ICA Working Group Meetingjuly 25, 2016

ICA Working Group Meetingjuly 25, 2016

ICA Working Group MeetingJuly 25, 2016

More Than Smart - Oakland, CA

Discussion of Stakeholder Comments

Stakeholders of the ICA Working Group provided comments on the Demonstration A implementation plans submitted by utilities. These comments were discussed at the July 25th in-person meeting of the ICA Working Group and summarized below. Stakeholder comments are taken verbatim from submissions.

DPA Site Selection:

“The location of Demo A projects should be discussed to ensure the ACR direction for “as broad a range as possible of electric characteristics” is met. PG&E provided a thorough discussion of its criteria and a presentation of how the two DPAs compared to the total territory for each criteria. The WG could ask for similar details from SDG&E and SCE before supporting their proposed locations. In addition, a cursory review of population data shows that over half of Californians live in dense metro centers near the coast (LA, San Francisco Bay/ San Jose, San Diego) which could have shorter circuits and less load variability than inland locations. Of the three utilities, only SCE includes a truly urban area near the coast (Johanna DPA). The WG should discuss potential shortcomings of running Demo A on circuits that could have low IC values relative to substation capacities based on the DPA selected. If addition, it is not clear what circuit/load conditions will trigger ICA protection and safety, and whether these conditions exist in the current DPA selections.”

Discussion summary and recommendations:

Circuits for comparative analysis have to have a broad set of analysis, but the electrical circuits do no need to be completely representative.

Stakeholders had questions about:

  • Total service areas in DPAs, how growth might evolve, and how growth is used as input into ICA methodology
  • How much variation is there within existing DG across DPAs (solar PV, EVs, storage, etc.)
  • IOU response that there is currently not a lot of diversity because there is not a lot of DG out there, but they are interested in looking at in subsequent DRP demos

Modeling software:

“PG&E and SDG&E will use LoadSEER for load modeling and forecasting in the ICA. SCE states, "SCE has tested the LoadSEER software package from Integral Analytics and determined that the program’s current functionality does not adequately meet the forecasting needs of the SCE distribution planning process.” “

What are the unique needs of SCE’s distribution planning process that make LoadSEER inadequate?

Discussion summary and recommendations:

Stakeholder comments:

  • WG stakeholder issues can be broken into short term and long term considerations:
  • Short term: engineering methodology (what are we getting to, future enhancements over time, etc.)
  • Long term: model for what is publicly available (is there a single model used by IOUs that others can use openly?)
  • Comparative assessment based on shared IEEE circuits
  • EPRI has already done the analysis for detailed v. streamlined across circuits so it would be a good data point to see how it compares, so WG wouldn’t have to go to EPRI to do it again
  • EPRI feeders aren’t publicly available, there is specific data that can’t be publicly shared with WG (would need proper NDA to access)
  • WG also cannot access formal model process to understand the assumptions
  • CSI data is publicly available, IOU express agreement to pull out data from EPRI for broad consumption and have Jeff (EPRI) follow up for what information EPRI can provide at no-cost
  • Stakeholders interested in having IOU look at full suite of feeders
  • IOUs have question on whether those are indicative of feeders in CA
  • Comment that there are other models out there besides EPRI, including Stanford Center (DOE funded) and Kevala’s private platform, as examples
  • Affirmed importance that same assumptions, equations, and methodologies are used in comparative results
  • Working Group stakeholders overall do not think IEEE feeder adequately models the system, and would like to see how IOUs did the analysis and show how the model does relative to others
  • Discussed that interim report (Nov 3) should include assumptions going into the analysis
  • IOU comparison: each utility would do 6 circuits total (3 of own, 3 of others)
  • WG expressed interested that a common iterative approach is needed in the fall season
  • Data remains a big barrier
  • Issues teed up for longer-term recommendations of WG:
  • How will regulator guide IOUs toward common platform?
  • More dynamic open platform?
  • Dynamic modeling capability later on?
  • Action items:
  • IOUs to do additional IEEE circuit
  • Reach out to EPRI with clarity of what’s publicly available data for CSI methodology

Methodology:

“This was the first item in the ICA WG recommendations from June 1. ORA envisioned this would be the joint utility filing that compared the methodologies per ORA’s March 3, 2016 post-workshop comments (p. A-2.), but the Demo A plans do not provide a comparison across the IOUs. While more information has been added to the revised plans, key details are still missing:

• Spatial resolution of analysis vs. reporting,

• When and how is short-circuit analysis performed relative to power flow?

• Loads and analysis for 2, 576, or 8760 hours?

• How does the streamlined method for each IOU compare to the others and to the EPRI method? PG&E’s Figure 4 is nearly identical to SCE Figure 2, but the flow to/from the “ICA Calculations” is different,

• Per the May 2 ACR, description of ICA criteria and how they are applied will not be provided until the interim report due 3Q2016;

• Even if the benchmark/comparison assessment shows results to be comparable for a set of reference circuits, the details of each methodology should be transparent to help understand how they will be applied to the full set of circuits.”

Discussion summary and recommendations:

• Need further discussion into understanding how ICA and transmission work down to the distribution level:

• What are the implications for interconnection process

• How to get transmission process analysis into this?

• Intent is long-term focus of WG: interested in including whole system analysis, hosting capacity in transmission planning to distribution level, how the processes meet and coordination of processes

• Action items: coordinate further discussion with ISO

Methodology:

“As shown on the tables included in each of the utility’s plans regarding the “power system criteria to evaluate capacity limits”, there is some difference remaining between what criteria each utility intends to include in the Demonstration A analysis. In particular, under the category of Safety/Reliability, PG&E has indicated their intent to evaluate Islanding, Transmission Penetration, and Operational Flexibility (see PG&E at A-16). SCE and SDG&E, however, indicated that they do not intend to include Islanding, and SCE also is not planning to address Transmission Penetration at this time (see SCE at 10, SDG&E at 10). It would be helpful to understand the reasons for the differences in approach and the potential implications this will have on the Demonstration Projects, the comparability of the results, and their applicability to future efforts.”

Discussion summary and recommendations:

Stakeholders agree that further discussion is needed. There is additional interest in fundamental restrictions/DER in abnormal conditions: changes in Rule 21, changes with adoption of standards, overall islanding/protection scenarios need to be studied within the analysis (who is testing/publishing, results and level of added risk)

  • ICA may help inform Rule 21 so that data will inform where load is and make interconnection process more efficient

Methodology:

“Each of the IOU plans discussed some of the content to be included in the intermediate report to be released at the end of Q3 2016. The IOUs highlight that this report will include more details and discussion on the limitation criteria used. Encourage discussion of these criteria in the working group prior to this report being released. This process will be beneficial to ensure alignment between the IOUs and allow for transparency to the working group. As an example, the IOUs are not aligned on the safety limitation criteria to include in the ICA determination. Alignment of the limitation criteria may be desirable overall, but will be necessary for comparing methods.”

Discussion summary and recommendations:

See above discussion on islanding, safety criteria, etc.

Data and Maps:

“Presenting/displaying ICA results for multiple scenarios of capacity (no reverse power flow to transmission system, maximum capacity regardless of reverse power) and growth (2-year growth, growth scenarios I and III). How do the IOUs intend to report/display these different combinations of results? “

Discussion summary and recommendations:

  • Demo A results should include reverse power flow
  • Different growth scenarios included in mapping required by ACR
  • Overall questions answered by IOU-submitted Demo A mapping proposal (available online for stakeholder comment)

Comparative assessment:

“There is a need for a plan to compare the streamlined and iterative results and computation times, both to each other and across IOUs. It will be important to discuss the collection of test feeders to use and the method for comparison. A rigorous comparison of results and computation time will help inform future decisions for evaluating ICA system wide.”

Discussion summary and recommendations:

Want robust and clear analysis for consistency. Addressed in modeling software discussion above.

Benchmarking and comparative assessment:

“This analysis was critical to WG support for the Demo A proposals, and is a key learning objective. However, the current plans do not provide enough detail to determine if the testing will yield meaningful results. For example, PG&E refers to an ERPI project with 16 feeders (p. A-33), SCE clearly states that six circuits will be used (p.22), and SDG&E did not state a number of circuits. This topic resulted in an animated discussion at the last meeting, particularly in terms of testing between methods vs. IOUs, and it remains an open issue.”

Discussion summary and recommendations:

Agreed that continued discussion of comparative analysis is needed.

Validation of results:

“The ruling stated that both Energy Division and ORA would be responsible for viewing and validating the inputs, models, limit criteria and results. Discuss the limitations (if any) on market participants’ ability to review this content and the conditions under which this will be possible. Working Group stakeholders interested in participating at this level of detail.”

Discussion summary and recommendations:

  • Issues around providing information in CSI, IEEE is public information and their inputs are non-customer load input
  • What are the inputs, and will they be transparent?
  • On the CSI circuit, individual customer sign off is prohibited because there are thousands of customers
  • WG members would like to get initial ideas of what are performance attributes

Reference circuits:

“PG&E says it is "willing to explore using the circuits in the demo and anonymizing them or EPRI California Solar Initiative circuits that were representative of the three IOUs”

SCE says it will “work with other IOUs to apply both methods on six reference circuits (two circuits from each IOU’s demo A study with anonymization)”

SDG&E says “SDG&E and the other IOUs will leverage the use of test circuits to perform the ICA and benchmark against each other.””

Discussion summary and recommendations:

See earlier discussion regarding modeling software.

Use cases:

“Stakeholders agree that there is need for WG agreement on ICA use cases to help inform discussions about temporal and spatial granularity, result formats, and frequency of updates.”

Discussion summary and recommendations:

  • Use cases need spatial and temporal aspects
  • Question of how scenarios are developed?
  • How do DER adoption patterns affect what hosting capacity might be?
  • How do you account for adding future hosting capacity in areas where it’s not needed now but might be needed later?
  • Agree that the use case discussion needs to help understand who is using each case (users), locational granularity (where on the system, when, frequency, level of accuracy)
  • Agree to change “procurement” to “sourcing, pricing, and programs”
  • Interest in understanding fast track process, IOUs may remove 3 minute limit for fast track
  • Want to ensure that for now, ICA discussion stays separate from LNBA, while acknowledging linkage and how ICA can inform LNBA results, like overlay of costs, hosting capacity, etc.
  • Overall consensus for PG&E slides on planning use case, except for last bullet on grid needs
  • Overall discussion of use cases further detailed in a draft document (to be summarized by MTS)

Other – learning objectives:

“Currently, each plan includes the same eight “learning objectives” (SCE adds two additional objectives) that are generally reiterations of the CPUC requirements. Requirements are not the same as learning objectives which are often in the form of questions to be answered through the proposed projects. For example, a key objective for all IOUs could be “Find the best ICA methodology and granularity of inputs, analysis, and results.” However, the learning objectives should be different for each IOU given that SCE and SDG&E are starting from scratch regarding the streamlined analysis, and PG&E has already applied this method across its entire service territory. Examples of utility specific objectives could be:

• SDG&E – Can SDG&E implement PG&E’s method, or improve on it, given SDG&E’s smaller service territory and computational needs?

• SCE –Will the iterative method work on SCE’s large territory?

• PG&E – What it required to extend the ICA beyond the substation bus and service transformer?

Overall, learning objectives should provide concrete objectives that can be used on an ex poste basis to determine if the Demo projects have been successful. “

Discussion summary and recommendations:

  • The above bullets should amend comparative assessment discussion
  • Load forecasting includes integration of all the tools, project development, etc.; forecasting becomes part of the planning process
  • Interest in smart inverter technologies
  • How do demo A and demo B speak to each other and potential of analyzing “fatal flaw scenario”
  • Relationship between two methods and supplementing/replacing Fast Track results

Other – schedule:

“Schedules for Demo A should provide details and dependencies such that the WG can understand the scope and timing of individual tasks. This will help the WG prioritize topics for future meetings. The schedules should also provide time for methodological adjustments should the initial comparison/benchmark tests show accuracy limits for the streamlined method or computational limits for the iterative method. Also, milestones for related projects, e.g. EPIC 2.23, should be shown. “

Top View