QP 311 Kingdom Community Wind Project Interconnection Feasibility Study. July, 2010 FINAL REPORT. Prepared by:

Similar documents
Sheffield-Highgate Export Interface SHEI. VSPC Quarterly Meeting October 18, 2017

3 Reliability Issues and Solutions

Elbert County 500 MW Generation Addition Interconnection Feasibility Study Report OASIS POSTING # GI

PID 274 Feasibility Study Report 13.7 MW Distribution Inter-Connection Buras Substation

100 MW Wind Generation Project

Location Upgrade Need Category Affected DU s Lead DU Reasons for selecting Affected utilities Reasons for Lead DU

Gateway South Transmission Project

Interconnection System Impact Study Report Request # GI

EL PASO ELECTRIC COMPANY (EPE) FACILITIES STUDY FOR PROPOSED HVDC TERMINAL INTERCONNECTION AT NEW ARTESIA 345 KV BUS

Feasibility Study Report

THE NECESSITY OF THE 500 KV SYSTEM IN NWE S TRANSMISSION SYSTEM TO MAINTAIN RELIABLE SERVICE TO MONTANA CUSTOMERS

Project #148. Generation Interconnection System Impact Study Report

Interconnection Feasibility Study Report GIP-226-FEAS-R3

Service Requested 150 MW, Firm. Table ES.1: Summary Details for TSR #

Interconnection Feasibility Study Report GIP-222-FEAS-R3

Interconnection Feasibility Study Report Request # GI Draft Report 600 MW Wind Generating Facility Missile Site 230 kv Substation, Colorado

Stability Study for the Mt. Olive Hartburg 500 kv Line

Generator Interconnection Facilities Study For SCE&G Two Combustion Turbine Generators at Hagood

Feasibility Study Report

Generation Interconnection Feasibility Study For XXXXXXXXXXXXXXXXXXXXXX MW generator at new Western Refinary Substation

Supplemental Report on the NCTPC Collaborative Transmission Plan

ABB POWER SYSTEMS CONSULTING

SPS Planning Criteria and Study Methodology

Emera Maine Representative: Jeffrey Fenn, P.E., SGC Engineering LLC

Generator Interconnection System Impact Study For

TOLTEC POWER PARTNERSHIP TOLTEC POWER PROJECT INTERCONNECTION STUDY SYSTEM IMPACT STUDY

Final Draft Report. Assessment Summary. Hydro One Networks Inc. Longlac TS: Refurbish 115/44 kv, 25/33/ General Description

Interconnection System Impact Study Final Report February 19, 2018

Interconnection Feasibility Study Report GIP-IR373-FEAS-R1

CUSTOMER/ TWIN ARROWS PROJECT

Transmission Competitive Solicitation Questions Log Question / Answer Matrix Harry Allen to Eldorado 2015

Feasibility Study Report

Georgia Transmission Corporation Georgia Systems Operations Corporation

PJM Generator Interconnection Request Queue #R60 Robison Park-Convoy 345kV Impact Study September 2008

Project #94. Generation Interconnection System Impact Study Report Revision

Interconnection Feasibility Study Report GIP-084-FEAS-R2

Updated Transmission Expansion Plan for the Puget Sound Area to Support Winter South-to-North Transfers

SYSTEM IMPACT STUDY EC300W ERIS FINAL REPORT. El Paso Electric Company

Falcon-Midway 115 kv Line Uprate Project Report

Emera Maine Representative: Jeffrey Fenn, P.E., SGC Engineering LLC

DUKE ENERGY PROGRESS TRANSMISSION SYSTEM PLANNING SUMMARY

Feasibility Study. Customer Kingman Area Photovoltaic Generation Project Interconnection

System Impact Study Report

DETOUR GOLD CORPORATION SYSTEM IMPACT ASSESSMENT FOR DETOUR LAKE PROJECT

Interconnection Feasibility Study Report GIP-023-FEAS-R1. Generator Interconnection Request # MW Wind Generating Facility Inverness (L6549), NS

Interconnection Feasibility Study Report GIP-157-FEAS-R2

El Paso Electric Company

ATTACHMENT Y STUDY REPORT

Connection Engineering Study Report for AUC Application: AESO Project # 1674

15 Nelson-Marlborough Regional Plan

PJM Generator Interconnection R81 Emilie (Fords Mill) MW Impact Study Re-Study

SYSTEM IMPACT RESTUDY H252W ERIS REPORT. El Paso Electric Company

OCTOBER 17, Emera Maine Representative: Jeffrey Fenn, P.E., LR/SGC Engineering LLC

Interconnection Feasibility Study Report GIP-369-FEAS-R1

Feasibility Study for the Q MW Solar Project

Wheeler Ridge Junction Substation Project Description and Functional Specifications for Competitive Solicitation

El PASO ELECTRIC COMPANY 2014 BULK ELECTRIC SYSTEM TRANSMISSION ASSESSMENT FOR YEARS

Generation Interconnection Impact Study Results for Eurus Energy America Corporation s Eurus Combine Hills I, LLC Power Project.

Midway/Monument Area TTC Study

Appendix D Black Hills Project Summary

MILLIGAN SOLAR PROJECT

Electric Power Research Institute, USA 2 ABB, USA

Feasibility Study for the Q MW Solar Project

Consulting Agreement Study. Completed for Transmission Customer

ENERGY RESOURCE INTERCONNECTION SERVICE STUDY DRAFT REPORT

Evaluation of the Performance of Back-to-Back HVDC Converter and Variable Frequency Transformer for Power Flow Control in a Weak Interconnection

Outer Metro 115 kv Transmission Development Study. (Scott Co, Carver Co and Hennepin Co)

PES Cook Islands KEMA Grid Study Final Report

SYSTEM IMPACT STUDY REPORT LA450S GENERATION STUDY. El Paso Electric Company

XXXXXXXXXXXXXXXXXXXXXXXXX TRANSMISSION/GENERATION FEASIBILITY STUDY FATAL FLAW AND FEASIBILITY ANALYSIS

TransWest Express Project

Reliability Analysis Update

SMUD 2014 Ten-Year Transmission Assessment Plan. Final. December 18, 2014

FIRSTENERGY S PROPOSED SOLUTION AND REQUEST FOR CONSTRUCTION DESIGNATION

Seabrook Substation Reliability Improvement Project

NEWFOUNDLAND AND LABRADOR HYDRO GULL ISLAND TO SOLDIERS POND HVDC INTERCONNECTION DC SYSTEM STUDIES VOLUME 1

TRANSMISSION PLANNING CRITERIA

SPIDER Modeling Sub-Group DER Modeling, CAISO Experience

Q95 Vicksburg 69kV. System Impact Study. APS Contract No Arizona Public Service Company Transmission Planning.

The Long-Range Transmission Plan

TEN YEAR PLANNING GUIDE SHASTA LAKE ELECTRIC UTILITY

Guide. Services Document No: GD-1401 v1.0. Issue Date: Title: WIND ISLANDING. Previous Date: N/A. Author: Heather Andrew.

North Oregon Coast Area Study. Adam Lint

Merger of the generator interconnection processes of Valley Electric and the ISO;

Transmission Improvements Plan for 575 MW Network Service Request Wansley CC 7 Generation Facility (OASIS # ) Georgia Transmission Corporation

XXXXXXXXXXXXXXXX Transmission Interconnection Feasibility Study

NETSSWorks Software: An Extended AC Optimal Power Flow (AC XOPF) For Managing Available System Resources

PLANNING, ELIGIBILITY FOR CONNECTION AND CONNECTION PROCEDURE IN EMBEDDED GENERATION

CIS-IEEE 2017 Conference Renewable Energy Session Renewable Energy s Impact of Power Systems

Refurbishment and Life Extension of Existing Facilities in Vermont

A member-consumer with a QF facility shall not participate in the Cooperative s electric heat rate program.

EL PASO ELECTRIC COMPANY SHORT CIRCUIT ANALYSIS FOR XXX S PROPOSED GENERATION INTERCONNECTION

AMERICAN ELECTRIC POWER 2017 FILING FERC FORM 715 ANNUAL TRANSMISSION PLANNING AND EVALUATION REPORT PART 4 TRANSMISSION PLANNING RELIABILITY CRITERIA

Targeted Application of STATCOM Technology in the Distribution Zone

MTEP13 SPM #2- Central Region. March 25, 2013

Sub Regional RTEP Committee - Southern. August 19, 2011

EE 456 Design Project

Energy Storage Interconnection Initiative

Merchant Transmission Interconnection PJM Impact Study Report. PJM Merchant Transmission Request Queue Position X3-028.

Generation Interconnection Facilities Study For

Transcription:

FINAL REPORT QP 311 Kingdom Community Wind Project Interconnection Feasibility Study July, 2010 Prepared by: i For ISO New England and Vermont Electric Power Company

Table of Contents Executive Summary... iv 1 Background... 1 1.1 Study Objective... 1 1.2 Project Description... 1 1.3 Technical Specifications... 3 1.3.1 Project Generator Modeling Data... 3 1.3.2 Project Transformer Modeling Data... 4 1.3.3 Project Transmission Line Modeling Data... 4 2 Study Area... 6 2.1 Transmission System... 6 3 Base Case Development... 7 3.1 Base Case Origin and Year... 7 3.2 Area Load... 7 3.3 Planned Projects... 7 3.4 Base Case Naming Convention... 7 3.5 Voltage Operational Analysis... 8 3.6 Analytical Tools... 9 4 Steady State Analysis Methodology... 9 4.1 Steady State Voltage Limits... 9 4.2 Steady State Thermal Limits... 9 4.3 Steady State Base Case Dispatch and Interface Conditions... 10 4.4 Steady State Contingency List... 11 5 Steady State Analysis Results... 12 5.1 Baseline System... 12 5.1.1 Voltage Performance... 12 5.1.2 Thermal Performance... 13 5.2 QP 311 In-Service... 14 5.2.1 Voltage Performance... 16 5.2.2 Thermal Performance... 17 6 Short Circuit Analysis... 18 7 Delta V on Capacitor Switching... 19 8 Cost Estimates for Required Network Upgrades... 19 9 Conclusion... 20 Appendices Appendix A - Project Information... A Appendix B - Study Methodology...B Appendix C - Preliminary One-Line Diagram...C Appendix D - Steady State Contingency List... D Appendix E - Steady State Base Case Summaries...E Appendix F - Steady State Base Case Draw Files... F Appendix G - Steady State Contingency Voltage Results... G Appendix H - Steady State Contingency Thermal Results... H ii

List of Tables Table 1-1 QP 311 Generator Modeling Data... 4 Table 1-2 QP 311 Transformer Modeling Data... 4 Table 1-3 QP 311 Transmission Line Modeling Data... 4 Table 1-4 QP 311 Collector String Modeling Data... 5 Table 4-1 Steady State Voltage Criteria... 9 Table 4-2 Steady State Thermal Criteria... 10 Table 4-3 Local Area Dispatches... 11 Table 5-1 Peak Load Contingencies Resulting in Voltage Violations Pre-Project... 13 Table 5-2 Losses Due to Addition of Q311 Project... 14 Table 5-3 Newport Area Pre Contingency Voltages... 15 Table 5-4 Additional Capacitors for K42 Contingency... 15 Table 5-5 Reactive Compensation Proposed for the Project... 16 Table 5-6 Post Contingency Capacitor Switching at Jay Tap Switching SS... 16 Table 6-1 Kingdom Community Wind Short-circuit Fault Duties... 19 List of Figures Figure 1-1 Pre Project Simplified One Line Diagram... 2 Figure 1-2 Post-Project Simplified One Line Diagram... 3 Figure 2-1 Project Geographical Map... 6 Figure 3-1 Steady State Base Case Naming Convention... 8 iii

Executive Summary RLC Engineering, LLC (RLC) conducted a Feasibility Study (the Study ) under the ISO New England Inc. (ISO) Open Access Transmission Tariff ( Tariff ) Schedule 22-Standard Large Generator Interconnection Procedures ( LGIP ). The Study was performed on behalf of ISO New England Inc. (ISO) and Vermont Electric Power Company (VELCO) for the Interconnecting Customer at Queue Position 311 (QP 311) to construct a 63 MW Wind Farm Project (the Project ) located in Lowell, Vermont in Orleans County. The wind farm is proposed to consist of twenty-one 3.0 MW Vestas V90 wind turbines connecting into the new Vermont Electric Cooperative (VEC) 46 kv Lowell Substation. The Project has a proposed inservice date of October 2012. Based on a prior study conducted by VELCO and VEC, the following changes are proposed to occur prior to the Project in-service date: o Construct Jay Tap Switching Station 4 miles west of the North Troy Substation on the existing 46 kv transmission line (the VT Public Service Board (PSB) has opened Docket 7604 to review VEC s Jay Tap Switching Station construction permit application) o Install five 46 kv breakers at Jay Tap Switching Station o Install four 2.7 MVAr shunt capacitors at Jay Tap Switching Station The primary objective of this Study was to determine if interconnecting QP 311 (the Project) would have significant adverse impact on the reliability and operating characteristics of the VELCO or VEC transmission systems, the transmission facilities of another Transmission Owner, or the system of a Market Participant. Steady state conditions and short circuit testing were analyzed in this Study. The purpose of the Study was to: (i) Analyze the steady-state and short circuit impact of the Project (ii) Determine any upgrades to the transmission system that would be required to mitigate any adverse impacts that the Project could otherwise pose on the reliability and operating characteristics of the New England transmission system (iii) Determine any upgrades required to mitigate any degradation to transmission transfer capability Project Description To accommodate the interconnection of the Project, a portion of the existing 34.5 kv transmission system around Lowell will need to be upgraded to 46 kv and a new 46 kv line will need to be constructed to interconnect with the VELCO 115 kv transmission system. The following is an overview of the facilities required for the interconnection of the Project: iv

Install a new 115/46 kv autotransformer on the K41 Line between Highgate and Moshers Tap to connect to the Jay Tap 46 kv Substation Install one 46 kv breaker between the Jay Tap Switching Station bus and the 46kV terminal of the autotransformer Upgrade 2 miles of existing 46 kv transmission line from Jay Tap Switching Substation to a tap point (Crossroad) to 795 kcmil ACSR and add three 46 kv Switches at Crossroad tap point Construct 2 miles of new 46 kv transmission line from Crossroad tap point to the Jay 17 Substation with 795 kcmil ACSR Upgrade the 34.5 kv Jay 17 Substation to 46 kv and install a 46/12.47 kv transformer at Jay 17 Substation Upgrade 10.4 miles of transmission line (to 795 kcmil ACSR) from Jay 17 Substation to VEC Lowell Substation Construct a VEC Lowell 46 kv Substation and install three 46 kv breakers and add a 46/12.47 kv transformer Construct 5.4 miles of new 46 kv transmission line (795 kcmil ACSR) connecting QP 311 Substation to VEC Lowell 46 kv Substation Construct a 46/34.5 kv substation in Lowell, Vermont (QP 311 Substation) Install twenty-one 3.0 MW Vestas V90 wind turbines equipped with their own dedicated 1000 V/34.5 kv GSU Construct two 34.5kV collector strings connecting the wind turbines with each tying into the QP 311 Substation at 34.5 kv See Section 1.2 for a complete discussion of interconnection details. Steady State Steady state voltage and thermal analyses examined system performance without the proposed Project in order to establish a baseline for comparison. System performance was re-evaluated with the Project and compared with the previous baseline performance to demonstrate the impact of the Project on area transmission reliability under the guidelines of the Network Capability Interconnection Standard (NCIS). Several redispatch conditions under the NCIS were evaluated at each load and transmission operating configuration. Steady state analysis was evaluated at a summer 2013 peak load level of 31,470 MW and at a summer 2013 shoulder load level of 22,024 MW for ISO New England. The shoulder load (D1) dispatch represented the block load being supplied from Hydro Quebec (HQ) and the Highgate HVDC converter at full output to stress the area in an exporting condition. The peak load (D2) dispatch represented the block load being supplied from Vermont and the Highgate HVDC converter out of service to stress the area in an importing condition. Additional sensitivity dispatches were performed at peak load with the Highgate HVDC converter on and the block load supplied from either Vermont or HQ. The objective of modeling these various dispatch scenarios was to examine the proposed Project and the ability of the transmission system to reliably serve customer demand under various stressed system conditions. v

Based upon the steady state results, the Project as originally proposed posed an adverse impact on the reliability and operating characteristics of the transmission system, and would require transmission system upgrades. For many single element contingencies (K28, K38, K42 and K60) voltage violations of reliability criteria were associated with the Project due to heavy reactive power losses on the 46 kv transmission system and the post-project transmission configuration. With reactive losses exceeding 20 MVAr along the 46 kv transmission corridor connecting the project to the Jay Tap Switching Station and over 20 MVAr of reactive losses within the Project facilities, the Project generators alone were incapable of providing the reactive support needed to sustain acceptable voltage criteria. Also, the post-project transmission configuration has the 12W Switch open at N. Troy which removes the reactive support of the Jay Tap Switching Substation shunt capacitors to the Newport area which is deficient in reactive resources when the block load is served by Vermont. Based upon the steady state results, for pre-project and post-project dispatches with the block load on Vermont, contingencies involving the loss of the Irasburg 115/46 kv transformer (H39 transformer and K41 Line stuck breaker) resulted in cases diverging due to the loss of voltage support for the Montgomery, Eden Corners and Johnson Substation areas. The reliability issues associated with these contingencies were not addressed in this Feasibility Study. Several system upgrades and project modifications were analyzed to provide necessary system voltage support. The following upgrades and modifications are recommended to reliably maintain system voltage criteria: Add 4 MVAr dynamic reactive support at the Project s 34 kv bus The QP 311 46/34.5 kv transformer tapped winding must be modified to the 102.5% tap Increase the size of each of the four planned 2.7 MVAr shunt capacitor banks at the Jay Tap Switching Substation to 5.4 MVAr for a total increase to the system of 10.8 MVAr Provide automatic capacitor bank switching at Jay Tap to maintain reliable 46 kv system voltage (or provide dynamic reactive power control) The Jay Tap Switching Substation 115/46 kv transformer tapped winding must be set to the 97.5% tap Add a 5.4 MVAr cap at the Newport-B2 bus for area voltage support Add a transfer trip scheme to trip the QP 311 units upon loss of 46 kv line or Jay Tap Switching Station 115/46 kv transformer When the Project is dispatched locally under stressed exporting conditions (maximum generation and shoulder load), no thermal reliability issues exist which require mitigation. The 46kV transmission lines between Irasburg and Central Vermont Public Service Company s Johnson Substation are at thermal capacity by generation in the area following contingencies of the K38 Line (Lyndonville-Sheffield 115kV Line). An additional sensitivity redispatch demonstrated that lowering the Sheffield dispatch with the Project was effective to relieve the loading on the underlying 46 kv transmission system. vi

Short Circuit Short circuit analysis was performed to determine the fault current levels on the new 46kV transmission system proposed for the Project. The maximum fault duty was approximately 4400A for a three phase to ground fault and 4900A for a single phase to ground fault at the Jay Tap 46kV S/S with the Project in-service. These fault duties are low and should be considered in the engineering design studies for the transmission system upgrades required for the Project. Cost Estimate VEC provided a cost estimate of approximately $1,184k to construct the necessary required network upgrades. The cost projections are based on installed costs and do not include the cost for environmental assessment, permitting or temporary facilities required for outage protection. VEC has communicated that subject to receipt of timely regulatory approval, the upgrades can be implemented in time to fit the planned commercial operations schedule for the Project. With the addition of the upgrades listed above to the Project as currently defined, the Project poses no significant adverse impact on the reliability and operating characteristics of the VELCO or VEC transmission systems, the transmission facilities of another Transmission Owner, or the system of a Market Participant. vii

1 Background 1.1 Study Objective The primary objective of this Study was to determine if interconnecting QP 311 (the Project) would have significant adverse impact on the reliability and operating characteristics of the VELCO or VEC transmission systems, the transmission facilities of another Transmission Owner, or the system of a Market Participant. Steady state and short circuit conditions were analyzed in this Study. 1.2 Project Description The Project consists of the following proposed electrical components and construction activities: Split the K41 Line 12.57 miles from Mosher s Tap end with a three-breaker ring bus and install a new 115/46 kv autotransformer to connect the 46kV Jay Tap Switching Station with the VELCO 115kV transmission system Install one 46 kv breaker between the Jay Tap Switching Station bus and the 46kV terminal of the autotransformer Construct a 2 mile 46 kv 795 kcmil ACSR transmission line from the Jay 17 Substation to a tap point (Crossroad) on the existing 46 kv transmission line 2 miles west of the North Troy Substation Upgrade the existing 46 kv transmission line between Jay Tap Switching Substation and the tap point (Crossroad) to 795 kcmil ACSR Install three 46 kv Switches at the Crossroad tap point Upgrade the 34.5 kv Jay 17 Substation to 46 kv and replace the 34.5/12.47 transformer with a 46/12.47 kv transformer Construct a VEC Lowell 46 kv Substation Install three 46 kv breakers at VEC Lowell Substation Install a 46/12.47 kv transformer at VEC Lowell Substation Upgrade 10.4 miles of transmission line (to 795 kcmil ACSR) from VEC Lowell 46 kv Substation to Jay 17 46kV Substation Construct a new 46/34.5 kv substation in Lowell, Vermont (QP 311 Substation) Install one 46 kv breaker at QP 311 Substation Install twenty-one 3.0 MW Vestas V90 wind turbines equipped with their own dedicated 1.0/34.5 kv GSU (actually 3-wdg but modeled as 2-wdg) Construct two collector strings connecting the wind turbines with each tying into the QP 311 Substation at 34.5 kv Construct 5.4 miles of 46 kv transmission line (795 kcmil ACSR) connecting QP 311 Substation to VEC Lowell 46 kv Substation Install a 34.5 kv breaker on each collector line at the QP311 substation Post-Project, area support for the VEC Lowell Substation and Jay 17 Substation will shift from the VELCO Irasburg # 42 Substation to the new 115/46 kv Jay Tap Substation. However, the existing Montgomery, Eden Corners and Johnson Substations will remain supported out of the VELCO Irasburg #42 Substation. A transmission line which connects the new VEC Lowell 46 kv Substation and existing 46 kv line to Irasburg # 42 Substation for emergency purposes 1

remained open for the entire Study. In addition, the North Troy substation will be separated from the Jay Tap and Richford 46 kv substations. Figure 1-1 below shows the area transmission configuration pre-project. Figure 1-2 below shows the transmission configuration post-project. The Project will be modeled in detail as shown in Appendix C. The Project has a proposed in-service date of October 2012. To Highgate 115 kv K41 Line To Moshers Tap 115 kv Switch 118 closed Switch 14W Open Jay Tap Switching Station 46 kv Switch 12W Closed North Troy To Newport Center To E. Berkshire 2 Miles 2 Miles Switch 12E Closed Richford To Jay Substation #40 2.7 MVAr 2.7 MVAr 2.7 MVAr 2.7 MVAr Jay 17 Substation 34.5 kv Xfmr 34.5/ 12.47 kv 7.5 MVA Pre Project Configuration VEC / Lowell 34.5 kv Substation Xfmr 34.5/12.47kV To Lowell 12kV To Montgomery / Eden Corners / Johnson Xfmr 46 / 34.5 kv VELCO Irasburg Substation #42 Figure 1-1 Pre Project Simplified One Line Diagram 2

To Highgate 115 kv K41 Line Moshers Tap 115 kv GM Eleven units each consisting of 3 MW and -0.875 to 0.609 MVAr* M Project GSU 34.5 / 1.0 kv 3.16 MVA* GSU 34.5 / 1.0 kv 3.16 MVA* Switch 118 closed pre and post Project To E. Berkshire Collector #1 Switch 14W Open pre and post Project Xfmr 115 / 46 kv 40 / 66.7 MVA 8.5 % Z 2.7 MVAr 2.7 MVAr Richford Kingdom 46 kv Bus Jay Tap Switching Switch KCW-E Station 115 / 46 kv Open 5.4 Miles 2 Miles Switch KCW-W Closed To Jay (40) Crossroad 2 Miles 2.7 MVAr 2.7 MVAr North Troy 2 Miles Switch 12W Open Post Proj. Switch KCW-S Closed 10.4 Miles Switch 12E Closed To Newport Center Jay 17 Substation 46 kv Xfmr 46 / 12.47 kv 7.5 MVA G Ten units each consisting of 3 MW and -0.875 to 0.609 MVAr* Collector #2 Xfmr 46 /34.5 kv 40 / 66.7 MVA 7.5 % Z VEC / Lowell 46 kv Substation N.O. Xfmr 46 / 12.47 kv 7.5 MVA Queue 311 Substation Xfmr 46 / 34.5 kv 7.5 MVA To Lowell 12kV * Twenty-one units and GSU s to be modeled in detail and not as an aggregate. Reference Appendix C for detai To Montgomery / Eden Corners / Johnson VELCO Irasburg #42 Figure 1-2 Post-Project Simplified One Line Diagram 1.3 Technical Specifications The following tables contain data as provided by the developer and ISO for the Project. The installed maximum capability for the Project is 63 MW. Each of the twenty-one 3.0 MW Vestas V90 wind turbines has its own 1.0/34.5 kv GSU and was modeled in detail as shown in Appendix C. 1.3.1 Project Generator Modeling Data The Vestas V90 wind turbine MVAr capability, as provided by the developer, has a power factor range of 0.96 leading (consuming) to 0.98 lagging (generating). The generators were set to operate in power factor control mode. The units were set to a fixed MVAr output that attempted to maintain unity power factor at the point of interconnection (Kingdom 46 kv Bus) while maintaining acceptable area voltages. Table 1-1 below provides the generator operating characteristics. 3

QP 311 Generator Modeling Data MVAr Capability Type MVA PF range MW Leading / Consuming Vestas V90 wind turbine Lagging / Generating 3.06-0.96 to 0.98 3.0 0.875 0.609 Table 1-1 QP 311 Generator Modeling Data 1.3.2 Project Transformer Modeling Data Table 1-2 below summarizes the data required to model the Project transformers. The interconnection and GSU transformers are fixed tap transformers with a range of +2 x 2.5%. QP 311 Transformer Data Transformer Tap Ratio Present Tap Nameplate Capacity (MVA) R (pu) X (pu) GSU 34.5/1.0 Center 3.16 0.0065 0.0948 Interconnection 46/34.5 102.5% Transformer at QP 311 (High Side) 40/53.3/66.7 0.0022 0.0750 Transformer at Jay 115/46 97.5% Tap Switching Station (High Side) 40/53.3/66.7 0.0025 0.0850 Jay 17 SS 46/12.47 Center 7.5/9.3 0.0050 0.0700 VEC Lowell SS 46/12.47 Center 7.5/9.3 0.0050 0.0700 Table 1-2 QP 311 Transformer Modeling Data 1.3.3 Project Transmission Line Modeling Data Table 1-3 and Table 1-4 summarize the modeling data for the transmission lines for the Project. Transmission line ratings were determined using available reference data. QP 311 Transmission Line Data Transmission Line Conductor Length MVA (Miles) Rating R (pu) X (pu) QP 311 SS to Lowell SS 795 KCMIL ACSR 5.40 76 0.0367 0.1598 VEC Lowell SS to Jay 17 SS 795 KCMIL ACSR 10.40 76 0.0707 0.3078 Jay 17 SS to Crossroad tap point (pole #155) on existing 46 kv line 2miles west 795 KCMIL ACSR 2.0 76 0.0134 0.0602 of North Troy SS Crossroad tap point (pole #155) to Jay Tap Switching SS 795 KCMIL ACSR 2.0 76 0.0134 0.0602 Table 1-3 QP 311 Transmission Line Modeling Data 4

QP 311 Collector 1 Data From To Conductor Type Length (FT) MVA R (pu) X (pu) Gen 1 Gen 2 1/0 AWG 800 10.0 0.0147 0.0065 Gen 2 Gen 3 1/0 AWG 1000 10.0 0.0184 0.0082 Gen 3 JB-1 1/0 AWG 1600 10.0 0.0294 0.0130 Gen 4 JB-1 1/0 AWG 200 10.0 0.0037 0.0016 JB-1 JB-2 500 KCMIL 820 23.4 0.0044 0.0048 Gen 5 JB-2 1/0 AWG 200 10.0 0.0037 0.0016 JB-2 JB-3 500 KCMIL 810 23.4 0.0044 0.0048 Gen 6 JB-3 500 KCMIL 200 23.4 0.0011 0.0012 JB-3 Coll Tap 750 KCMIL 200 28.4 0.0009 0.0010 Gen 7 JB-4 1/0 AWG 200 10.0 0.0037 0.0016 Gen 8 JB-4 500 KCMIL 1200 23.4 0.0065 0.0071 Gen 8 Gen 9 1/0 AWG 800 10.0 0.0147 0.0065 Gen 9 Gen 10 1/0 AWG 810 10.0 0.0149 0.0066 Gen 10 Gen 11 1/0 AWG 950 10.0 0.0175 0.0077 JB-4 Coll Isol 500 KCMIL 300 23.4 0.0016 0.0018 Coll Isol Coll Tap 795 KCMIL DRAKE 300 70 0.0007 0.0029 Coll Tap KCW 34 795 KCMIL DRAKE 5800 70 0.0128 0.0562 QP 311 Collector 2 Data From To Conductor Type Length (FT) MVA R (pu) X (pu) Coll 2 Tap Gen 12 500 KCMIL 500 23.4 0.0027 0.0029 Gen 12 Gen 13 1/0 AWG 900 10.0 0.0166 0.0073 Gen 13 Gen 14 1/0 AWG 850 10.0 0.0156 0.0069 Gen 14 Gen 15 1/0 AWG 820 10.0 0.0151 0.0067 Coll 2 Tap JB3-3 500 KCMIL 3150 23.4 0.0169 0.0185 JB3-3 Gen 16 1/0 AWG 200 10.0 0.0037 0.0016 JB3-3 JB3-2 500 KCMIL 810 23.4 0.0044 0.0048 JB3-2 Gen 17 1/0 AWG 200 10.0 0.0037 0.0016 JB3-2 JB3-1 500 KCMIL 810 23.4 0.0044 0.0048 JB3-1 Gen 18 1/0 AWG 200 10.0 0.0037 0.0016 JB3-1 Gen 19 1/0 AWG 1000 10.0 0.0184 0.0082 Gen 19 Gen 20 1/0 AWG 850 10.0 0.0156 0.0069 Gen 20 Gen 21 1/0 AWG 825 10.0 0.0152 0.0067 Coll 2 Tap KCW 34 795 KCMIL DRAKE 6800 70 0.0151 0.0658 Table 1-4 QP 311 Collector String Modeling Data 5

2 Study Area 2.1 Transmission System The primary area of concern for this study is the northwestern portion of VEC s service territory as shown in Figure 2-1 below. The sub transmission system in this area is heavily networked. The Project interconnects into the sub transmission system near the North Troy 46 kv Substation and also ties into the K41 115 kv transmission line between Highgate and Moshers tap. The Project is connected by a radial 46 kv Line as shown above in Figure 1-2 Project Interconnection Project Location Figure 2-1 Project Geographical Map 6

3 Base Case Development 3.1 Base Case Origin and Year The base case originated from VELCO and included a model of VELCO s sub-transmission system. The base case was revised to reflect proposed area projects. 3.2 Area Load Using the NEPOOL 2009 Capacity, Energy, Load and Transmission (CELT) Report and the methodology described in Appendix B 1, steady state analyses using a 2013 peak load forecast of 31,470 MW and a 2013 shoulder load forecast of 22,024 MW were completed. 3.3 Planned Projects The following list of planned facilities was present in the base case received from VELCO: Comerford QP 148 Sheffield Wind Project QP 172 Swanton Project QP 224 Lyndonville Station The following list of planned facilities was added to the base case received from VELCO: Wind QP 166 (Q195 now a three-terminal line with section to Littleton closed) Biomass Project QP 229 Biomass QP 251 Biomass QP 307 Lyndonville Transmission Project Capacitors Vermont Southern Loop transmission Project QP 274 3.4 Base Case Naming Convention Steady State base case designations were formatted as follows: 7

pk13_ p1_dx Dispatch Identifier D1 = Highgate On D2 = Highgate Off Project Identifier p0 = Project out of service p1 = Project in service Load Level Identifier PK = peak load SH = shoulder load Followed by load model year Figure 3-1 Steady State Base Case Naming Convention Several cases were developed for the Study at a summer 2013 peak load level of 31,470 MW and at a summer 2013 shoulder load level of 22,024 MW for ISO New England. Dispatch D1 - The shoulder load (D1) dispatch represented the block load being supplied from HQ and the Highgate HVDC converter at full output to stress the area in an exporting condition. Dispatch D2 - The peak load (D2) dispatch represented the block load being supplied from Vermont and the Highgate HVDC converter was analyzed out of service to stress the area in an importing condition. Sensitivity Dispatch 3- The peak load (D3) dispatch represented the block load being supplied from Vermont and the Highgate HVDC converter at full output. Sensitivity Dispatch 4- The peak load (D4) dispatch represented the block load being supplied from HQ and the Highgate HVDC converter at full output to stress the K42 Line. The objective of modeling these various dispatch scenarios was to examine the proposed Project and the ability of the transmission system to reliably serve customer demand under various stressed system conditions. 3.5 Voltage Operational Analysis The Study included an accurate and detailed model of the Project. All collector branches, individual high and low-voltage busses for the wind generators and GSU's were modeled using the configurations, network impedances, unit reactive capabilities and facility ratings provided. The detailed model allowed analysis of real and reactive power flows and losses across individual elements of the Project and made it possible to accurately test and monitor particular 8

voltage control strategies. Being able to monitor the terminal voltage at each individual wind turbine generator made it possible to ensure units at the end of the collector strings remain within voltage limits. The Project was adjusted (upgrades were added) to compensate for the reactive losses of its collector system and interconnection facilities. The Study concluded that by adding a 4 MVAr dynamic reactive device at the Project s 34kV bus for voltage control and setting the winding of the QP 311 46/34.5 kv transformer tap to 102.5% of nominal allowed voltages at the Project collector strings and generator buses to be maintained within criteria for area contingencies (Loss of K42, K41W and K38 Lines) which otherwise resulted in voltages outside criteria. 3.6 Analytical Tools A steady state analysis was performed using the GE Power Systems, PSLF load flow software package, Version 17. Short-circuit analyses were completed using the Aspen One Liner Program. 4 Steady State Analysis Methodology Steady state thermal and voltage analyses examined system performance without the proposed Project in order to establish a baseline for comparison. System performance was then reevaluated with the Project and compared with the previous baseline performance to demonstrate the impact of the Project on area transmission reliability. 4.1 Steady State Voltage Limits Table 4-1 identifies the voltage criteria used by VELCO in the primary Study area for steady state voltage assessment. Voltage Class Acceptable Voltage Range Pre-Contingency (Normal Conditions) Post-Contingency (Emergency Conditions) 230 kv and above 0.98 to 1.05pu 0.95 to 1.05pu 115 kv 0.95 to 1.05pu 0.95 to 1.05pu Below 115 kv 0.95 to 1.05pu 0.90 to 1.05pu Table 4-1 Steady State Voltage Criteria 4.2 Steady State Thermal Limits Table 4-2 contains the thermal loading performance criteria applied to transmission lines and transformers in the Study. 9

System Condition Pre-Contingency (all lines in) Post-Contingency Time Interval Continuous Maximum Allowable Facility Loading Normal Rating Less than 15 minutes after Short Time Emergency contingency occurs (STE) Rating More than 15 minutes after Long Time Emergency contingency occurs (LTE) Rating Table 4-2 Steady State Thermal Criteria 4.3 Steady State Base Case Dispatch and Interface Conditions Two load dispatches were analyzed for the study. D1 - Shoulder Load with Highgate on at 210 MW and the block load shifted to HQ to create a maximum export condition D2 - Peak Load with Highgate off and the block load on Vermont. Additional sensitivity dispatches were also performed at peak load with Highgate on and the block load on either Vermont or HQ. For the Study, four base cases were developed to analyze the impact of the Project on area reliability under stressed system conditions. Case A - Pre-Project (No Jay Tap 115/46 kv Substation) Case B - Project On-Line with no redispatch Case C - Project On-Line with redispatch against remote generation (Western Massachusetts) Case D - Project On-Line with redispatch against local generation (Sheffield and Swanton) Table 4-3 identifies these dispatch scenarios. 10

Generator Base Case Dispatch 1 (Shoulder Load) A - Pre Project B C D Dispatch 2 (Peak Load) * Sensitivity with Highgate on and block load either on Vermont or HQ A - Pre Project B C* D* * Wind (QP 166) 99 99 99 99 99 99 99 99 Biomass (QP 229) 29 29 29 29 29 29 29 29 Biomass (QP 251) 71 71 71 71 71 71 71 71 Vermont Yankee 667 667 667 667 667 667 667 667 Gorge (QP 274) 42 42 42 42 42 42 42 42 Moore 121 121 121 121 161 161 161 161 Comerford 144 144 144 144 144 144 144 144 Wind (QP 172) 40 40 40 19 40 40 40 19 Coventry 8 8 8 8 8 8 8 8 Barton 0 0 0 0 3 3 3 3 Swanton GT 42 42 42 0 42 42 42 0 QP 311 0 63 63 63 0 63 63 63 Highgate 210 210 210 210 0 / 210* 0 0 / 210* 0 / 210* PV 20 130 130 130 130 130 130 130 130 Highgate Falls 9 9 9 9 9 9 9 9 (Swanton Hydro) Sheldon Springs Hydro 3 3 3 3 6.5 6.5 6.5 6.5 Fairfax Hydro 3 3 3 3 3 3 3 3 Lower Lamoille (Peterson, Milton, Clark Falls) 14 14 14 14 14 14 14 14 Altresco (W. Mass) 66 66 3 66 66 66 3 66 Table 4-3 Local Area Dispatches Detailed interface transfer and dispatch summaries for each of the baseline cases are included in Appendix E. Draw files representing the baseline cases are included in Appendix F. 4.4 Steady State Contingency List The original contingency file provided by VELCO was reviewed and modified for the Study. Contingency analysis was conducted with approximately 50 contingencies encompassing single element, transformer, generation and 115kV stuck breaker outages within the northwestern portion of Vermont s transmission system. The Highgate Special Protection System (SPS) is always enabled (armed) and its action is modeled as appropriate in the contingencies listed in Appendix D. When the flow through the Highgate HVDC converter is from Quebec to Vermont, the SPS will reduce the converter import into Vermont to a pre-specified level (usually 150 MW) for loss of any of the following Lines: 11

Georgia Sandbar (K19 Line) Georgia Essex (K21 Line) Sandbar Essex (K22 Line) Sandbar So. Hero Plattsburgh (PV-20 line) Essex Williston (K23) and Essex - Berlin (K24) o If the K23 or K24 line is out of service, loss of the other line will cause runback to the pre-specified level Contingencies annotated with RB* include the Highgate SPS to model the runback. For this study it is assumed that a transfer trip scheme will be in place to trip the QP 311 units upon loss of the transmission line connecting QP 311 (VEC Lowell) 46 kv Substation to Jay 17 46 kv Substation or loss of the Jay Tap 115/46 kv transformer. Appendix D provides a listing of the contingencies used in the Study. 5 Steady State Analysis Results 5.1 Baseline System 5.1.1 Voltage Performance Baseline System All Lines In Under shoulder load conditions, steady state voltage analysis reported no violations of reliability criteria for the baseline with all lines in-service for any of the base cases. Under peak load conditions in Dispatch 2 (including sensitivity dispatches), steady state voltage analysis reported violations of normal criteria (less than 95%) for the baseline with all lines inservice for 46kV buses in several Vermont load regions. During peak load conditions with the block load supported by Vermont, the area has insufficient transmission voltage regulation and reactive resources to maintain voltages above the normal criteria. These voltage issues are mainly distributed throughout the underlying 46kV and 34kV subtransmission networks. Baseline System Post Contingency Under shoulder load conditions in Dispatch 1, post-contingency voltage analysis reported no violations of reliability criteria for the baseline with one exception, the Stowe DCT contingency which causes loss of 115kV and 34.5 kv lines supporting the Stowe area. Under peak load conditions in Dispatch 2 (and sensitivity dispatch 3) with the block load on Vermont and Highgate offline, contingencies involving loss of the Irasburg 115/46 kv H39 transformer resulted in solutions diverging due to the loss of voltage support for the Montgomery, Eden Corners and Johnson Substation areas. Under peak load conditions in Dispatch 2 with the block load on Vermont and Highgate offline, post-contingency steady state voltage analysis reported violations of reliability criteria (less than 90%) for multiple contingencies. The following contingencies listed in Table 5-1 resulted in 12

voltages below criteria in the baseline pre-project cases. The Barre 34.5 kv capacitor bank is available but offline based on the switching criteria in the study methodology. Outage Violations kv Barre X63 Xfmr Barre / Granite Area Busses 34.5 D-204 Newport / Wenlock Area Busses 46 K19 East Fairfax / Sandbar Area Busses 34.5 K19_RB East Fairfax / Sandbar Area Busses 34.5 K28 Newport / Richford / Wenlock Area Busses 46 K28_Stuck Bkr Newport / Richford / Wenlock Area Busses 46 K38 Newport / Richford / Wenlock Area Busses 46 K42 Newport / Wenlock Area Busses 46 K60 Newport / Richford / Wenlock Area Busses 46 Stowe Area Busses Between Johnson and Middlebury 34.5 Stowe DCT Area Busses From Newport Through Middlebury 115-34.5 Fairfax X67 Xfmr East Fairfax Area Busses 34.5 U199/X178 Ashland/Beebe 34.5 Irasburg H39 Xfmr Case Diverged with Block on Vermont K41 Stuck breaker Case Diverged with Block on Vermont Table 5-1 Peak Load Contingencies Resulting in Voltage Violations Pre-Project Reliability issues associated with these contingencies were not addressed in the Study. In the Sensitivity Dispatch 3 with Highgate online and block load on Vermont the following differences were seen from the D2 dispatch which had Highgate offline: o The D-204 Line remained within voltage criteria o The K42 Line outage resulted in voltages above the 105% criteria for the East Fairfax and VEC Pleasant Valley buses In the Sensitivity Dispatch 4 with Highgate online and block load on HQ the following differences were seen from the D2 dispatch which had Highgate offline and the block load on Vt: o The D-204 Line remained within voltage criteria o The K28 Line contingency resulted in voltages just above the 105% criteria for the Sheffield area o The K41 stuck breaker and Jay SS 40 transformer contingencies resulted in voltages above the 105% criteria for the Irasburg area o The K42 Line outage resulted in voltages above the 105% criteria for the East Fairfax and VEC Pleasant Valley buses o The Richford Transformer contingency resulted in voltages above the 105% criteria for the Jay area 5.1.2 Thermal Performance Baseline System All Lines In Under both shoulder and peak load conditions (including sensitivity dispatches), steady state thermal analysis reported no violations of reliability criteria for the baseline with all lines inservice. In Dispatch 1 with shoulder load and maximum generation, the Littleton to Comerford 13

D204 Line is near 100% of its normal rating which confirms the Northern Vermont and New Hampshire 115 kv areas are at their maximum export limit. Baseline System Post Contingency Under both peak and shoulder load conditions, steady state thermal analysis reported violations of reliability criteria for the baseline system post contingency. These violations are pre-existing thermal overloading concerns. Under Dispatch 1 for shoulder load conditions, loss of the Barre 115/34 kv X63 transformer resulted in overloading the Berlin to Mountain View 34 kv line, which is outside the local study area. The U-199/X-178 contingency overloaded the 230 kv Comerford to Littleton D204 Line to 105.9% and the 115 kv Littleton to Q195 Tap line to 104.8 % of their LTE ratings. Loss of the K21 overloads the 115 kv Sandbar to Essex K22 Line to 102.6% of its LTE rating, which is mitigated by the Highgate Runback SPS. Under Dispatch 2 peak load conditions (including the sensitivity dispatches with Highgate online and the block load on either Vermont or HQ), loss of the Barre 115/34 kv X63 transformer and F206 Line resulted in overloading 34.5 kv circuits in central Vermont, which are outside the local study area. For the K19 and Fairfax 115/34 kv X67 transformer contingencies, overloads involving the 34 kv Nason St to Nason V line were reported. Additionally the St Albans 115/34 kv transformer contingency overloads the parallel St Albans to Nason 115-34 kv transformer. For the Sensitivity Dispatch 4 with Highgate online and the block load on HQ, loss of the F206 Line also resulted in loading the Lowell VEC to VEC 21 Tap line to just over 100% LTE. 5.2 QP 311 In-Service When the Project is in service approximately 42 MVAr in reactive power losses were reported. These losses occur from the Jay Tap Switching Station to the point of interconnection and include the Project collector strings and transformer losses. Table 5-2 demonstrates where the losses occur for QP 311. Location MVAr Loss Jay Tap Switching Station to VEC Lowell 46 kv Line Losses 21.4 Project Interconnection 46 kv Line Losses 6 Project Collector, Project 46/34.5 Transformer and GSU Losses 14.3 Total Reactive Power Losses 41.7 Table 5-2 Losses Due to Addition of Q311 Project As discussed in Section 1.3, the Vestas V90 units are capable of providing approximately 0.6 MVAr per machine for a total of almost 13 MVAr for the entire wind farm. Along with the addition of a 4 MVAR dynamic reactive device as discussed in Section 3.5, the Project compensates for less than half of the total increase in losses when the Project is at full output. Another change impacting voltage/reactive power performance with the Project is the change in the 12W switch position at North Troy to Normally Open. With the 12W open, the four 2.7 MVAr capacitor banks at Jay Tap Switching Station are isolated from the Newport area which 14

exacerbates the low voltage conditions in Dispatch 2 under peak load conditions with Vermont serving the block load in the Newport area. With the 12W open after the Project, reactive support was added at the Newport 46 kv bus to help alleviate these low voltages. The addition of a 5.4 MVAr capacitor bank at the Newport-B2 bus brought bus voltages within the immediate Newport area above pre-project levels as demonstrated in Table 5-3. Newport Area Voltages Post-Project Pre Project Before Newport Cap Newport B-2 46 kv 0.94pu 0.91pu 0.96pu S Bay2 46 kv 0.94pu 0.92pu 0.96pu Coventry 46 kv 0.92pu 0.92pu 0.96pu Moshers Corner 46 kv 0.95pu 0.92pu 0.96pu Newport2 46 kv 0.95pu 0.92pu 0.96pu N. Troy 46 kv 0.95pu 0.92pu 0.96pu Table 5-3 Newport Area Pre Contingency Voltages Post-Project 5.4MVAr Shunt at Newport 46 kv Under peak load conditions, steady state voltage analysis reported violations of reliability criteria post contingency with the Project as originally proposed. Loss of the K42 resulted in the case solution diverging due to voltage collapse. This demonstrated the need for additional reactive resources in order to alleviate the reactive losses resulting from the Project. Additional shunt capacitors were evaluated at VEC Lowell 46 kv substation as well as at Jay Tap substation. Adding an additional 10.8MVAr of shunt capacitors at the Jay Tap Switching Station maintained voltages within criteria for the K42 contingency as demonstrated in Table 5-4. A 5.4MVAr capacitor at VEC Lowell 46 kv substation did not maintain reliability criteria for loss of the K42 Line. The recommendation for the Project is to increase the four 2.7 MVAr capacitor banks to 5.4 MVAr capacitor banks. This could be accomplished by adding or changing capacitor cans in each bank. K 42 Contingency Loss of Highgate St. Albans - Georgia Post-Project No Additional Shunt Pre-Project Capacitor VEC Lowell 46 kv or Jay Tap 115/46 kv SS Post-Project Additional 5.4 MVAr Shunt Capacitor VEC Lowell 46 kv SS Post-Project Additional 10.8 MVAr Shunt Capacitor Jay Tap 115/46 kv SS Highgate 115 kv 0.96pu 0.94pu 0.97pu Jay Tap Switching 115 kv N/A 0.95pu 0.99pu Moshers Tap 115 kv 0.98pu 0.97pu 1.0pu Irasburg 115 kv 0.99pu Case Diverged 0.97pu 1.0pu Jay Tap Switching 46 kv 0.93pu 1.0pu 1.0pu VEC Lowell 46 kv N/A 1.0pu 1.0pu Kingdom 46 kv N/A 1.0pu 1.0pu Table 5-4 Additional Capacitors for K42 Contingency 15

Table 5-5 demonstrates the reactive compensation proposed for the Project. These upgrades are in-service for all the analyses performed with the Project in-service. Location # of Devices Amount Total (MVAr) MVAr Increase Jay Tap Capacitors 4 2.7 10.8 Newport 34 kv Capacitors 1 5.4 5.4 Generator capability 21 0.609 12.8 DVAR 1 ±4.0 ±4.0 Additional Reactive Compensation 25 to 33.0 Table 5-5 Reactive Compensation Proposed for the Project 5.2.1 Voltage Performance QP 311 All Lines In Under shoulder load conditions, steady state voltage analysis reported no violations of reliability criteria for the Project configuration with all lines in-service. Under peak load conditions, steady state voltage analysis reported violations of reliability criteria for the Project configuration with all lines in-service for several 46 kv buses. These violations were seen pre-project and the addition of the Project and associated upgrades do not aggravate these low voltage conditions. QP311 Post Contingency In all dispatches, loss of the 46 kv transmission line from Jay Tap Switching Station to the Project results in high voltages for the remaining substations connected to Jay Tap 46 kv Substation. Voltages of 1.10pu occur and would require post-contingency automatic capacitor switching. Table 5-6 demonstrates the amount of shunts needed to be tripped for acceptable voltage criteria in the shoulder load D1 dispatch. The application of a dynamic reactive device at Jay Tap 46 kv Substation may be needed to control the temporary high-voltage condition based on the Transmission Owner s design criteria. Also in Dispatch 1, the Stowe DCT remains unacceptable but closely matches the pre-project results. Post Contingency Capacitor Switching at Jay Tap Switching SS Device Tripped Jay Tap Switching Voltage Jay Voltage Rich Voltage None 1.11pu 1.10pu 1.10pu 1 Bank of 5.4MVAr 1.09pu 1.09pu 1.08pu 2nd Bank of 5.4MVAr 1.07pu 1.06pu 1.06pu 3rd Bank of 5.4MVAr 1.05u 1.04pu 1.04pu All Banks 1.03pu 1.03pu 1.02pu Table 5-6 Post Contingency Capacitor Switching at Jay Tap Switching SS Under the Dispatch 2 peak load cases (including the sensitivity dispatches), a significant number of voltage reliability criteria violations exist as seen in the pre project case. The results clearly demonstrate that the Project does not significantly impact the area reliability under this dispatch and load condition. 16

Under Dispatch 2 peak load dispatch conditions with the block load on Vermont and Highgate offline, contingencies involving loss of the Irasburg 115/46 kv H38 transformer resulted in solutions diverging due to the loss of voltage support for the Montgomery, Eden Corners and Johnson Substation areas. This is a pre-project condition and the Project as configured has no affect. 5.2.2 Thermal Performance In the three base cases with the Project, two scenarios (Case B and C) were developed with no redispatch in the local area. The third scenario (Case D) was developed with a local redispatch with Swanton removed and Sheffield reduced. Refer to Table 4-3 for specific details. The reasons for including the analysis with multiple redispatch conditions were to determine if local congestion exists and whether the local units used in the redispatch cases were critical to support of load in the various northern Vermont load pockets. By having multiple redispatch cases, the analysis was able to distinguish between an adverse impact and local reliability dependencies on specific generation. QP 311 All Lines In Under Dispatch 1 for shoulder load conditions, steady state thermal analysis reported violations of reliability criteria for the Project in-service configuration when the project was not redispatched with local generation (Cases B and C). In this scenario excess generation in the area occurs and the Comerford to Littleton Tap D204 line overloads to 108% of the normal rating. This is related to area dispatch and is not experienced when local generation is dispatched off to accommodate the Project (Case D). Therefore, local area congestion may require restrictions of generation during off-peak load periods. Under Dispatch 2 for peak load conditions (including the sensitivity dispatches), steady state thermal analysis reported no violations of reliability criteria for the Project configuration with all lines in-service. QP 311 Post Contingency Under Dispatch 1 for shoulder load conditions, loss of the Barre 115/34 X63 transformer resulted in overloads outside the local study area. Other contingencies including loss of the U-199/X-178 Line, B202 Line, K21 Line, K38 Line, K39 Line and K42 Line all resulted in overloading in the study area when the project was not redispatched with local generation (Cases B and C). This is related to area dispatch and is not experienced when local generation is dispatched off to accommodate the Project (Case D). Therefore, local area congestion may require restrictions of generation during off-peak load periods. Under Dispatch 1 for shoulder load conditions, reported loadings increased with the Project for the K38 contingency with local generation dispatched off (Case D). With the Project, some load at Lowell and Jay 17 is transferred to the new 46 kv line that interconnects the Project to the Jay Tap Switching Station. This configuration change combined with generation from the Project and remaining Sheffield generation online results in higher flows on the 46 kv transmission lines between Irasburg and CVPS Johnson Substations. An alternative redispatch with Sheffield offline and Swanton at partial output showed lower loadings on this path. No upgrades are recommended since loading was below the emergency rating and an alternative redispatch scenario relieved the congestion issue. 17

Under Dispatch 2 for peak load conditions (Highgate off and block load on Vermont), when area generation was dispatched off (Case D) to accommodate the Project, the K19, K19 with runback (RB), St Albans 115/34 kv transformer (X61 or X64) and East Fairfax 115/34 kv transformer (X67) contingencies resulted in higher overloads on the Nason V - St Albans and the Nason St - Nason V transmission lines. This is a result of peak load conditions, lack of area generation and Swanton generation reductions. The overloads were similar when local generation was not redispatched to accommodate the Project (Cases B and C). Under the Sensitivity Dispatch 4 (block load being supplied from HQ and the Highgate HVDC converter at full output to stress the K42 Line) when area generation was not dispatched off to accommodate the Project the K38 contingency resulted in the Highgate-St Albans 115kV Line (K42) exceeding LTE rating. Redispatching local area generation (Case D) relieves the loading on the K42. 6 Short Circuit Analysis Short circuit studies were conducted to assess the impact of the Project on fault current levels within the VELCO area. The twenty-one VESTAS wind turbine generators were represented as two separate aggregate machines consisting of eleven and ten units. The wind generators were modeled as an equivalent synchronous generator per manufacturer s technical manuals. The subtransient impedance is used for short-circuit fault calculations from the wind generators. The generator step-up transformers (GSU s) were also modeled as two sets of aggregates by multiplying the base MVA by the number of corresponding units to obtain the new MVA rating; the per-unit impedance is the same for the aggregate as for a single transformer. The Kingdom 46/34.5 kv transformer, along with the collector string impedance was modeled as described in Table 1-2 and Table 1-4. The transformer was modeled as a grounded wye delta with the 34.5 kv side being grounded as per the preliminary one line in Appendix C. The 115 / 46 kv Jay Tap transformer was modeled as a three winding wye wye with a delta tertiary as per VELCO s typical installation and as described in Table 1-2. One line drawings indicate the voltage for many of the buses as 46 kv. The supplied ASPEN model has these same buses modeled as 48 kv. Line impedances were entered as per unit values based on the 48 kv base of the ASPEN model. Following VELCO s short-circuit analysis guidelines, faults were simulated with an assumed pre-fault voltage Flat option, with 1.05 p.u. voltage. For X/R calculations where X was not defined, a value of X = 0.0001 p.u. was used. For calculations where R was not defined, the ANSI X/R ratio was used assuming R = max( X / g, Rc) with Rc = 0.0001pu and g = 125 for generators, 40 for transformers, and 10 for all others. Aspen Oneliner Version 11.5 program was utilized to determine short circuit values. The below gives a comparison of the short-circuit fault duties of affected buses and includes the fault duties with and without the Project in service. 18

Bus Fault Location Bus Voltage (kv) Without Kingdom Wind Generation Three Phase ANSI X/R Single Phase Three ANSI X/R Phase With Kingdom Wind Generation ANSI X/R Single Phase ANSI X/R Three Phase Delta Amps Single Phase Jay 115 3001 5.4 2870 5.1 3396 5.7 3090 5.2 395 220 Jay Tap 48 3263 10.1 3924 10.2 4390 9.4 4888 9.8 1127 964 Jay 17 48 2527 7.9 2746 5.5 3782 8.1 3563 5.1 1255 817 VEC/Lowell 48 1592 6.2 1651 4 3357 8.5 2530 3.7 1765 879 Kingdom 48 1335 5.8 1159 4.4 3569 11.3 1772 4.4 2234 613 Kingdom 34.5 1552 6.8 2150 7.2 6218 14.3 6342 16.6 4666 4192 Table 6-1 Kingdom Community Wind Short-circuit Fault Duties Fault duties are low and should be considered in engineering design studies of the new 46 kv transmission system supporting the Project. No issues at these levels are expected. 7 Delta V on Capacitor Switching With the recommended change in capacitor size following the addition of the Project, a study was conducted to determine the relative change in voltage (Delta V) on capacitor switching. Based on the available short-circuit duty listed in Table 6-1 above, the short-circuit MVA at the Jay Tap 46 kv bus without the Project is 260 MVA. Therefore, the maximum capacitor size to remain within 3% voltage change is 7.8 MVAr. The largest capacitors to switch at Jay Tap is 5.4 MVAr and within the acceptable range. The 5.4 MVAr capacitors will have an expected Delta- V of 2%. 8 Cost Estimates for Required Network Upgrades The following is a cost estimate provided by VEC to construct the necessary required network upgrades. The cost projections are based on installed costs and do not include the cost for environmental assessment, permitting or temporary facilities required for outage protection. VEC has communicated that subject to receipt of timely regulatory approval, the upgrades can be implemented in time to fit the planned commercial operations schedule for the Project. 1. Increase from 4 x 2.7MVAr 46kV capacitor bank at the Jay Tap station to 4 x 5.4MVAr bank and associated switching equipment for voltage control. Assume 2 switches per 5.4MVAr capacitor rack (3 switches for 10.8MVAr with CB switching one stage of fixed 2.7MVAr) and 2.7MVAr per layer allowing for the switching on and off of 2.7MVAr of capacitors incrementally if required. Material: $260k (based on VEC Jay Switching Station quote for 10.8MVAr bank) to include 2 switches and 10.8MVAr of capacitors less the estimated cost of 2 steel racks. $100k for 2 additional switches per breaker bay x 2 = $200k. Total Material: $460k. Estimated Labor: $50k. Total Estimated Cost + 20% contingency: $612k. 2. Addition of a 5.4MVAr capacitor at the 46kV Newport-B2 bus. Based on a single 5.4MVAr capacitor bank for VEC Jay Switching Station material. Material: $143k for 1 switch and 5.4MVAr capacitors including the cost of two steel racks each with 2.7MVAr of capacitors on first layer (expandable to 5.4MVAr per rack by installing additional 2.7MVAr of capacitors on second layer). $60k for vacuum circuit 19