Stan Holland Senior Engineer Nick Hatton Staff Engineer Overview - - PowerPoint PPT Presentation

Year 10 Task Force Report to Studies Subcommittee

January 15, 2020

Stan Holland Senior Engineer Nick Hatton Staff Engineer

Overview

• Purpose of Year 10 assessment
• Data Sources
• Tools and Approach
• Results
• Conclusions
• Recommendations

2

Purpose

• Determine if there are any reliability risks in

the 2028 Anchor Data Set, specifically with-

• Resource adequacy
• System stability
• Transmission congestion
• Analysis to be done using a Power Flow (PF)

model and a Production Cost Model (PCM)

• Prepare data summaries and tools that other

assessment teams can use at their discretion for baseline comparisons

3

Data Sources

• 2028 ADS PF version 1.0 (PF and Dynamics)
• Same as the 2028 Heavy Summer 1A base case
• Includes comments from Western planning groups
• 2028 ADS PCM Phase 1 version 2.2
• Maintains same topology and resources as the 2028

ADS PF version 1.0 (imported into PCM tool)

• Other data added to facilitate an hourly commitment

and dispatch

• 2028 ADS PCM Phase 2 version 2.0
• An update of the phase 1 case that includes

additional generation retirements and replacements

• Not used in Year 10 assessment

4

Tools and Approach

• Stability assessment used GE PSLF
• 7 standard disturbances
• Post transient analysis of two unit Palo Verde Outage
• Resource adequacy assessment used GE PSLF

and ABB GridView

• Voltage violations
• Frequency response
• Unserved load
• Transmission congestion assessment used GE

PSLF and ABB GridView

• Branch and transformer overloads
• Branch and path congestion

5

Stability Results

Event Load Tripped (MW) Gen Tripped (MW) Other

Chief Joe Brake 2-Unit Palo Verde outage 2,747

FR: 1881 MW/0.10 Hz

Colorado River Red Bluff 250 Gates-Midway | Diablo-Midway 1,800 Brownlee-Hells Canyon 3 120

FR: 1682 MW/0.10 Hz

Daniel Park-Comanche 235 Pacific DC Intertie block 96

6

FR (Frequency Response): If generation was tripped in an outage, the frequency response metric was calculated for that disturbance. NERC uses this metric to calculate the Interconnection Frequency Response Obligation (IFRO) for an entire interconnection. The IFRO for the Western Interconnection is typically around 880 MW/.10 Hz. Details on the IFRO can be found here.

7 Note: The disturbance modeled here reflects generator governor droop limitations, so a return to 60 Hz is not expected in this plot.

8

Post Transient Results

• 145 buses with a voltage change greater than 5%. All

were in 5-to-6% range, which is considered acceptable when building a base case.

• Five branches and eight transformers were overloaded

in the pre-analysis case.

• Nine branches and 34 transformers were overloaded in

the post transient case. The highest overload not present in the pre-analysis base case was 128.6%.

9

Resource Adequacy Results

10

The only area to have unserved load in the PCM was CFE for three hours on July 19th. This chart shows that there was no additional import capacity from California available during the “event”.

Unserved Load for CFE in PCM

11

The CFE unserved load was caused by the loss of two large combined cycle units. Local CT and steam units responded, as did units in California; however a small amount of load was ultimately unserved.

500 1,000 1,500 2,000 2,500 3,000 3,500 4,000 7/14/2028 7/15/2028 7/16/2028 7/17/2028 7/18/2028 7/19/2028 7/20/2028 7/21/2028 7/22/2028 7/23/2028

CFE Load/Gen Balance Snapshot - 2028 ADS Ph1 v2.2 19-11-25

DG/DR/EE Other Combustion Turbine Steam - Other Combined Cycle Small Hydro RPS Biomass RPS Solar Wind Geothermal Steam - Coal Hydro+ES Nuclear Demand Dump MW

Friday

Change in Generation

20,000 40,000 60,000 80,000 100,000 120,000

Coal Gas Hydro Nuclear Solar Wind Other Renewable Other Thermal

Generation Capacity Comparison (MW)

2017 Actual 2028 ADS_P1v2.2 2028 ADS_P2v2.0

12

Retire Diablo Add Antelope

Transmission Congestion Results

• The term “congestion” commonly refers to a branch
• r path that is consistently near it’s limits, either by

design or due to changes and assumptions.

• WECC often gauges congestion by a set of

utilization metrics, where high utilization is:

• U75 = Flow at 75% or more of the rating for 50% or more
• f the hours, or
• U90 = Flow at 90% or more of the rating for 20% or more
• f the hours, or
• U99 = Flow at 99% or more of the rating for 5% or more
• f the hours.

13

14

0% 20% 40% 60% 80% 100% 120% Percent of Hours

Most Heavily Utilized Paths - 28ADS_PCM_V2.2 2028 ADS PCM V2.2

U75 U90 U99

Path 83 MATL is the only WECC path that exceeded the U99 metric. The modeling of a carbon tax in Alberta may be a factor in the high south to north flows.

Conclusions (Power Flow) (1 of 2)

• Disturbances that included generation drop showed a

strong generation response.

• The Gates-Midway and Diablo-Midway outage

identified a large amount of load tripped.

• Still under investigation
• Please see report for additional discussion.
• Four voltage criteria violations were seen in the PDCI

block outage at load buses in the Northwest.

• PDCI block has associated RAS that was not used
• Please see report for additional discussion.

15

Conclusions (Power Flow) (2 of 2)

• The post transient analysis had some overloads
• Overloads could be indicative of a system deficiency or a

modeling problem.

• Some entities include overloads in the Year 10 base case to

illustrate the need for additional, yet unfunded or unapproved system enhancements

• Corrective action for these overloads may have already been

identified, so they wound not be issues that need further resolution

16

Conclusions (PCM)

• The generation from the imported power flow

was sufficient to meet the 1-in-2 load forecasts except for 3 hours in CFE.

• The increase of solar and wind in California led

to an increase in the amount of curtailed (spilled) generation.

• The shift from coal-fired to renewable

generation decreased the CO2 emissions

17

Recommendations

• This study should be used as a baseline
• Ensure voltage support and cascading

potential are included in future studies.

• DG was not included in the Power Flow case.

Future studies will include DG and results may vary based on it’s inclusion.

18

Contact:

Chair Shilpa Toppo Liaisons Dick Simons Nick Hatton Stan Holland

19