feeders down to the line section or node on two Distribution Planning Areas (DPA).3

In addition to the ACR, the May 2016 ruling established the ICA Working Group (WG). The purpose of the WG is to monitor and provide consultation to the IOUs on the execution of Demonstration Project A and further refinements to ICA methodology. CPUC Energy Division staff has oversight responsibility of the working group, but it is currently managed by the utilities and interested stakeholders on an interim basis. The utilities jointly engaged More Than Smart, a 501(c)3 non-profit organization, to facilitate the WG. The Energy Division may at its discretion assume direct management of the working group or appoint a working group manager.

Between May 2016 and this Final WG Report, the WG met 16 times. Discussions focused on developing the ICA have been facilitated by MTS. The WG has benefitted from contributions by a large range of stakeholders. A complete list of contributing stakeholders is provided in Appendix A. The WG expects to continue its efforts through August 2017 as it begins to address long-term refinement items.

In December 2016, Pacific Gas & Electric (PG&E), Southern California Edison (SCE), and San Diego Gas and Electric (SDG&E) submitted their final Demo A reports, representing a substantial milestone for the WG. These reports summarize demo results, lessons learned, and the IOUs’ recommendations on methodology calculation and feasibility of implementation of the ICA across the entire distribution system.

1 Assigned Commissioners Ruling, November 2014. (http://docs.cpuc.ca.gov/PublishedDocs/Efile/G000/M141/K905/141905168.PDF)2 http://www.cpuc.ca.gov/General.aspx?id=5071 3 Assigned Commissioner’s Ruling, May 2016. (http://docs.cpuc.ca.gov/PublishedDocs/Efile/G000/M161/K474/161474143.PDF)

2





This Final Report contains the full Working Group’s recommendations for the ICA methodology and related actions coming out of stems from the IOU’s December filings. The , constituting the full Working Group’sWG offers recommendations on the significant progress made to date, recommendations on how to improve the methodology going forward and how the results may be used to inform decision-making on the part of the Commission, utilities, providers of distributed energy resources, and customers. These include recommendations on the following three categories:

[1.] Uses of ICA: the WG identifies two primary use cases for the ICA. The first and most developed an immediate is for use of ICA to facilitate a significantly streamlined interconnection process. The second less developed use case is for , and an ongoing use of ICA within various system planning processes to identify when and where capacity upgrades are needed on the distribution system, as a result of DER growth.. [SCS: there is no reason to limit this to changes as a result of DER growth, the ICA could be used for all sorts different types of changes] The WG report clearly outlines immediate, near and long term methodological refinements to enable the use of ICA within the interconnection process, and lays out considerations for the planning use case, with a goal of developing methodology recommendations for use within the planning context in the near-term (and in coordination with ongoing planning proceedings at the PUC).

[2.] Development of Common IOU methodology: the IOUs tested ICA under two separate methodologies, known as the “iterative” and the “streamlined” methodology, in Demo A. A majority of WG members recommend that the iterative methodology is used for interconnection purposes, with added refinements detailed further in this report. PG&E recommends a “blended” approach in using both methods for interconnection4. The streamlined methodology may provide value into the planning process, and will continue to be discussed as the WG further defines the uses of ICA in system planning.

1.[3.] Timeline: the WG recommends that the IOUs implement ICA across their entire service territory within 12 months of a PUC final decision on ICA methodology, and include the identified recommendations from this report. Further, the WG recommends that the Commission establish two processes to incorporate modifications to the ICA:

a. As the WG continues to refine and enhance the ICA based on new studies and developed methods, it is requested that the Commission establish a process to allow the ICA WG to approve multiple enhancements to the methodology as they are developed in consultation with the WG during the long-term refinement process.

[b.] The WG makes the following refinements based on best available knowledge and in-depth discussions among a broad stakeholder group, though there are recommendations contingent upon costs and modeling software capabilities. It is requested that the Commission establish a process for the IOUs to consult with the PUC Energy Division if one or more IOUs determines that a methodology component (specific to the added refinements) cannot may not be included in the first system-wide rollout of ICA, due to these issuesissues not possible to predict at this time (i.e. those

4 See PG&E’s final Demo A report: http://drpwg.org/wp-content/uploads/2016/07/R1408013-PGE-Demo-Projects-A-B-Final-Reports.pdf

3

Laura Wang, 02/19/17,

Placeholder language: IOUs to provide timeline information and rationale further down in document: Is it still the opinion of some stakeholders that implementation should be 12 months from filing of report?





discovered during the rollout process). For those changes, the WG expects that IOUs will prioritize their resolution in the next iteration of ICA.

The CPUC envisions that an approval of a final ICA methodology common across all utilities will be made in a Proposed Decision in early 2017.

[SCS: Re: below list Should this just be in an Appendix as noted above? Also, I think there needs to be a caveat added below since the vast majority of the organizations listed participated minimally, if at all. It doesn’t seem quite right to suggest that they all contributed or even were following what happened. I attempted something but I wasn’t quite sure how best to convey this.]The “Working Group” references all active parties participating in ICA WG meetings, which include the IOUs, government representatives, DER developers, nonprofits, and independent advocates and consultants. The following stakeholder groups attended at least one meeting or webinar of the ICA WG, but there were only a handful of organizations that participated actively throughout and helped to draft this Final Report:

- ABB Group- Advanced Microgrid

Solutions- Alcantar & Kahl- AMS- Artwel Electric- Bloom Energy- CAISO- California Energy

Storage Alliance- California Energy

Commission- California Public

Utilities Commission- CPUC Office of

Ratepayer Advocates - California Solar

Energy Industries Association

- City of Burbank- Clean Coalition- Community Choice

Partners- Community

Environmental Council (Community

Renewable Solutions LLC representing)

- Comverge- DNV GL- ECCO International

Inc.- Energy and

Environmental Economics

- Electric Power Research Institute

- Energy Foundation- Environmental

Defense Fund- Gratisys Consulting- Greenlining Institute- Helman Analytics- ICF International- Independent Energy

Producers Association

- Independent advocates

- Independent consultants

- Integral Analytics

- Interstate Renewable Energy Council

- Kevala Analytics- Lawrence Berkeley

National Laboratory- Lawrence Livermore

National Labs- Natural Resources

Defense Council- Northern California

Power Agency- NextEra Energy- New Energy Advisors- Nexant- Open Access

Technology International

- Pacific Gas and Electric Company

- PSE Healthy Energy- Quanta Technology - Sacramento

Municipal Utilities District

4





- San Diego Gas & Electric

- SEIA- Shute, Mihaly &

Weinberger LLP - Siemens- Smart Electric Power

Alliance

- SoCal REN- SolarCity- Solar Retina- Southern California

Edison- Stem Inc. - Strategy Integration- Sunrun

- SunPower- The Utility Reform

Network- UC Berkeley- Vote Solar

The ICA WG met consistently to discuss proposed methodology for Demonstration A and review final Demo A reports. A full summary of WG documents including meeting agendas, presentation slides, and participant lists may be found in the Appendix.

Scope and Organization of the Report All three IOUs submitted their Demo A reports at the end of December 2016 and the maps and downloadable data was made available in January and February of 2017. These reports lay out in detail the assumptions and calculations used within the ICA methodology. Additionally, the IOUs each separately made their own recommendations on which final methodological choicesmethodology (i.e., using a streamlined or iterative methodology) should be used going forward in a system-wide rollout of ICA. based on two identified use cases of ICA discussed by the WG (ICA is expected to be used to facilitate a streamlined interconnection process, and to provide useful input in guiding utility system planning efforts). These recommendations were are made to inform the CPUC in making a proposed decision on final ICA methodology. The WG took those recommendations into account but has reached different conclusions in some areas.

In order to provide the CPUC with useful recommendations while allowing the IOUs to continue implementation of the ICA, the WG is submitting this report now, but expects to file separate considerations indicating where cost effectiveness should be strongly considered by the Commission in deciding on a common methodology. A second filing will both give the IOUs the opportunity to summarize a general cost range for implementation of various methodological options, and allow the WG to consider additional whether different or addditional recommendations are necessary based upon the cost estimates provided.

The WG acknowledges that continued deliberation with regards to cost impacts and cost recovery will likely occur in a separate forum. It is also acknowledged that theencrouages the IOUs tocan continue to engage in some work related to the full system roll-out, such as data clean-up efforts, independent of a CPUC proposed decision.

Further, the WG acknowledges there is a need to balance the incorporation of methodological changes and still implement the ICA on the entire distribution system in a timely manner. Further, there are

5


Current placeholder language – WG understanding that recommendations may differ after considering cost estimates from IOUs.

Sky C. Stanfield, 02/19/17,

I participated in the WG on behalf of IREC, there probably sin’t any reason to list my firm.





several meaningful recommendations made in this report that were discussed among the WG, but not necessarily required to be tested as a part of Demo A. For these recommendations, the IOUs and stakeholders jointly engaged in discussion regarding the need for the changes and the practical feasibility of need and practical feasibility on whether the change should be incorporating themed in the initial system-wide rollout vs. , or to setting them it as a longer-term goals. For those identified for inclusion in the first system-wide rollout, the WG understands that IOUs may encounter unexpected issues in their implementation. To this end, the WG requests that the Commission establish a formal process whereby the utilities may submit a request to the PUC Energy Division regarding a change in methodology that is needed as a result of unforeseen circumstances, for further consideration. For those changes that cannot be resolved in the first rollout, the WG expects that IOUs will prioritize their resolution in the next iteration of ICA and work with stakeholders on any refinements that may be necessary as a result of new learning.

Similarly, the WG expects that additional work developing over the next six months on long-term refinements of the ICA (as specified under the ACR) will result in the development of additional methodology or refinements that may be included in either the first system rollout, or subsequent iterations of the ICA. Given that the ICA is understood to improve over time, given incorporations of best efficiency practices and additional methodology, the WG requests that the Commission establish a process whereby improved iterations of the ICA may be incorporated through a WG approval process [SCS: Seems like the Commission might prefer that the Energy Division be the actual one “approving” anything?], rather than through a formal request through the PUC. This gives IOUs flexibility to phase in additional ICA enhancements as they are developed through the long-term refinement process (e.g, fixed power factor functions) with expediency. The ACR additionally specifies multiple items the WG should focus on to continue refining the ICA methodology. The WG filed an interim long-term refinement report in December 2016 detailing work to-date on those items, and sorting topics into a tiered system to develop a rough schedule for WG work in 2017. Since the filing of that report, the WG has identified additional long-term refinement items after reviewing the IOUs’ final Demo reports. Those items are included in an amended Table5, and the WG will endeavor to prioritize this list at the beginning of its long-term refinement work. For the WG, “long-term refinement” details WG activity 6 months after the filing of this final report, beginning March 1, 2017 and ending August 31, 2017.

To this end, the WG agrees to identify consensus items where parties have built consensus, and identify specific parties places where there is non-consensus and alternative proposals have been made.

[SCS: Since our WG really only wrote recommendations principally on things that needed to be changed or refined from the Demo or where a choice was presented, we may want to clarify that WG silence on specific aspects of the methodology outlined in the final utility reports should be considered as support for, or at least non-opposition to, the utilities proceeding as they did in the Demo.]

5 See “Next Steps” section

6





Discussion Topics, Points of Consensus, Proposed Alternatives if Any, Recommendations, Next StepsThe sections below detail the further refinements to ICA methodology the WG recommends for incorporation. Where possible, recommendations are mapped to the specific section in the ACR.

The WG recommendations can be defined in three categories:[1.] ICA Use cases of ICA1.[2.] Standardization of methodology across IOUs2.[3.] Short-term activities defined under ACR Section 3.1 for WG consultation

A high-level summary of recommendations for the interconnection use case, modified from the methodology tested in Demo A, is provided here: [SCS: ? is something supposed to be filled in here? or are you referring to the section below?]

Use Cases

At the beginning of the WG process, it was agreed that the WG should identify the specific uses of ICA and make recommendations on ICA methodology based on these concrete use cases, to the full extent possible. It is expected that methodological considerations, with regards to frequency of updates, hourly profiles, and other modeling options, may change based on the use of ICA.

Through multiple meetings, the following use cases were discussed:

[1.] Informing interconnection siting decisions and fFacilitating a significantly streamlined interconnection process

ICA results can be used to streamline the interconnection study process, decrease costs of interconnection, and improve project certainty by providing information to customers and third parties about the amount of DER capacity that can be interconnected at a specific location. It is expected that the development of ICA will coordinate closely with the separate Rule 21 proceeding. Thus, the WG proposes that an interconnection use case to be adopted by the Commission include the following considerations. Utilities also specifically point out a need to coordinate the application of ICA with regards to timing, configuration, and equipment. [SCS: I am not sure I understand what is meant by the last sentence here]

a. Developers should be able to submit a Rule 21 application if a project is below the identified ICA value for their technology/project type and receive approval for those screens the ICA has evaluated, in coordination with the Rule 21 proceeding to ensure safety and reliability.

b. ICA values are expected to replace multiple technical criteria (screens) in Rule 21, which are arbitrary or not location-specific. [SCS: Was this intended to address the fast track size limit issue? As written it is not very meaningful to me. I suggest the following replacement: “ICA values are expected to replace the need for some size limits and technical screens in the

7


Tom Russell – I tried to put your comment here, let me know if this works





Rule 21 Fast Track process that currently rely on broad assumptions that can be better addressed through the site-specific analysis contained in the ICA.”]

[c.] The ICA will work towards replacing arbitrary fast track screens within Rule 21 where the ICA results can address the technical concern that the screen originally sought to address. In the first system rollout, the ICA is expected to replace screens F (Short Circuit Current Contribution), M (Aggregate Generation ≤ 15% of the line section peak load), and N (Penetration Test). In the long term, the ICA is expected to resolvereplace additional screens, including Screens G (Short Circuit Interrupting Capability), L (Transmission Dependency / Stability Test), O (Power Quality and Voltage Fluctuation), and P (Safety & Reliability), as well as address single-phase lines, pending additional information and study. [SCS: out of respect for the utilities, I think the word arbitrary is probably not quite right, the screens were developed for a reason, but the advent of the ICA is just a more precise tool. I added the general descriptors of the screens from the Rule 21 flow chart since for most readers the letters are meaningless. Alternately these could be added in a footnote on the same page to make it easier to read)]

i. Screen G may need development to add to the components being considered in ICA, as it specifically evaluates interrupting capability (ICA currently looks at reduction of reach). [SCS: This needs further explanation, as written I don’t think most readers (myself included) would understand the relationship between interrupting capability and reduction of reach]

[ii.] Screen L is focused on transmission stability and dependency that cannot’t be analyzed by an ICA that currently only evaluates the distribution system impactsICA, thus, – this screen will still need to be evaluated used to identifyfor transmission constrained areas.

ii.[iii.] Screen O may have issues with voltage stabilization and cause overvoltage based on the line configuration and the specific transformer configuration of the applicant. [SCS: Needs revision, Screen O itself doesn’t cause anything to happen, does the following revision capture what was intended better?: Screen O will continue to be required in the near-term as the first rollout of the ICA will not identify all issues with voltage stabilization. The ICA will not identify all cases were overvoltage may still occur based on the line and transformer configuration of the applicant and thus the use of Screen O is still necessary in the near-term.)

iii.[iv.] Screen P can be impacted by transformer configuration due to its ability to see ground faults and must be evaluated when specific applicant information is provided. [SCS: again this doesn’t make sense as written, screen P is not “impacted”, I assume you mean the distribution system can be impacted by .. but I am a little less sure how to re-word the rest.]

[d.] Further development of ICA may identify additional criteria or screens that can be replaced or otherwise reviewed in an expedited mannerrecommended for expedited review, in coordination with the Rule 21 proceeding.

c.[e.] Projects should qualify for Fast Track if the applicable project size is less than the lesser of either the ICA at the Point of Interconnection or the existing Fast Track eligibility limits.

8





d.[f.] The ICA should be updated frequently enough to allow for a meaningfully faster (if not ultimately “automatic”) interconnection process for projects that are proposed below the ICA value at their point of interconnection.

[i.] Some WG stakeholders believe that, at a minimum, the ICA should be run at least annually system-wide and be, updated on a “case-by-case” basis at least monthly (i.e. when new projects are added or other system changes occur that meet a condition threshold) at a minimum monthly basis to allow the ICA figure shown on the maps to be valid for interconnection requests at nearly all times (or earlier, to the extent feasible).

i.[ii.] SCE supports system-wide monthly updates for the initial rollout, and consideration of additional functionality and higher levels of frequency in subsequent iterations, such as case-by-case updates, weekly or on demand updates (contingent upon funding ability). [SCS: is this really any different than above? It seems to me both are getting at the same goal of an ICA that is updated at least monthly system-wide (I don’t think it is important to the stakeholder group whether the entire ICA is done system wide or on a case-by-case monthly, so long as the ICA value is considered accurate at least once a month for the whole system. I make this point to suggest we should consolidate and state the consensus where possible to help the Commission.]

ii.[iii.] PG&E notes that these envisioned condition-based updates require significant front-end coding and development to implement properly, and may create additional costs and/or delay implementation. [SCS: is this really needed here PG&E? It seems to me like this point could be made all over this document, it would be helpful if it was more clear, i.e. Does PG&E not support the above condition based upon PG&E’s expectation regarding how much it would cost? Or are you just noting it will cost something?]

e.[g.] As a long-term vision, the ICA should be updated on a real-time or daily basis to the extent possible, reflecting values that can be used in an automated interconnection process. Future enhancement should work towards this goal, while considering issues such as the following (as highlighted by IOUs’ experience): [SCS: It seems to me like most stakeholders could agree to incorporating the two caveats below into the recommendation if there is a desire for this to be consensus driven. Namely, I think stakeholders understand that as we move towards a more automated process for being able to update the ICA on a more regular/frequent basis that there will be a need to address how “queued” projects are treated and how the potential “actual” conditions associated with projects will be predicted ahead of time. I believe both SCE and PG&E are getting at essentially the same issue below]

i. PG&E: integration issues can be dependent on specific application information that can’t be known ahead of time to be reflected in ICA. Further, generation queuing, commercial operation dates, and planned work/transfers can all have a unique impact on certain locations in the system, which must be considered application-by-application with manual engineering review, since specific elements can’t be predicted or known ahead of time.

9





ii. SDG&E: there is the opportunity to claim queue positions via speculative projects – the implementation of such a function should additionally come with a change to Rule 21 for more strict enforcement of applicant timelines and milestone provisions.

f.[h.] Presentation of ICA values:[i.] ICA information will be presented in both online maps and downloadable data

formats. The ICA results themselves are information is technology-agnostic, although they may be displayed on the maps as specific to certain technologies (i.e. solar PV) to provide relevant information to the largest group of users.

[ii.] The ICA value used for the interconnection review should be the same ICA value shown on the online maps – as noted above, at a minimum, the online ICA maps should change within a month of the ICA value changingshould be current system-wide on at least a monthly basis and the maps should thereby reflect the updated values at least monthly. [SCS: edited to match the recommendations above and b/c the way it was written it sounded to me like the maps would change within a month of the update, not when the update is done to be current]

[iii.] The ICA value used in the interconnection review process shouldwill provide monthly and hourly data about hosting capacity limitations that enables a developer to design a system that takes full advantage of the available hosting capacity at their proposed point of development. The use of this information in the interconnection process will be, contingent upon verification that the system design addresses the appropriate hosting capacity limitations based upon the temporal changes. In addition, there will need to be and increased communication and visibility built into the interconnection agreement process to ensure operational profile of these projects abide by ICA hourly limitations.

Table 1: Summary of Recommendations for Interconnection Use Case For full detail, please reference specific report sections.

[SCS: regarding item 2 below, I think the caveats of SCE and PG&E are confusing, it isn’t clear to me exactly what is being objected to and I suspect there is actually consensus here if framed correctly. As I understand it stakeholders want an ICA figure that is accurate at least monthly, I am not sure why SCE would want it “not more than monthly” beside the same cost and practical considerations already taken into account by the stakeholders. Similarly, with PG&E’s comment, do they mean they want to update it constantly, or what? Time is clearly relevant here and I think what PG&E is suggesting is more that we shouldn’t require information to be updated on a specific temporal basis if there is no change, which stakeholders likely agree with. As it is, I am not sure the Commission will really understand what SCE and PG&E want as written.].

[SCS: Re Item 12 below: I think the IOUs were going to write up some explanation of their reasoning behind the amount of time it would take to include in the report. The idea was that we would all be able to review that and decide if we could support the 12 months from decision approach if better explained. I think for the stakeholders it is important that the utilities commence whatever work can

10





reasonably be expected to not change pending CPUC decision and we would like to better understand why 12 months is needed for everything else.]

Majority Consensus Recommendation: first system rollout

Non-Consensus and Alternatives

WG activity on Long-Term Refinement (6 months)

Refer to Report Section

1. Methodology Iterative PG&E: “blended” approach (see final Demo report)

2. Update frequency

Non-IOU stakeholders: At least monthly, on case-by-case basisSCE: no more than monthlyPG&E: dictate updates by conditions, not time frame

Use cases

3. Hourly profile 96 24, 576, 8760 <missing report section here? Or would include in a cost considerations section>

4. Voltage regulating devices

Float Power system criteria

5. Operational flexibility

Publish two ICA values: 1) no reverse flow across SCADA operated devices, 2) reverse flow up to substation busbar

Develop new methodology to eliminate or minimize use of heuristic

Power system criteria

6. Circuit models Incorporate changes to circuit models

Alternative methodology to address pre-existing conditions

Circuit models

7. Smart inverters

Begin work with software vendors to determine best means of incorporating smart inverter data when methodology is developed

Begin studies on reactive power, static volt/VAR, fixed power factor – other studies TBD

3.1.b

8. Maps Publish PV value rather than agnostic generationSearchable database

Single phase line sectionsMachine readable data

3.1.c

9. Computational efficiency

Reduction of nodesCost considerations

3.1.d

10. ORA success 3.1.e

11


ORA requests review and comments on proposed success criteria


All 3 computational efficiency methods approved?


At 2/15 call, IOUs asked stakeholders to review IOU published maps again and reconfirm whether its functionalities serve requests.


Timing correct?


From ORA evaluation of success criteria


Still need discussion on cost considerations here?


Pulled from stakeholder subgroup recommendations doc, placeholder before cost considerations convo?


Will cite Section #s and pages later during final report formatting


Include with





criteria and reference circuits

11. Smart meter 3.1.f12. Timelines Non-IOU stakeholders:

12 months from filing of WG final reportSCE and SDG&E: 12 months from PUC decisionPG&E: June 2018, potentially longer based on methodology

3.1.g

13. Scope System Wide

2. Distribution system annual planning

It was determined that there is a role for a planninga use case for the ICA focused on planning is useful, as it is expected that the ICA can help determine and guide where and when future capacity may and may not be needed. It is envisioned that ICA results may guide sourcing and procurement of DER solutions with additional locational granularity in the future. The three IOUs all propose to use the streamlined methodology in the planning context, as the iterative methodology creates a large amount of granular and complex data, and requires considerable resources to conduct multiple scenario analyses. However, many components of this use case remain less certain, due to multiple ongoing efforts in other CPUC proceedings that will inform how ICA will be used in system planning, as well as further clarity into the utility annual planning process itself. Further, the multiple ways ICA may be incorporated into planning (from guiding grid modernization investments, to how DERs may be evaluated as solutions in the IRP) are quite variable in the level of detail (e.g., granular hourly profiles, frequency of updates, etc.) they require from the ICA methodology. Because many open questions remain about the precise definition of the planning use case, the WG was not able to make specific recommendations regarding the appropriate methodology that would ultimately serve the use case best. Finally, it was determined that the need to incorporate ICA in planning purposes, while highly important, is less immediate, particularly compared to the use of ICA in streamlining interconnection.

Thus, the WG proposes to further define the planning use case as a key high-priority long-term refinement issue to begin in March, and outlines several considerations for the planning use case going forward:

a. Taking this approach will allow the WG to form a specific list of uses of ICA in planning, evaluate the methodological needs for each use case, and determine whether the iterative or streamlined method may better serve that use case

b. The streamlined methodology continues to be a viable option for use. It is understood that many steps the IOUs will take to employ the streamlined methodology for the first system-wide rollout will additionally benefit the deployment of a streamlined method later down the road.

12





c. Achieving these identified uses may require a blended approach (using aspects of both the iterative and streamlined methodology).

[d.] The WG appreciates and understands the benefits that employing a streamlined method offers, and looks forward to better identifying its limitationsevaluating its application to the planning use case in further WG meetings.

The following suggestions provide a good starting point to further WG discussion, though the WG requests further guidance from the PUC on specific uses of ICA in these contexts:

a. The scale, pace and prioritization of ratepayer funded grid modernization investments should be guided by projected ICA values. ICA may be one tool to guide and prioritize ratepayer-funded investments for grid modernization.

b. ICA may be used to evaluate the impact of DER to serve as potential solutions to complement the Integrated Resources Planning process

c. The current system capacity revealed through the ICA may be combined with location-specific projections of DER growth (i.e., DER growth scenarios) to come up with projections of system needs.

d. The ICA and LNBA may be considered in tandem to identify opportunities where additions to hosting capacity can enable DER growth that would avoid distribution system upgrades that would be more expensive than the cost of increasing hosting capacity on specific circuits.

Recommendations on iterative versus streamlined methodology

The IOUs were directed to develop a “streamlined” hosting capacity analysis based on four steps: 1) establish distribution system level of granularity; 2) model and extract power system data; 3) evaluate power system criterion to determine DER capacity; and 4) calculate ICA results and display on an online map. The IOUs were also asked to develop a power flow-based “iterative” methodology for comparison purposes, and submit a detailed comparative summary of both methodologies based on Demo A results.

The streamlined method uses an abstraction approach, applying a set of equations and algorithms to evaluate power system criteria at each node on the distribution system. The streamlined method first performs a baseline power flow simulation to acquire the initial conditions of the circuit that will be used in the streamlined calculations. These conditions can be, but are not limited to, electrical characteristics such as thermal ratings, resistance, voltages, current, fault duties, etc. The streamlined method then evaluates the full set of criteria, including thermal, voltage, protection, and safety limits independently to determine the maximum hosting capacity at a given node or component of the system. The Ssimpler methods utilized in the streamlined methodology may not capture some of the more dynamic effects on the more complex circuits. However, the ability to utilize simpler equations and algorithms within a database can enable faster computations on large datasets.

The iterative method performs an iterative power flow simulation at each node on the distribution system. The iterative method solves for thermal and voltage conditions simultaneously using voltage simulations. Fault flow simulations are used for protection criteria not dependent on power flows. Due

13





to the large number of iterations required, iterative analysis can result in long processing times, especially when expanded to large numbers of distribution circuits. However, the use of an iterative simulation parallels what IOUs would perform as part of an interconnection study. This technique is expected to provide more confidence in representation of integration capacity on more complex circuit conditions.

The IOUs presented a comparison summary of Demo A results using both methodologies, and outlined recommendations within their Demo reports on which methodology, or portions of methodology, they believe should be employed in a full system-wide rollout. The rationale behind these recommendations is based on lessons learned from the Demo projects and full system-wide implementation considerations, including computational efficiency, ability of CYME/Synergi software, and costs.

Should we include summary of IOU recs in Demo reports, or just language to refer to full Demo reports? [SCS: Seems like a reference is fine, in light of the limited time line I think other areas of this report warrant that attention more]

Leave Placeholder Here for Potential Cost Discussion

After multiple meetings, a majority of the WG recommended that the iterative methodology be used for the interconnection use case (with the following refinements detailed in this report). [SCS: I strongly believe this recommendation needs to be explained, it is not very meaningful to the Commission without an explanation, particularly since full consensus was not reached, I attempted an explanation below, but think further detail could be more helpful.] This recommendation from the majority of the WG is driven by a couple of principle factors. First, the results of the Demonstration project in the final utility reports raised significant concerns for the WG regarding the accuracy of the streamlined results. Since precise accuracy is necessary to allow the ICA results to be relied upon in the interconnection process without forcing additional caveats and manual review, this appeared to disqualify the streamlined process for application in the interconnection use case. Second, the WG considered using a combined method, as recommended by PG&E, but found that it undermined many of the goals of the interconnection use case which seek to move towards an automated process that requires less manual review by engineers and would enable the ICA information displayed on the map to be the same as what is applied in the interconnection process. Third, while the WG is very aware of the computational intensity of the iterative method, it appears that there may be reasonable ways to reduce the data intensity and also utilize more efficient computing resources to address some of these concerns. In the long run, it seemed likely to the WG that the costs of the computing issues could be reasonably managed as technology and understanding of the ICA methodology advance. Finally, the WG is sympathetic to the concerns of PG&E, particularly those that arise out of unique circumstances associated with the current state of their software and associated operations. However, the WG stakeholders strongly want to maintain a consistent approach across all three IOUs in order to facilitate future advancements and limit confusion across the state.

PG&E separately proposes a “blended” approach, whereby both the streamlined and iterative methods are used for the interconnection use case – the streamlined method would be applied

14


Please let me know if your party would like to include an alternative recommendation here





to an overall analysis for the whole system, and iterative would be utilized to analyze specific conditions within the interconnection process. [SCS: I would find it helpful if the preceding and following comments could be clarified to make it clear whether PG&E’s desire to use a blended approach is largely driven by their own unique circumstances, or whether they would be making that recommendation either way. I also think it would be helpful for the Commission if PG&E explained why the tradeoff of not having an ICA map that matches the interconnection outcome is worth making]

o PG&E notes that adoption of this blended approach will require fewer engineering resources. PG&E projects that if iterative is required for use then a new team would be needed to manage the ICA process. It is projected that if streamlined is adopted and the iterative approach is adopted more efficiently on application basis then the new work load can be more efficiently managed with current engineering resources.

o PG&E is also undergoing existing planned work on modifications to the gateway to (1) utilize the new GIS system implemented in 2016, (2) expand gateway to include substation models, and (3) expand ability to have service transformers in the models. If recommendations require the incorporation of planned modifications and automated iterative across the whole system then significant additional work would be required on the gateway and could postpone work to include substation and service transformers. Also, if PG&E’s recommendation of application based used of iterative is not adopted, then more engineering resources would have to be hired and trained in order to perform the regular iterative ICA analysis. Adopting the application based iterative and system wide streamlined recommendation would allow PG&E to more efficiently use existing resources and tool capabilities.

The WG understands the value of running ICA using the streamlined methodology, and continues to consider its use as a possibility in the planning use case (to be developed as a priority long-term refinement item).

ACR Requirements – ICA Modeling and Input Assumptions

Modeling and extracting power system data: The IOUsS used LoadSEER, a load forecasting analysis tool, to develop load profiles at the feeder, substation, and system levels. Load forecasts play a critical role in forecasting grid conditions used to establish ICA values. As DER load growth is identified to be addressed through a parallel track of the DRP proceeding (Track 3, subtrack 1), The WG recommends that the findings and recommendations from the CPUC workshop on Load and DER forecasting, as well as all findings from this DRP sub-track, be incorporated into the ICA to coordinate the load forecasting methods used in determining ICA value.

Power system criteria methodology: ICA results are dependent on the most limiting power system criteria. The four criteria used for Demo A are:

15





1. Thermal criteria: amount of additional load and generation that can be placed on the distribution feeder without crossing the equipment ratings

2. Power quality/voltage criteria: voltage fluctuation calculated based on system voltage, impedances and DER power factor. Voltage fluctuation of up to 3% is part of system design criteria for all three utilities.

3. Protection criteria: required amount of fault current fed from the sub-transmission system due to DER operation.

4. Safety/reliability criteria: operational flexibility that accounts for reverse power flow issues when DER/DG is generating into abnormal circuit operating scenarios. Other limitations supporting the safe and reliable operation of the distribution system apply. [scs: such as? are they actually applied as part of the methodologies? If not I suggest striking this last sentence. I also think if operational flexibility is the only real limit being identified here it should be renamed accordingly. I make this recommendation understanding that operational flexibility is key to safety/reliability, but it is not the only factor on that topic and precision matters in understanding the results]

The WG developed recommendations regarding the input assumptions in for the power system criterion used. To the extent possible, the WG identified where methodological changes may apply to specific use cases.

Power quality/voltage criteria: The IOUs take various approaches to how they treat voltage regulating devices within the iterative methodology. Devices may be “locked” or “unlocked”. Currently, CYME software does not have this capability, while Synergi does allow devices to float. Through WG meetings, it was explained that the CYME module used for Demo A locked voltage devices to better allow for modeling convergence. The WG recommends that, for the iterative methodology, voltage regulating devices not be are not locked down, and that the IOUs work with software vendors to encourage its inclusion into the first system-wide rollout. The WG is open to continued discussion on the number of iterations of adjustment if needed.

Safety/reliability, or “operational flexibility”: Demo A tested two scenarios for operational flexibility: (1)one prohibiting reverse flow of power across any SCADA-operated device, and (2) evaluating maximum DER capacity, irrespective of power flow direction. To calculate the no-backflow [SCS: do you mean the values for (2)? that seems to be the “backflow” not “no backflow” case?] values, IOUs excluded the safety/operational flexibility criteria when determining the minimum or limiting ICA value for each node, such that the minimum of four, rather than five ICA criteria values were used.

The IOUs concluded in their Demo A projects that no reverse flow limits are necessary within the ICA for reasons beyond operational flexibility. Specifically, the no reverse flow limits are needed in some cases to prevent violations of the power quality and voltage limits, as existing voltage regulators (both field and substation) may not currently be adequate to allow for backflow, and existing control settings may not be adequate to properly manage increased levels of DER (some controls are programed to existing system conditions).

16





The intent of the safety/reliability is constraint is to ensure that all operational flexibility is preserved when DERs are added to the grid. The SCADA-operated devices represent points at which the grid can be reconfigured, either permanently or temporarily. The ability of the grid to tolerate reverse flow depends on the configuration. By prohibiting reverse flow at these points, the ICA determines the DER adoption that produces no reverse flow in any configuration

The WG recognizes that the operational flexibility criterion as implemented and described above is based on engineering practices that allow for calculation of the operational flexibility criteria across all circuits. Consequently, the results of Demonstration A show that operational flexibility, as currently modeled by the IOUs, is a significant (possibly the most significant) limit to ICA that produces conservative results. A goal of the development of the ICA methodology was to avoid heuristics to the degree possible and to favor engineering analysis. The Working GroupWG agrees that the operational flexibility criterion could accurately be described as a heuristic.

The operational flexibility criterion based on no reverse power flow across SCADA-operated devices was implemented in demonstration A because no other options for ensuring operational flexibility were identified and determined to be feasible given the current understanding of the capabilities of either the iterative or streamlined methods. The Working GroupG agrees that this was a reasonable short-term path, but believes that developing an improved approach to evaluating DER adoption limits related to operational flexibility should be an ICA development priority.

The operational flexibility criterion based on no reverse power flow across SCADA-operated devices is a reasonable short-term solution to the preservation of operational flexibility, and should therefore be carried forward to show results both with and without this constraint in the first system-wise rollout of the ICA in order to start that process without waiting for further refinement of the criterion.

The first system-wide rollout of ICA results will publish two values: 1. Operational flexibility limitations on SCADA devices, and 2. No operational flexibility limit that allows power to transfer up to the substation busbar, but does not allow power flow across the transformer (maintaining backflow constraint).

Publishing both values will better indicate the hosting capacity where this factor is mitigated or determined to be non-constraining through Supplemental Review. It is important to note that this second value is different than the value tested in Demo A. It should be replaced with an improved, less conservative approach based on engineering analysis as soon as one can be developed, which is expected to replace Screen P (the Safety and Reliability Screen) in the Rule 21 process.

Many WG members place high priority on development of an improved, non-heuristic approach to evaluating operational flexibility limits value as a key long-term refinement item.

For these WG members, it is envisioned that the WG develop an improved, less conservative approach based on engineering analysis that evaluates whether a limit on operational flexibility

17





actual results in any safety or reliability impacts. This new approach may be enabled by an improved understanding of the ICA’s ability to evaluate a large number of scenarios and configurations or by a discussion of how the utilities study the operational flexibility impact of an interconnection application that requires such a study.

This view is not shared across IOUs, which would like to examine whether the problem may be solved in future years independent of new WG methodology, through the implementation of other potential solutions. The WG will determine a more detailed priority list of items in the beginning stages of the long-term refinement process.

The WG and the PUC should consider the question of the interplay between ICA and operational flexibility. If increased DER adoption has the potential to become a consideration in operational flexibility, how can we quantify the impact of the change in operational flexibility? What kind of change in operational flexibility is appropriate to reach policy goals related to DER adoption? Are there technical and/or policy solutions to expand ICA while still preserving operational flexibility? Understanding these questions may require a separate research initiative or pilot project.

Circuit modelsUpdating circuit models Currently, updating the models in both CYME and Synergi is difficult. Currently, when new data from GIS or other sources is introduced to the circuit models, both CYME and Synergy lose convergence. Without the ability to accurately update information contained within the models, they have not yet proven themselves for a full circuit rollout. Therefore, the WG recommends the IOUs’ models demonstrate the ability to incorporate changes to their circuit models before a full circuit modeling takes place.

Pre-existing conditionsThe WG identified a challenge whereby circuit models sometimes display containing violations of one or more of the power system criteria before the DER adoption is modeled (i.e. a pre-existing condition on the circuit is responsible for the violation) and thus the ICA results show a zero-sum hosting capacity. A targeted DER solution may not exacerbate impact or, in some cases, could even improve the existing violation criteria. However, it may be difficult to determine whether adding a DER solution worsens a violation criteria or creates an entirely new violation. During the distribution planning cycle, utility planners check if any violations occur due to changing loads, and updated models are created to reflected how the utility will periodically resolve these issues. However, it is not clear how these are reflected within ICA. The WG acknowledges there would be multiple challenges to reflect pre-existing conditions within the ICA.

To address this condition, the IOUs understand that (1) ICA should be limited by pre-existing condition when adding DER degrades the pre-existing condition and (2) that ICA not be limited by pre-existing condition when adding DER improves the pre-existing condition. For example, in a situation when low voltage exists in an area, adding generation may improve the low voltage condition but adding load may degrade the pre-existing conditions. In this particular example, generation ICA would not be limited by the pre-existing condition but load ICA would be limited by the pre-existing condition. It must be noted

18


What is the timing consideration? First system rollout or LTR?


Taken from ORA rec – see 3.1.e


Can IOUs provide more clarity here?





that in some cases, such as substations with LTC control, adding generation to a low voltage pre-existing condition may actually further degrade the low voltage condition rather than improve the low voltage condition. For this reason each pre-existing condition should be evaluated to determine if adding DER would degrade or improve the pre-existing condition.

The IOUs would need to create automated processes that may can be able to be developed as part of the development of the ICA implementation plan to efficiently evaluate the feeders and substations for pre-existing conditions. These processes would determine if any pre-existing conditions exist and to determine if adding DER would improve or degrade the detected pre-existing condition and take the necessary action to determine when ICA can be allowed or when ICA must be limited by the pre-existing condition. It is expected that this process would require much IT front end to automate and/or significant engineering resources to properly consider on a regular update frequency. To date there has been no effort to quantify these potential existing costs to determine their magnitude.

PG&E’s specific viewpoint is that while iterative is ideally what is desired in the long term, it is not likely that analyzing iterative while considering pre-existing conditions, as stated above, is going to the best next step forward in the short term. Based on PG&E’s 7 years of model gateway experience and our recent experience with standing up the automated CYME DG Screening Tool, PG&E recommends that (1) streamlined be adopted which has less reliance on pre-existing conditions and (2) allow pre-existing conditions to be fixed only within the Rule 21 process (updated to account for ICA) as needed for more resource efficiency. PG&E’s sees requirements to use iterative methodology and automated approaches that account for pre-existing condition to add an excess rate base burden that does not promote smart incremental ICA adoption.

Load Forecast AssumptionsStakeholders of the WG posed questions about theon assumptions used in load forecasting, including questions on how weather conditions are included. In Demo A, IOUs aligned load allocation methodology with current interconnection practices. [SCS: I think this section should be removed unless stakeholders identify specific concerns or recommendations coming out of the IOU responses.]

Short-term activities of the ICA WG

The ACR outlines seven discrete activities for WG consultation related to Demo A (ACR Section 3.1). The IOUs consulted with the WG on each of these topics in 2016. A summary of those topics, discussions, and recommendations are included below.

3.1.b: Recommend methods for evaluation of hosting capacity for the following resource types: i) DER bundles or portfolios, responding to CAISO dispatch; ii) facilities using smart inverters

With regards to DER bundles or portfolios responding to CAISO dispatch:

19


Include IOU discussion here, or does provided information lead to a further recommendation?





For Demo A, the IOUs explored means of further exploring how hourly limits can differ depending on the DER technology or portfolio, without calculating separate ICA for each DER type. It was concluded that location (distance from substation), rather than temporal variance, may be a more significant driver of ICA variation. The IOUs recommend continuing the use of a technology-agnostic approach in a full system-wide rollout of ICA, rather than making assumptions about CAISO dispatch, or other specific resource types. The WG is in agreement with the continued use of a technology-agnostic approach.

With regards to smart inverters:

It’s envisioned that smart inverters can modify the ICA in two ways: smart inverters can support greater hosting capacity on the system when deployed in aggregate, and projects can use smart inverters to address stated hosting capacity limitations.

Within Demo A, the IOUs began analysis to start understanding the impact of smart inverters on ICA due to their ability to reduce voltage impacts usingdue to their reactive power capabilities. These capabilities were tested on a limited basis using the streamlined method, and were not tested using the iterative method.

The utilities each did a version of a study to determine if the presence of smart inverters with limited capabilities on a distribution feeder would increase the integration capacity. The capabilities tested were a static volt/VAR curve (SCE) and fixed power factor (PG&E and SDG&E). These functions were only assumed to be operational when it did not impact real power output. This “active power priority” results in limited use of the advanced inverter functionality, especially at mid-day hours when the circuit is in need of voltage support. Despite this limitation, the studies found significant increases in integration capacity with the existence of smart inverters.

To balance the need for new methodology study and development before incorporation into ICA software modules, and to align development processes so that consideration of smart inverters may be included into ICA modules as methodology is finalized, inverter standards are finalized through the IEEE process, and smart inverters enter the market, the WG makes the following recommendations:

Additional study, methodology development, and software enhancement to incorporate smart inverters in the near term (with the goal of inclusion into the first system-wide rollout, or next possible iteration of ICA):

a. The WG agrees that additional study of smart inverters is prioritized as a long-term refinement item, and proposes to engage in further studies and methodology development with the goal of including smart inverters into the ICA modules at the soonest possible iteration. For methodology developed during this phase of the ICA WG, the WG will recommend which refinements may be included in the initial system-wide rollout, and which may be included in future iterations.

b. Studies during the long-term refinement period will consider two overarching questions: 1) at what point do smart inverters have an impact on increasing hosting capacity? 2)

20


Can we do in advance of final IEEE standards? Brad Heavner Q





once smart inverters are implemented as common practice, how much will they impact hosting capacity?

c. The WG identified several recommended studies to undertake, including static volt/VAR and fixed power factor functions with consideration of reactive power priority6. Additional studies, building on the initial IOU Demo A study, may be identified.

d. It is understood that CYME and Synergi software have the capability to incorporate smart inverter data, including locations and settings of deployed assets. The IOUs will endeavor to work with software vendors so that smart inverter data can be input as it is collected.

e. Currently, fixed power factor and dynamic volt/VAR are not included in ICA modules as both consensus methodology and automated processes to include these values do not exist. As the WG refines methodology for these two values, the utilities will work with CYME and Synergi to incorporate these values into the ICA module.

f. Additional studies have been recommended by some WG members to further longer-term goals for smart inverter inclusion. These are listed here, and may be revisited as a long-term refinement topic:

i. How changes in grid operations may enable greater DER penetration while remaining within steady state voltage standards.

ii. How ICA may potentially evaluate optimal reactive power to systematically address voltage limits. This may allow developers to locate DERs with advanced inverters at optimal locations.

iii. How ICA may consider dynamic inverter functions, which may include settings to be changed by season, TOU period, weekday vs. weekend use, etc.

3.1.c: Recommend a format for the ICA maps to be consistent and readable to all California stakeholders across the utilities’ service territories with similar data and visual aspects (Color coding, mapping tools, etc.).

The WG discussed ICA map formats in the July WG meeting. The ACR specifies requirements for how ICA results shall be available via utility maps. To reach common fundamental principles guiding the ICA map formats, the joint IOUs presented a proposal for displaying ICA results, including structure of mapping layers (substations, circuits, line segments all visible) and which information will be viewable in map format and which will be included in the downloadable data set.

The following attributes are visible across all three IOU maps: 1) circuit; 2) section ID; 3) voltage (kV); 4) substation; 5) system; 6) customer breakdown percentage (agriculture, commercial, industrial, residential, other); 7) existing generation (MW); 8) queued generation (MW); 9) total generation (MW);

6 New inverter standards under pending changes in IEEE 1547 would require inverters to be sufficiently oversized to create the headroom that enables advanced functionality even when solar panels are at full production levels. A small increase in inverter size can result in a large increase in capacity to produce reactive power. It is therefore likely that active power priority is an excessively conservative assumption.

21


Can SDG&E confirm?





10) ICA with uniform generation (MW); 11) ICA with uniform load (MW); 12) integration capacity of a typical/generic PV system (MW). All map content can be downloaded via ESRI open data.

All IOUs make the following information available via downloadable data set: 1) Demo A final report; 2) ICA Translator; 3) load profiles; 4) customer type breakdown; 5) detailed ICA results by circuit.

Given the amount of data produced in calculating ICA results, the IOUs recommend either limiting future downloadable data to only actionable data with practical applications, or exploring alternative solutions to publishing interactive load profiles besides web-based map applications.

The WG envisions that there may be some differences between the interconnection use case and planning use case with regards to map and dataset needs. These following recommendations should be incorporated in the first system-wide rollout for the interconnection use case:

A stakeholder subset of the WG recommends that the ICA data allow for identification and screening of ICA values by user criteria (such as “list or show all line sections with ICA capacity of 2MW, sorted by size or sorted by substation”). Achieving this goal means the utilities should create a searchable database to ID locations by ICA value and area screens.

i. A stakeholder subset of the WG recommends that the default ICA map display a generic solar PV profile rather than a uniform generation profile. The WG would like to provide a review of utility assumptions for the generic PV profile.

The WG identifies incorporation of single phase circuits as a high priority item for long-term refinements beginning Q1 2017, due to its usage within the interconnection use case. IOUs have indicated that CYME and Synergi models both already have the capability to run analysis on single phase line sections, once methodology is developed (see interim ICA long-term refinements report). The WG recommends that the IOU online maps immediately displayinclude where all single phase line sections are located to enable residential developers to identify which three phase circuits they are connected to, within the first system rollout.

The WG has already identified issues related to data access as an important long-term refinement item, and would like to highlight discussion on providing data in machine readable formats as a discussion topic. Discussion of practical implementation will also highlight cybersecurity concerns.

3.1.d: Evaluate and recommend new methods that may improve the computational efficiency of the ICA tools and process in order to calculate and update ICA values across all circuits in each utility’s service territory in updated ICAs more frequently and accurately.

The IOUs presented on three proposed methods that may be used in ICA to improve computational efficiency at the September and October WG meetings, with the purpose of reducing the number of data points needed to calculate in ICA without reducing the quality of results. These methods focus on 1) hourly reduction and mapping, 2) node filtering, and 3) criteria bounding. Each IOU employed

22


Placeholder language: After 2/15 call, it was suggested that stakeholders review the IOU demo map functionality and then revisit these recommendations





different levels of computational efficiency methods in their demo projects. The WG is in general consensus with regards to the methodology underlying these computational efficiency refinements.

i. Hourly load profile reduction methods analyze fewer loading conditions. For example, an ICA using a 576 hourly profile (which uses minimum and maximum load days for every month, for 12 months – 24 x 2 x 12) may be efficiently reconstructed by reducing the number of hours analyzed with similar loading conditions.

ii. Node filtering methods improve efficiency by limiting the number of nodes analyzed – when nodes are within close proximity to each other with no customer loads in between, or nodes exists only for simulation purposes, those nodes have the same level of ICA due to similar levels of impedance and loading conditions.

iii. Reduction of limitation categories for feeders with a high short circuit duty. For those specific feeders, the voltage fluctuation screens and protection limitation screens do not need to be evaluated, as they will not affect the final ICA value.

It is further identified that, while these computational efficiency measures allowed IOUs to efficiently complete the Demo A projects, another set of questions remain regarding computational efficiency measures that will allow IOUs to perform ICA across their entire distribution systems. Demo A results demonstrated that average times to run ICA can significantly differ depending on whether the streamlined methodology or iterative methodology is being employed. Some of these differences are due to the methodological choices, while others are software and hardware related factors.

While it is expected that continuous improvements to ICA methodology will, over time, lead to additional efficiency and accuracy, full deployment of ICA will lead to a significant number of data points and storage needs, each with cost considerations. [SCS: I drafted the below as a further placeholder that explains where I think the WG is at the moment, since it is seeming unlikely we will get and be able to consider a cost estimate at this point.]

The WG has asked the utilities to prepare a cost estimate framework that would assist the WG, and ultimately the Commission, in evaluating the costs associated with the different methodological and computational components of deploying the ICA. There are a number of different factors that may affect the cost and time associated with running the iterative approach, in particular, and the WG decided that it could not make sufficiently detailed recommendations without understanding the magnitude of the costs of each of the different choices. The need for this detailed estimation did not become fully apparent to the WG until after it had a chance to digest the final Demo A reports and thus there was not sufficient time to complete and consider the analysis prior to filing this report. The WG is thus planning to consider the cost estimates and make further recommendations based upon them at a later date [fill in target date].

While each utility is constrained in the near term with the current capabilities of software and their vendors, the utilities should endeavor to provide estimates that are approximate rather than necessarily

23





reflecting cost/limitations of current power flow modeling software, vendors, and utility processes that may be reasonably expected to change over time.

3.1.e: Evaluate ORA’s recommendation to require establishment of reference circuits and reference use cases for comparative analyses of Demo Project A results. The IOUs were directed to work towards additional consistency between IOUs’ calculation methodology and assumptions, for both the iterative and streamlined approach. To ensure a common approach between IOUs, the IOUs were asked to compare methodologies against reference circuits, for discussion and approval by the WG. This comparative assessment also revealed differences between methodologies when running the streamlined versus iterative methods. Full exploration of these differences are detailed in the separate IOU Final Demo A Reports.

The IOUs used the IEEE 123 test feeder as the reference circuit for comparative analysis as it employs a public data set of power flow results. The IOUs first compared power flow results between the power system analysis tools (PG&E and SCE employ CYME, and SDG&E employs Synergi), and then across ICA results. The WG expresses consensus interest in further testing alignment on a more representative California feeder.

The IOU demo reports include a joint report component concluding that overall, the ICA results do not have significant variation across the IOUs for both the iterative and streamlined methodologies, with the slight variations attributed to how power flow models are treated between CYME and Synergi.

The CPUC Office of Ratepayer Advocates (ORA) include 12 metrics for success for evaluating ICA. ORA provided the WG with a table of these criteria on January 10, 2017, with a brief description of whether the IOUs have met the criteria. The most recent version of the table is provided below:

ORA Criteria SCE SDG&E PG&E Comments1. Accurate and meaningful results

A. Meaningful scenarios ACR scenarios best for use cases?

B. Reasonable technology assumptions

Need plan to incorporate smart inverter data

C. Accurate inputs (i.e. load and DER profiles)

Track 3 Track 3 Track 3

D. Reasonable tests (i.e. voltage flicker)

No concerns/alternatives from working group

E. Reasonable test criteria (i.e. 3% flicker allowed)


24


Should we include a summary here from reports?


Placeholder, potential cost estimates to go into this section





F. Tests and analysis performed consistently using proven tools, or vetted methodology

Tools being developed as part of Demo A and LT refinements

G. Meaningful result metrics provided in useful formats

See #5 See #5 See #5 Duplicative

2. Transparent methodology IOUs have been open to information requests

3. Uniform process that is consistently applied

LT Item LT Item LT Item QA/QC of custom Python scripts TBD

4. Complete coverage of service territory

Not required at this point

5. Useful formats for results6. Consistent with industry, state, and federal standards


7. Accommodates portfolios of DER on one feeder

Uniform Gen map, plus DER translator

8. Reasonable resolution –Spatial Optimal (lower) resolution TBD;

nodal reduction proposal –Temporal Optimal (lower) resolution TBD;

576 vs. 24 hours9. Easy to update based on improved and approved changes in methodology

QA/QC of custom Python scripts TBD

10. Easy to update based on changes in inputs (loads, DER portfolio, DER penetration, circuit changes, assumptions, etc.)

Tweaks to circuit models in CYME/Synergi required for convergence are currently lost when new data from GIS and other data sources is incorporated into power flow circuit model

11. Consistent methodologies across large IOUs

See #3 See #3 See #3 Duplicative

12. Methodology accommodates variations in local distribution system

LegendCriteria met, OK to proceedAdditional work pending as long

25





term refinementDelay full scale circuit modeling until resolved

Is this the right go/nogo objective? If not, what?

The legend describes how close to the IOUs are to meeting the Criteria. Green means that the IOUs have met the criteria, so it is ok to proceed with full scale circuit modeling. Yellow means that these are areas that have been identified as longer term refinement goals. The WG understands that not all of the requirements can or need to be met in order to begin performing the full scale circuit modeling.However, the WG expects the IOUs to meet these criteria as the ICA is refined over time. Red means that the IOUs have not met the criteria, and the IOUs cannot begin full scale circuit modeling until these criteria are met.

The IOUs currently meet or will continue to work on meeting the majority of the criteria. However, the WG believes not all of the IOUs have met criteria 5 and criteria 10. In regards to Criteria 5, while SCE provided access to their mapping tools relatively early on in the process, both SDG&E and PG&E have yet to provide working versions of their ICA mapping tools to the WG. SDG&E has not provided a working link to the WG. PG&E has provided access, but access to data sets behind the map is not available as part of the . Until full access to these tools are provided, the WG cannot review the usefulness of the formats, and therefore cannot recommend that the IOUs have met the criteria.In regards to Criteria 10, updating the models in both CYME and Synergi is difficult. Currently, when new data from GIS or other sources is introduced to the circuit models, both CYME and Synergy lose convergence. Without the ability to accurately update information contained within the models, they have not yet proven themselves for a full circuit rollout. Therefore, the WG recommends the IOUs’ models demonstrate the ability to incorporate changes to their circuit models before a full circuit modeling takes place.

3.1.f: Establish a method for use of Smart Meter and other customer load data to develop more localized load shapes to the extent that is not currently being done.

To develop localized load shapes, IOUs took customer loading profiles derived from smart meter energy consumption data, and aggregated customer-level information up to the distribution transformer level. This distribution transformer level of information was used to disaggregate circuit-level loading profiles. This locational data gives greater confidence in load allocation.

3.1.g: Establish definite timelines for future achievement of ICA milestones, including frequency and process of ICA updates.

Following the completion of Demo A, the IOUs plan to perform final system-wide implementation of ICA. The WG engaged in multiple discussions surrounding expediency around this implementation, given the size and complexity of this project.

The IOUs provide the following comments regarding the implementation of ICA on a system-wide basis:

26


WG members, please review the IOU explanations of timelines to justify 12 month rollout. Still need to determine whether consensus rec on timeline


There was a stakeholder question on how AMI data employed differently - need to list differences here? Does it impact a recommendation here, or is there consensus in how smart meter data is used in ICA methodology?


This has changed though since the last iteration of this chart? PG&E and SDG&E both online at drpwg.org

Brian Goldman, 02/19/17,

Due to the fact that the WG has not yet commented on these inputs, these views are ORA’s own for the moment. If any party is in disagreement with ORA’s conclusions, please include potential revisions in future drafts.





The IOUs understand the urgency of implementing an approved ICA methodology system wide and the IOUs are committed on to implementing the ICA Methodology in the most expeditious manner. The IOUs also ask the stakeholder to understand that what is being asked is a very large and complex project which has not been attempted by any utility. For reference, in Demo A, SCE performed an ICA on 82 distribution feeders and the time allowed for that was 4 months. In the system wide implementation, SCE has to implement ICA on more than 4,500 distribution feeder which exponentially higher in magnitude with significant reduction in time as compared to what was done in Demo A.

While a final decision is pending, the IOUs will continue to work on preparation activities such as preparing network models, data sources, work force plans and implementation procedures. Once the PUC issues a final decision, IOUs anticipate 12 months will be necessary for implementation. More specific details on what IOUs work activities prior to and after the decision are outlined below (applicable to all three IOUs):

Work to commence while a decision is pending: Model creation and validation: SCE engineers to create distribution system models. Activity can

start prior to PUC Final Decision, but it is estimated to last 10-months. Preparation of data sources: Preparation of data sources such as, SCADA Historian, GIS, and

Distribution Management System is required.

Work to commence after decision (12 months): [SCS: it isn’t clear to me if each utility has different estimates or if the notes below should be edited so they speak on behalf f all the IOUs?]

Implement ICA Methodology – SCE estimates 4 months of development once final ICA Methodology is established. Work cannot start prior to PUC final decision as development requires all assumptions and functionality be outlined prior to start of solution design. In addition, based on Demo A work, various iterations of testing are required to stabilize code (e.g., troubleshoot bugs) to render solution production ready. Code will not be stabilized until after various distribution circuit models have been analyzed. Vendor engagement is required.

Run ICA – Perform the ICA on the distribution system models. Based on the ICA Methodology requirements (e.g., number of hours, periodicity of updates) computing resources need to be configured and computing resource management systems may need to be developed. Work with vendor community is required. [SCS: What is the estimate on how long this will take?]

Quality Assurance and Control– In support of ICA Methodology implementation activities, and to support SCE in the publication of most accurate results, quality control and quality assurance systems and processes need to be designed, developed, and implemented. [SCS: What is the estimate on how long this will take?]

Publication of Results – Develop interfaces between ICA Results databases, mapping databases, and other data sources required by the PUC final decision. Edit map symbology to meet ICA

27





requirements.[SCS: What is the estimate on how long this will take?] Note that once PUC makes a decision, some of these activities will need to be completed in parallel in order to complete the project in the 12 months.

Separately, PG&E recommends that the ICA by implemented by June 2018, to coordinate with PG&E’s planning process (currently distribution planning analysis and engineering review occur in the January to May timeframe). PG&E notes that if adopting recommendations of PG&E to use the “streamlined” method for system wide analysis and the “iterative” method on an as requested or pre-application basis, then it is expected that less engineering resources are needed to implement this efficient approach. PG&E projects that if iterative is required for use then a new team would be needed to manage the ICA process. It is projected that if streamlined is adopted and the iterative approach is adopted more efficiently on application basis then the new work load can be more efficiently managed with current engineering resources.

PG&E is also undergoing existing planned work on modifications to the gateway to (1) utilize the new GIS system implemented in 2016, (2) expand gateway to include substation models, and (3) expand ability to have service transformers in the models. If recommendations require the incorporation of planned modifications and automated iterative across the whole system then significant additional work would be required on the gateway and could postpone work to include substation and service transformers. Also, if PG&E’s recommendation of application based used of iterative is not adopted, then more engineering resources would have to be hired and trained in order to perform the regular iterative ICA analysis. Adopting the application based iterative and system wide streamlined recommendation would allow PG&E to more efficiently use existing resources and tool capabilities.

IOU within 12 months, following a proposed decision on final methodology from the PUC. In the interim, the IOUs will continue to evaluate resource requirements necessary to enhance efficiency and scalability of full ICA deployment.

SDG&E?

Long-term refinement activities with potential for inclusion into initial ICA methodology

In December 2016, the WG filed an interim status report on long-term refinement items. This status report includes both the suggested list of topics for further refinement, as outlined in the ACR, as well as additional topics proposed by WG members. It was noted that, as the report was due before IOUs published final Demo A reports, some recommendations for long-term methodology refinement will be adjusted after full review of final demo results and incorporated into the final WG report.

28





3.2.a: Single phase feeders: The ACR identifies expansion of ICA methodology to single phase feeders as a long-term refinement item to begin after the submission of the final ICA WG report. In discussions to date, the IOUs have proposed to evaluate the impact of performing single-phase ICA on at least one circuit in each Demo A DPA. This test evaluation would examine the capability of single phase radials to accept additional loads of DER. This test evaluation should then inform a recommended methodological approach for approaching single phase ICA over an entire distribution system.

- The WG identifies incorporation of single phase circuits as a high priority item for long-term refinements beginning Q1 2017, due to its usage within the interconnection use case. To facilitate the analysis of single phase circuits once methodology is developed, the WG recommends the following refinements to ICA for incorporation into the first system-wide rollout:

o IOUs should further develop their gateway and circuit modeling with the understanding that single phase line sections will eventually be incorporated

o All three IOUs’ maps should visiblye include all single phase line sections to enable residential developers to identify which three-phase circuits they are connected to. Even though ICA values are not yet available for single phase lines, their locations should be shown, and it should be possible to clearly identify what three phase circuit they are connecting to.

- Given that most retail DG (e.g., household level rooftop solar installations) are located on single phase circuits, future iterations of the ICA will provide further insight on the degree to which hosting capacity limits are reached due to wholesale versus retail distributed generation. [SCS: I am not sure if I understand this point? Is this a methodology recommendation?]

Coordination with LNBA process The interim ICA long-term refinement report identified “actively engage to determine how ICA and LNBA may influence each other or be used concurrently.”

- The WG noted that understanding where planned grid upgrades are expected, and having those upgrades indicated in the LNBA online map will also be useful to ICA stakeholders.

Next Steps for the ICA WG

The WG looks forward to continuous improvement and development of additional methodological components for the ICA. The WG identified a tiered list of discussion topics in 2017 in the interim ICA long-term refinement report submitted December 2016. This list of topics has been updated since, and the WG aims to create a proposed working schedule as a priority item once work on long-term refinement items begin.

Topic Rationale ACRTier 1: Higher PriorityOperational flexibility criteria

Development of a non-heuristic approach to operational flexibility based on engineering analysis

Operational flexibility criteria

29





Smart Inverters Studies to facilitate inclusion of smart inverters into ICA at soonest possible rollout

3.1.b

Expansion to single phase feeders

Initial discussions should align with streamlined interconnection discussions, IOUs may start analysis on single circuit on a Tier 2 timeline

3.2.a

Data access Data discussion focused on understanding IT requirements to address market sensitive information, data sharing, and automated data analysis.

3.2.b, 3.2.d

Interactive maps Discussion focused on understanding IT requirements and benefits of increasing data directly visualized onto ICA maps.

3.2.c

Operational flexibility criteria

Development of an improved, less conservative approach to operational flexibility based on engineering analysis

Load forecast assumptions

Leverage work of CPUC DER and load forecast workshop Q1 2017

Tier 2: Lower PriorityIntegration of ICA into growth scenarios for decision making purposes

Refine how growth scenarios are implemented, understand how use cases for growth scenarios may impact ICA results, and make recommendations on incorporation of ICA into growth scenarios. Conversations will occur after February 2017 CPUC workshop on growth scenarios as part of Track 3 efforts.

Comparative analysis

Expansion of comparative analysis to more than one circuit and test more complicated circuits, as IOUs consider expansion of ICA to all circuits within service territories.

3.1.e

Independent validation of ICA

Proceeding in coordination with Track 3 efforts, and following comparative analysis discussion with an agreed-upon dataset and results for parties to compare and validate.

3.2.f

Method for reflecting the effect of potential load modifying resources on integration capacity

Begin development of methodology to include resource reliability and uncertainty factors into ICA, model resource impacts on ICA indicators, assess impacts of load-modifying resources, and include non-grid engineering analysis within ICA methodology

3.2.e

Definition of quality assurance and quality control measures

The WG will determine whether this long-term refinement issue identified in the ACR requires further discussion after reviewing the final Demo A reports

3.2.g

ICA in peak load conditions

Continue discussion of ICA that allows DERs to serve peak load conditions, while maintaining grid stability during low-load conditions

30





31





Appendix

Meeting Date Topic(s)May 12 – 1:00pm-3:00pmWebinar

Opening meeting

May 18 – 10:30am-12:00pmWebinar

Seeking input regarding 1) use of power flow analysis and 2) level of granularity

June 1- 9:00am-3:00pmIn person

First discussion of demonstration implementation plan before June 16th submission

June 9 – 9:00am-3:30pmIn person

Second discussion of demonstration implementation plan before June 16th submission

July 5 – 2:00pm-4:00pmConference call

Call to discuss submission of demonstration implementation plan

July 25 – 9:00am-3:30 pmIn person

Discussion of submitted stakeholder comments on demonstration implementation plansUse cases3.1.c/3.2.c – data and maps3.1.b – portfolio analysis

August 31 – 9:00am – 4:15pmIn person

Use cases3.1.b – smart inverters3.1.f – smart meter/customer load dataData access

September 30 – 9:00am-4:00pmIn person

3.1.e – comparative analysis3.1.b.i – portfolio analysis3.1.d – computational efficiencyData access

October 17 – 9:00 am-4:00pmIn person

Demo A update3.1.d – computational efficiency3.1.f – smart inverters3.1.e – comparative analysis3.1.b.i – DER portfolios3.2.a-g – long-term scoping discussion

November 18 – 9:00am-4:00pmIn person

Review of Working Group short term final report outlineLong-term scoping discussion of 3.2.a-g plus other topicsData

December 13 –webinar

Review of Working Group interim long-term report topics

January 6 – 9:00am – 4:00pmIn person

Review of Final IOU Demo A Reports

January 17 – 9:00am – 4:00pm Review of Final IOU Demo A Reports

32





In personJanuary 20 – 9:00am – 4:00pmIn person

ICA Recommendations

February 2 – 2:00pm-4:00pmWebinar

ICA Recommendations and development of report

February 14- 9:00am – 1:00pmWebinar

ICA Recommendations and development of report

March 1 - Final report due

867711.2

33


€¦ · web view2017. 2. 21. · much capacity of integrated distributed energy resources may be...

Documents