20
2019 Deliverability Study Report October 1, 2019 SPP Resource Adequacy

R deliverability st… · Web viewThe deliverable percentage has increased every study year for each fuel type except coal, natural gas, and wind. Since the 2016 Deliverability Study,

  • Upload
    others

  • View
    0

  • Download
    0

Embed Size (px)

Citation preview

Page 1: R deliverability st… · Web viewThe deliverable percentage has increased every study year for each fuel type except coal, natural gas, and wind. Since the 2016 Deliverability Study,

2019 De l iverab i l i ty S tudy Repor t

October 1, 2019

SPP Resource Adequacy

Page 2: R deliverability st… · Web viewThe deliverable percentage has increased every study year for each fuel type except coal, natural gas, and wind. Since the 2016 Deliverability Study,

Southwest Power Pool, Inc.

REVISION HISTORY

DATE OR VERSION NUMBER

AUTHOR CHANGE DESCRIPTION COMMENTS

9/23/2019 SPP Staff Initial Draft

i

Page 3: R deliverability st… · Web viewThe deliverable percentage has increased every study year for each fuel type except coal, natural gas, and wind. Since the 2016 Deliverability Study,

Southwest Power Pool, Inc.

CONTENTS

Revision History...............................................................................................................................i

1. Executive Summary.................................................................................................................5

1.1 Studied Capacity..........................................................................................................5

1.2 Results.........................................................................................................................6

1.3 Acknowledgements.....................................................................................................6

2. Introduction..............................................................................................................................7

2.1. Objective......................................................................................................................7

2.2. Applicability.................................................................................................................7

2.3. Study Process...............................................................................................................7

3. Study Inputs and Assumptions.................................................................................................9

3.1. Software...................................................................................................................... 9

3.2. SPP Balancing Authroity Area......................................................................................9

3.3. Base Models and Topology........................................................................................10

3.4. Generator Grouping...................................................................................................10

3.5. Transfer File...............................................................................................................10

3.6. Monitored and Contingent Elements........................................................................10

3.7. Excluded Contingencies.............................................................................................11

3.8. Other Study Assumptions..........................................................................................11

4. Results....................................................................................................................................12

4.1. 2019 Deliverability Study Results...............................................................................12

4.2. Trending Results........................................................................................................12

4.3. Conclusion................................................................................................................. 15

Appendix D: List of Acronyms.....................................................................................................16

2

Page 4: R deliverability st… · Web viewThe deliverable percentage has increased every study year for each fuel type except coal, natural gas, and wind. Since the 2016 Deliverability Study,

Southwest Power Pool, Inc.

LIST OF FIGURES

Figure 1: Chart representation of deliverable, committed, and available capacity...........................................................................................................9Figure 2: Regional map indicating the SPP BA boundary...............................11Figure 3: Trending Deliverability Study results for Coal, Natural Gas, and Wind resources..............................................................................................15Figure 4: Trending Deliverability Study results for Fuel Oil, Nuclear, and Hydro resources.............................................................................................16Figure 5: Trending Deliverability Study results for Ag. Byproducts, Municipal Waste, Solar, and other fuel type resources..................................................16

3

Page 5: R deliverability st… · Web viewThe deliverable percentage has increased every study year for each fuel type except coal, natural gas, and wind. Since the 2016 Deliverability Study,

Southwest Power Pool, Inc.

LIST OF TABLES

Table 1: SPP registered capacity by fuel type..................................................5Table 2: 2019 Deliverability Study Results by fuel type..................................6Table 4: 2019 Deliverability Study Results by fuel type................................12Table 6: Deliverable Capacity from the previous three studies.....................13

4

Page 6: R deliverability st… · Web viewThe deliverable percentage has increased every study year for each fuel type except coal, natural gas, and wind. Since the 2016 Deliverability Study,

1. EXECUTIVE SUMMARY

In accordance with Attachment AA1 of the SPP Open Access Transmission Tariff (OATT), Southwest Power Pool (SPP) is required to perform an annual Deliverability Study determining the available Deliverable Capacity of resources registered in the SPP Integrated Marketplace. The Deliverability Study determines how much of each resource’s capacity is deliverable to the SPP Balancing Authority Area only, specific delivery points or zones within SPP are not studied. The results of the 2019 Deliverability Study are valid for the upcoming Summer Season (2020) and the subsequent Summer Season (2021). Deliverable Capacity purchases allow a Load Responsible Entity (LRE) to satisfy the Planning Reserve Margin (PRM) portion of its Resource Adequacy Requirement without obtaining firm transmission service.

1.1 STUDIED CAPACITYThe capacity configuration by fuel type for the SPP Balancing Authority (BA) is shown below in Table 1.

Table 1: SPP registered capacity by fuel type

Fuel TypeMax Capacity2

(MW)Agricultural Byproducts 21

Coal 23,774Fuel Oil 1,343Hydro 3468

Municipal Solid Waste 13Natural Gas 35,583

Nuclear 2,115Other Fuel 63

Solar 215Wind 21,290

Total Capacity 87,885

1 Attachment AA (Section 10): https://www.spp.org/documents/58597/attachment%20aa.pdf2 Registered capacity as of September 1, 2019.

5

Page 7: R deliverability st… · Web viewThe deliverable percentage has increased every study year for each fuel type except coal, natural gas, and wind. Since the 2016 Deliverability Study,

1.2 RESULTSAfter performing the analysis, using TARA software, it was determined the total Deliverable Capacity was 83,509 MW and the deliverability percentage was 95.0% with the assumptions outlined for the study. Table 2 shows the 2019 Deliverability Study Results by fuel type.

Table 2: 2019 Deliverability Study Results by fuel type

Fuel TypeDeliverable

Capacity (MW)Max Capacity

(MW)Percent Deliverable

(Deliverable Capacity /Max Capacity)

Agricultural Byproducts 21 21 100.0%Coal 23,632 23,774 99.4%

Fuel Oil 1,153 1,343 85.9%Hydro 3,423 3,468 98.7%

Municipal Solid Waste 13 13 100.0%Natural Gas 34,015 35,583 95.6%

Nuclear 2,115 2,115 100.0%Other Fuel 61 63 98.2%

Solar 187 215 87.1%Wind 18,889 21,290 88.7%

Total Capacity 83,509 87,885 95.0%

1.3 ACKNOWLEDGEMENTSThe stakeholder review process was an integral part in this study, and SPP staff appreciates the participation and oversight of the Supply Adequacy Working Group (SAWG), Operational Reliability Working Group (ORWG), and Transmission Working Group (TWG).

6

Page 8: R deliverability st… · Web viewThe deliverable percentage has increased every study year for each fuel type except coal, natural gas, and wind. Since the 2016 Deliverability Study,

2. INTRODUCTION

2.1. OBJECTIVE The objective of the annual Deliverability Study is to evaluate and determine the available Deliverable Capacity of resources registered in the SPP Integrated Marketplace to the SPP BA Area.

2.2. APPLICABILITYThe results from the Deliverability Study indicate the available, but not the committed, capacity of a resource. An LRE may enter into a Deliverable Capacity contract with a Generator Owner that has available capacity not already committed.  The Deliverability Study results can only be used to arrange capacity contracts for the Summer Peak Season3.

The Generator Owner will determine how much available capacity a resource has by subtracting the Deliverable Capacity from the already committed capacity, which will determine the amount of available capacity it has for Deliverable Capacity purchases. Figure 1 shows four different examples of how the amount of available capacity can be determined.

Figure 1: Chart representation of deliverable, committed, and available capacity

2.3. STUDY PROCESS The Deliverability Study process begins with staff defining scope assumptions, which are vetted using the SPP stakeholder process. Once the scope is finalized and approved, input data is collected based on the requirements listed in the scope. The data is modeled using the appropriate software along with the assumptions for the study. This study is performed on resources registered in the SPP

3 The Summer Peak Season is defined as June 1 to September 30.

7

Page 9: R deliverability st… · Web viewThe deliverable percentage has increased every study year for each fuel type except coal, natural gas, and wind. Since the 2016 Deliverability Study,

Integrated Marketplace while incorporating a First Contingency Incremental Transfer Capability (FCITC) transfer limit analysis. Final metric results were presented to the appropriate SPP working groups. The results for each resource was populated in the Engineering Data Submission Tool (EDST) by October 1 so each Generator Owner has access to the results for each of their resources.

8

Page 10: R deliverability st… · Web viewThe deliverable percentage has increased every study year for each fuel type except coal, natural gas, and wind. Since the 2016 Deliverability Study,

3. STUDY INPUTS AND ASSUMPTIONS

3.1. SOFTWAREThe Deliverability Study utilized the Transmission Adequacy & Reliability Assessment (TARA) software package from PowerGEM©. TARA is a steady-state power flow software tool with modeling capabilities and analytical applications that extend beyond traditional power flow solution. Using linear (DC) and non-linear (AC) power flow calculations, TARA has the capabilities to perform data checking, N-1 reliability analysis, transfer limit calculation, preventive and corrective dispatch, critical facility identification, reactive analysis, outage analysis, model building, and region specific tools for generation deliverability and reserve requirements analysis. The multi-transfer limit function was used to perform the 2019 Deliverability Study.

3.2. SPP BALANCING AUTHROITY AREAThe map in Figure 2 shows the dispatching boundary of the SPP BA Area for the Deliverability Study4.

Figure 2: Regional map indicating the SPP BA boundary

4 Only the eastern interconnect portions were included for simulation purposes, i.e. all facilities and loads east of the DC ties.

9

Page 11: R deliverability st… · Web viewThe deliverable percentage has increased every study year for each fuel type except coal, natural gas, and wind. Since the 2016 Deliverability Study,

3.3. BASE MODELS AND TOPOLOGYSPP staff and stakeholders develop planning models representing specific system conditions as part of the Integrated Transmission Planning Near-Term (ITPNT) process5. Generator Owners submit generator capability data to SPP that reflect the current capability ratings in the planning models. Transmission projects are developed through the ITPNT process to address reliability needs. Based on system conditions, the ITPNT Balancing Authority (BA) model demonstrates expected transmission system flows under a Security Constrained Economic Dispatch (SCED) similar to the Integrated Marketplace dispatch and is used as the base model for the Deliverability Study. Additional information about the ITPNT BA model can be found in the ITPNT Scope6.

The topology included in the Deliverability Study utilized the 2020 summer season BA model from the 2019 ITPNT model build.

3.4. GENERATOR GROUPINGEach resource was grouped into plants based on the site and point of interconnection for each resource. A plant’s maximum output is the summation of the maximum output of all resources at the same site. In order to capture reliability constraints caused by simultaneously increasing multiple plants in a localized area, additional analysis was performed in order to group one or more plants together using a methodology based on Power Transfer Distribution Factors (PTDF). Generators that share PTDF impacts greater than 25% on any transmission element in SPP were grouped together. There were 379 resource groups analyzed in the study, which were comprised of 952 units.

A transfer file was created based on the difference between dispatched capacity from the planning model and nameplate capacity of the plant. For example, if a plant with the nameplate capacity of 200 MW were dispatched at 150 MW, the amount for the transfer analysis would be 50 MW. A transfer file was created for all plants in the study.

3.6. MONITORED AND CONTINGENT ELEMENTSThe monitored and contingent elements included all SPP transmission facilities 100kV and above as well as first tier transmission facilities 230kV and above.

Known System Operating Limits (SOLs) 69 kV and above were monitored during the analysis. The SOLs, which are studied in the annual flowgate assessment, were sourced from the most recent book of flowgates.

5 Link to the latest ITP Manual: https://www.spp.org/engineering/transmission-planning/6 Link to the latest ITPNT process scope: https://www.spp.org/engineering/transmission-planning/

10

Page 12: R deliverability st… · Web viewThe deliverable percentage has increased every study year for each fuel type except coal, natural gas, and wind. Since the 2016 Deliverability Study,

3.7. EXCLUDED CONTINGENCIESFrom the contingencies in Section 3.6, contingencies reported as invalid in the ITPNT process were excluded.

3.8. OTHER STUDY ASSUMPTIONS The initial dispatch amount in the base model is assumed automatically

deliverable to the SPP BA Area since the model dispatches around constraints and sets dispatch amounts for each resource.

As generation is uniformly increased for each group of resources, the SPP Balancing Authority Area demand uniformly increases.

A FCITC analysis of each transfer was performed to determine deliverability limits.

A three percent (3%) Transfer Distribution Factor (TDF) threshold was used to analyze constraints impacted by the transfer.

11

Page 13: R deliverability st… · Web viewThe deliverable percentage has increased every study year for each fuel type except coal, natural gas, and wind. Since the 2016 Deliverability Study,

4. RESULTS

4.1. 2019 DELIVERABILITY STUDY RESULTSAfter performing the analysis, it was determined the total Deliverable Capacity was 83,509 MW and the deliverability percentage was 95.0% using the approved assumptions outlined for the study. The amount of Deliverable Capacity increased from the 2018 Deliverability Study by 100 MW due to an increase of wind capacity. However, the overall deliverability percentage decreased from last year’s analysis by 1%. Table 3 shows the 2019 Deliverability Study Results by fuel type. Table 3: 2019 Deliverability Study Results by fuel type

Fuel TypeDeliverable

Capacity (MW)Installed

Capacity (MW)Percent Deliverable

(Deliverable Capacity /Max Capacity)

Agricultural Byproducts 21 21 100.0%Coal 23,632 23,774 99.4%

Fuel Oil 1,153 1,343 85.9%Municipal Solid Waste 13 13 100.0%

Natural Gas 34,015 35,583 95.6%Nuclear Fuels 2,115 2,115 100.0%

Other Fuel 61 63 98.2%Solar 187 215 87.1%Water 3,423 3,468 98.7%Wind 18,889 21,290 88.7%

Total Capacity 83,509 87,884 95.0%

Error: Reference source not foundError: Reference source not found Lists Deliverable Capacity and number of plants by deliverability percentage categories.

4.2. TRENDING RESULTSThe annual Deliverability Study process began in 2016; therefore, results were trended based on the past three studies performed. The incremental amount of Deliverable Capacity has been

12

Page 14: R deliverability st… · Web viewThe deliverable percentage has increased every study year for each fuel type except coal, natural gas, and wind. Since the 2016 Deliverability Study,

synonymous with the incremental amount of max capacity studied, which is represented in Table4.

Table 4: Deliverable Capacity from the previous three studiesDeliverability

StudyMax Capacity

(MW)Deliverable

Capacity (MW)Percent Deliverable (Deliverable

Capacity /Inst. Capacity)

2016 83,334 79,808 95.8%2017 85,496 82,340 96.3%2018 86,747 83,410 96.2%2019 87,884 83,509 95.0%

Figure 3, Figure 4, and Figure 5 display the Deliverable Capacity and percent deliverable results for each study year by fuel type. The deliverable percentage has increased every study year for each fuel type except coal, natural gas, and wind. Since the 2016 Deliverability Study, the total amount of installed wind capacity has increased 46% (~6,730 MW) while the total amount of coal and nuclear capacity has decreased 12.3% (~2,900 MW) and 20% (590 MW), respectively. The amount of max capacity from natural gas resources increased from 2018 to 2019 due to approximately 1,120 MW being converted from coal.

Figure 3: Trending Deliverability Study results for Coal, Natural Gas, and Wind resources

13

Page 15: R deliverability st… · Web viewThe deliverable percentage has increased every study year for each fuel type except coal, natural gas, and wind. Since the 2016 Deliverability Study,

Figure 4: Trending Deliverability Study results for Fuel Oil, Nuclear, and Hydro resources

Figure 5: Trending Deliverability Study results for Ag. Byproducts, Municipal Waste, Solar, and other fuel type resources

14

Page 16: R deliverability st… · Web viewThe deliverable percentage has increased every study year for each fuel type except coal, natural gas, and wind. Since the 2016 Deliverability Study,

4.3. CONCLUSIONThe analysis concluded the total Deliverable Capacity was 83,509 MW and the deliverability percentage was 95.0% with the assumptions outlined for the study. The amount of Deliverable Capacity increased from the 2018 Deliverability Study by 100 MW due to an increase of wind capacity. However, the overall deliverability percentage decreased from last year’s analysis by 1%.

15

Page 17: R deliverability st… · Web viewThe deliverable percentage has increased every study year for each fuel type except coal, natural gas, and wind. Since the 2016 Deliverability Study,

APPENDIX D: LIST OF ACRONYMS

BA Balancing AuthorityDC Direct Current EDST Engineering Data Submission ToolFCITC First Contingency Incremental Transfer CapabilityITPNT Integrated Transmission Planning Near-Term LOLE Loss of Load Expectation LRE Load Responsible EntityMW MegawattOATT Open Access Transmission Tariff ORWG Operational Reliability Working GroupPRM Planning Reserve Margin PTDF Power Transfer Distribution FactorsSAWG Supply Adequacy Working GroupSCED Security Constrained Economic DispatchSOL System Operating Limit SPP Southwest Power PoolTARA Transmission Adequacy & Reliability AssessmentTDF Transfer Distribution FactorTWG Transmission Working Group

16