Comment Report Form for WECC-01011
Posting 6
The WECC-0101 Variance Drafting Team (DT) thanks everyone who submitted comments on the proposed documents.
Posting
This document was last posted for a 30-day public comment period from March 10 through April11,2016.
WECC distributed the notice for the posting onMarch 9, 2016. The DT asked stakeholders to provide feedback on the proposed document through a standardized electronic template. WECC received comments fromten entities representing five of the eight Industry Segments, as shown in theWECC Standards Voting Sector Table that follows. One of the ten entities did not provide comments within the required comment window. Although the drafting team did not respond to those specific comments, the issues raised were resident in comments provided by others.[1]
Location of Comments
All comments received on the document can be viewed in their original format on the project page under the “Submit and Review Comments” accordion.
Changes in Response to Comment
In response to NV Energy, the following footnote was added to MOD-026 / E.B.7 to clarify that the language does not specify how the required information must be obtained; rather, it only specifies what must be obtained. The footnote reads as follows:
“This does not require open-circuit testing. The data can be obtained through alternative methods.”
Note: The footnote is attached to E.B.7. The numbering of the footnote may shift when the preamble of the document is removed and the variance attached to the underlying standard. The preamble is not part of the variance and will be stripped away when filed.
The phrase “to the Transmission Planner” was added to the MOD-026 Violation Severity Level table for E.B.4 in column two creating symmetry across the table.
In MOD-027, Attachment 1, row 3, column 2, the left parenthesis around (Requirement 2) was added to close the statement. Also in MOD-027, Attachment 1, row 1, column 2, punctuation was corrected after “350 MVA” from a period to a semi-colon.
Implementation Plan
The Implementation Plan was presented in detailed form in Posting 5 and has been retained there during the development phase of this project.
Action Items
There are no previous action items pending.
Action Plan
On April 19, 2016, the DT met to consider and address comments received during Posting 6 of this project. The only change was non-substantive. The project will be presented for ballot.
Contacts and Appeals
If you feel your comment has been omitted or overlooked, please contact W. Shannon Black, Consultant, WECC Standards Processes, at . In addition, there is a WECC Reliability Standards Appeals Process.
WECC Standards Voting Sector Table
The WECC Standards Voting Sectors are:
1 — Transmission Sector
2 — Generation Sector
3 — Marketers and Brokers Sector
4 — Distribution Sector
5 — System Coordination Sector
6 — End Use Representative Sector
7 — State and Provincial Representatives Sector
8 — Other Non-Registered WECC Members and Participating Stakeholders Sector
Commenter / Organization / WECC Standards Voting Sectors1 / 2 / 3 / 4 / 5 / 6 / 7 / 8
1 / Angela Gaines / Portland General Electric / X / X / X
2 / Michelle Amarantos / Arizona Public Service Company / X / X / X / X / X
3 / Leland McMillan / Talen Montana, LLC / X
4 / Patricia Robertson / BC Hydro / X / X / X / X
5 / Sergio Banuelos / Tri-State Generation and Transmission / X / X / X / X / X
6 / Jess Watkins / NV Energy / X / X
7 / Laura Nelson / Idaho Power / X / X
8 / Diana McMahon / Salt River Project / X / X / X / X / X
9 / Lynda Kupfer / Puget Sound Energy / X / X / X / X / X
Index to Questions, Comments, and Responses
Question
- The Drafting Team welcomes comments on all aspects of the document.
- The Drafting Team welcomes comments on all aspects of the document.
Summary Consideration: / See summary in the preamble of this document.
Commenter / Yes / No / Comment
Portland General Electric / Portland General Electric will submit a No vote, as there is insufficient evidence to merit the more stringent 5-year re-validation of models.
PGE believes the proposed WECC variances will not improve the reliability of the BES to a magnitude that justifies the added burdens of additional costs and allocation of already limited resources.
The drafting team appreciates PGE’s continued involvement in the standards development process.
Sufficiency of Evidence
The DT notes there is a continuing disagreement as to the quantity and quality of technical evidence provided in support of this project. The DT notes that the concerns have been asked and answered in the following postings and would direct Portland General Electric to the DT’s responses in the following documents:
- NERC Posting 1, Response to Comments, Calpine, Question 2/5
- NERC Posting 1, Response to Comments, Talen, Question 4
- NERC Posting 1, Response to Comments, BC Hydro/APS, Question 5
- WECC Posting 1, Response to Comment, Xcel
- WECC Posting 2, Responses to Comments, Calpine, APS, SPR, and Xcel
- WECC Posting 3, Response to Comments, Xcel/Clark/APS/Tri-State, see also Action Plan
- WECC Posting 4, Response to Comments, APS, see also the preamble
- WECC Posting 5, Response to Comments, Tri-State, APS, Colorado Springs Utility.
The DT notes there is a continuing disagreement as to benefit/burden analysis should the variances be approved. The DT notes that the benefit/burden analysis has already been addressed and would direct PGE to the DT’s responses in the following documents:
- NERC Posting 1, Response to Comments, Calpine, Question 4
- WECC Posting 2, Response to Comments, Calpine
- WECC Posting 3, Response to Comments, Tri-State
- WECC Posting 4, Response to Comments, PacifiCorp
- WECC Posting 5, Response to Comments, Tri-State
“Unlike other NERC regions, WECC has more than 15 years of experience of successfully implementing generator validation procedures. The quality of the WECC dynamic database was highlighted in the FERC/NERC September 8, 2011 outage report [15] and a FERC LBNL report on frequency metrics [6].”
Western Electricity Coordinating Council, Modeling and Validation Work Group Recommendations on WECC SAR-0101 and Power Plant Modeling Standards, May 30,2014 (AKA: Technical Paper 1) at page 4.
Arizona Public Service Company / While AZPS maintains our belief that the variances are not warranted for the reasons previously articulated in our prior filed comments, AZPS will confine our present remarks to concerns not previously raised.
MOD-027-1 E.B.1.3 and E.B.2.1 state that the minimum sampling rate is 10 times per second; however, many entities utilize data recording devices that are configured to and have an upper threshold to sample at no more than once every four seconds. Moreover, this data is communicated to EMS, which is configured to acquire and receive data at a rate of once every 2 seconds, and will need to be reconfigured to receive and read the data at the increased proscribed rate. Thus, to meet the rigor of the variance, many entities will be forced to replace equipment at each site and make changes to existing installed EMS. AZPS requests this prescriptive sampling rate be revisited or a technical justification be provided, with an explanation of how the more frequent sampling rate results in an increase in reliability commensurate with this sort of fleet-wide replacement.
Also, MOD-027-1 E.B.1.2 states that the "Measured power output and measured frequency shall both be recorded at either the point of interconnection or the generator's terminals." Historically, system frequency data is obtained from EMS which is not recorded at either the point of interconnection or the generator's terminals. AZPS believes E.B.1.2 should be removed or edited to account for EMS measured frequency data; otherwise, costly equipment installations will be required for many entities at each unit.
Sampling
The DT notes that APS’s concerns were previously asked and answered. The perceived value of the additional data will continue to be entity specific as is the benefit/burden analysis of the sampling rate. The DT would direct APS to the DT’s responses in the following documents, with the added caveat that the sampling rate was discussed in depth during multiple DT meetings:
- WECC Posting 2, Response to Comments, Idaho Power/SRP/PacifiCorp/PPL/Xcel
- WECC Posting 2, Response to Comments, see also Calpine on the related costs concern
- WECC Posting 3, Response to Comments, Xcel, see also APS regarding concerns over unnecessary burden
Model Validation and Sampling Rates
To establish a new model, or verify the dynamic characteristics of an existing model, the data used must contain sufficient information to prove the model’s dynamic abilities. In addition, the measured and simulated data must be of sufficient resolution in time and magnitude to observe the dynamic behavior.
It is generally understood that to accurately digitize a signal, it must be sampled at a rate thatexceeds the signal frequency by more than a factor of two (the Nyquist rate). In the field of system identification, a minimum sampling rate of five times the Nyquist rate is often used as the rule of thumb to correctly construct the data for the desired range or, to state another way, more than 10 times the highest frequency present in the measurement [1].
The dynamic range of a governor control system varies widely, but normally extends from below 0.01 Hz to above 1.0 Hz, although the response to the upper range of frequencies is attenuated as frequency increases. Governor models are designed for fidelity in frequency range to at least 1.0 Hz. This is necessary to reproduce the governor’s response to frequency transients and inter-area oscillations, which can range from 0.1 Hz to 1.0 Hz. Therefore, to properly observe governor response, data sampled at a rate of 10 Hz or higher should be used. This minimum sampling rate is consistent with frequency response measurement requirements in the grid codes of other countries [2].
For model development, in addition to using a high sampling rate for measured data, it is necessary to use a perturbation input that consists of a range of frequencies that exceeds the dynamic range of the model. To meet these requirements, staged tests for model development or validation are performed with inputs consisting of step changes, impulses, or a series of individual sine waves. Digitized measurements are normally sampled at rates far exceeding 10 Hz.
For revalidation of an existing governor model using disturbance data, the typical event will be a sudden change in system frequency due to a large generation trip. This disturbance will normally stimulate system inter-area modes as well as generator local modes of oscillation that range from 1-to-2 Hz or higher. Disturbance event monitoring equipment, such as phasor measurement units (PMU), sample data with sufficient precision and sampling rate suitable for validation of models; i.e., in excess of 20 Hz.
If there is any oscillation in the electrical power signal, which is used as the validation measurement, it will be aliased using a lower sampling rate, as shown below. The measurement, sampled at a rate of 1 Hz, leads to an incorrect estimate of the governor response and will not validate the model. Also note in this case that the governor output is oscillating at a frequency near 1.0 Hz, so validation of the governor model can only be achieved with data sampled at a frequency greater than 10 Hz.
MOD-027-1 E.B.1.2
The DT has opted to retain the language because of the granularity of the sampling rate. The locational aspect of the requirement is included to standardize the value of the data requested. Another measurement nearby may or may not be sufficient; however, the requirement seeks to eliminate the variables. The need to standardize the data is seen in the recent-years degradation of the model.
The DT recognizes that the stricter approach may result in increased costs. For source referencing of cost-related discussions, please see Response to Comments to PGE (above) in this posting.
Talen / Talen Montana, LLC (TALN) appreciates the opportunity to comment. Please accept the following as suggestions for improvement:
1. Recommended Changes to WECC-0101 MOD-026:
a. Accept OEM-calculated rotational inertia data (H values), due to being better than trip test-derived results (i.e. based on manufacturing drawings, and not subject to frictional effects or measurement uncertainty). This is a more restrictive criterion than the NERC standard, taking "restrictive" in the present context to mean "more accurate." Trip testing also subjects equipment tomeaningful wear and tear, thereby degrading its reliability, which would be an inappropriate outcome for a reliability standard.
b. Rephrase footnote 2 for clarity. It appears from the text of the variance that it should present a list of typical < 180-day changes, but it commingles permanent items (e.g. new AVR) as well as short-term ones (shifting between voltage and PF control).
c. State that any WECC-approved model type is acceptable (another clarification)
d. Change the revalidation interval to 10 years. TALN does not agree that requiring model validation every 5 years, rather than the 10 years required by the NERC standard, will lead to significant reliability benefits.
2. Recommended Changes to WECC-0101 MOD-027:
a. Require that a representation of boiler dynamics be included in fossil-unit models fossil-unit models. Such equipment responds to disturbances by opening the HP turbine control valves, creating an imbalance between steam-make and steam-take, until the boiler can ramp-up. Many governor models are based on an "infinite boiler" assumption; and while the NG pipelines feeding gas turbines and the lakes feeding hydro units may indeed be infinite for the purposes of disturbance response, the drop in HP steam pressure for fossil units must soon (within seconds) be counteracted or the unit may be destabilized or even trip. This would stand as a clarification to the NERC standard, informing GOs up-front as to the pass/fail criteria that will be applied.
b. Footnote 2 of the MOD-027 variance should be rephrased for clarity. It appears from the text of the variance that it should present a list of typical < 180-day changes, but it commingles permanent items as well as short-term ones. Also change, "going from droop control to constant MW control," to, "inhibiting or bypassing the droop response." It is normal for unit can be running at a MW setpoint and still have droop response enabled.
c. State that any WECC-approved model type is acceptable (another clarification)
d. Change the revalidation interval to 10 years. TALN does not agree that requiring model validation every 5 years, rather than the 10 years required by the NERC standard, will lead to significant reliability benefits.
The DT appreciates Talen’s continued participation in the standards development process. The DT notes that Talen submitted the same comments in NERC Posting 1, Question 4. The comment was asked and answered in the forum.
BC Hydro / BC Hydro appreciates that this Posting for Comment is only for new substantive issues. However, BC Hydro does not feel that previous comments submitted have been fully addressed.
1. BC Hydro does not believe that the WECC MOD-026 and -027 regional variance standards to the approved NERC MOD-026 and MOD-027 standards are necessary. From Posting 4, it was stated that the models have improved under the WECC Policy. However, the Posting also goes on to state that WECC Policies are voluntary and unenforceable. If WECC members have been following the Policy voluntarily to date, why does the drafting team and working group feel that the NERC MOD-026 and -027 standards would mean entities would stop following the Policy?
2. Revalidation every 5 years. Now that the models are improved, and a significant amount of entities have digital relays versus analog, why is the shorter time frame proposed? BC Hydro has been performing WECC testing (i.e. model validation) since the late 1990's and re-validation test results have not identified any significant reliability risks. As such, there does not appear to be any measurable technical benefit to perform the testing more frequently.
Why is the Variance Necessary if already Covered in a Policy?
As BC Hydro points out, modeling under the WECC Generator Validation Policy (WGVP) has greatly improved and, in fact, has been lauded by NERC/FERC. However, the WGVP remains voluntary and not every entity abides by the WGVP.
To clarify the enforcement component it may help to understand the WECC Document Categorization that impacts the WGVP. At the threshold, the WGVP is incorrectly named. It is not a policy at all as defined in the WECC Document Categorization Policy. Rather, the WGVP is categorized as a WECC Guideline. Unlike either a WECC Criterion that requires adherence or a WECC Regional Reliability Standard requiring compliance, a WECC Guideline (the WGVP) is purely advisory in nature. Because it is advisory, entities that do not currently follow the WGVP have no incentive to adopt the higher performance threshold contained therein. As is indicated in many of the comments received, if an entity is not currently abiding by the WGVP adoption of the higher performance standard may result in additional burden.
The enforcement component offered with a Regional Reliability Standard (RRS) provides incentive for all applicable entities to abide by the more stringent standard. Further, inclusion of the specifics in the proposed project will further refine the data reported and; thereby, the accuracy of the associated model.