Cross Cascades North Study Team Report

Cross Cascades North Study Team Report

Acknowledgements ColumbiaGrid Members & Participants Avista Corporation Bonneville Power Administration Chelan County PUD Cowlitz County PUD Douglas County PUD Grant County PUD Puget Sound Energy Seattle City Light Snohomish County PUD Tacoma Power Enbridge Power Other Contributors Portland General Electric Copies of this report are available from: ColumbiaGrid 8338 NE Alderwood Rd Suite 140 Portland, OR 97220 503.943.4940 www.columbiagrid.org April 2012 Photos provided by: Bonneville Power Administration, Grant County PUD, NW Power and Conservation Council, Seattle City Light, Chelan County PUD, istock Photo. Cover photo: Sam Churchill

Table of Contents Executive Summary Introduction Study Methodology Study Results Conclusions Next Steps APPENDIX A - Study Plan APPENDIX B - Investigation of Post Contingency Shunt Capacitor Switching Assumption APPENDIX C - Generation Displacement Method PV Contingency Results and Contingency Analysis with Critical WOCN PV Basecase APPENDIX D - TCRM Analysis Pg. 1 Pg. 3 Pg. 7 Pg. 9 Pg. 15 Pg. 16 Pg. 17 Pg. 26 Pg. 40 Pg. 64

Executive Summary 1 The West of Cascades North (WOCN) interface is one of two major transmission paths that deliver remote resources from east of the Cascade Mountains to west-side loads. WOCN is particularly important to serving the Puget Sound area in Washington and for transfers to British Columbia. The WOCN typically loads most heavily during unusually cold winter weather events. Otherwise, the path tends to be loaded well below its transfer limit of about 10,200 megawatts (MW). Beginning in 2009, ColumbiaGrid s annual system assessment raised a potential reliability concern on the WOCN path. The 2010 and 2011 system assessments again documented the possibility that a critical outage on WOCN during the winter could jeopardize voltage stability. When remote renewable generation (wind) is increased in the system planning models to displace west-side thermal generation (Centralia and others), heavy loading on the path occurs more frequently even during normal winter weather. ColumbiaGrid formed a Cross Cascades Study Team in 2010 as a result of the WOCN reliability concerns and the length of time it could potentially take to address them. The purpose of the team was to study and calculate how the path-loading limit could be increased using various transmission projects. The calculation methods used, which are explained in this report, incorporate both load growth and westside generation displacement. In the 10-year planning case modeled in the 2011 System Assessment, there were three conditions that exceeded the voltage stability limit during transmission outage contingencies on the WOCN: Peak 10-year winter loads West-side thermal generation displaced by remote east-side wind or hydro generation Limited post-contingency reactive response. The Cross Cascades Study Team s work was limited to comparing the Available Transfer

Capability (ATC) benefits of alternatives to increasing the WOCN interface voltage stability limit. The team initially identified projects with short (two to five-year) implementation times. Alternatives that involve building new lines entail a longer implementation time (more than five years) and were studied with the assumption the initial short-term projects were completed. The study team also assumed a suite of projects that is planned by the Puget Sound Area Study Team (PSAST) will have been completed, along with several other thermal or voltage mitigation projects. In addition to analyzing the ATC benefits of the projects, the study team used the Transmission Curtailment Risk Measure (TCRM) to evaluate the effect on the Northern Intertie and Puget Sound area of new line options during a wide range of operating conditions. The TCRM score indicated how the new lines would interact with the existing system. Two of the initial projects with short implementation times would increase the WOCN transfer limit. These projects are adding 24-ohm series capacitors on the Schultz-Raver 500kV #3 and #4 lines and reconductoring 26 miles of 2.5 conductor section on the Schultz-Raver #4 line. The longer-term projects are expected to be more complex. To fully realize the benefits of new cross Cascades line alternatives, like Chief Joe-Monroe #2 500kV, additional mitigation will be required outside of the WOCN interface, including addressing additional Portland area 115kV line overloads and voltage stability limits in the I-5 corridor. Based on its analysis and sensitivity studies, the Cross Cascades Study Team reached the following conclusions: The Chief Joe-Monroe #2 option showed more benefit than other options that added a new cross Cascades transmission line. It was followed closely by the Sickler- Monroe 500 kv line. New transmission line options that terminate at Monroe Substation (Chief Joe-Monroe #2 or Sickler-Monroe) reduce the west-side south-to-north TCRM and increase the north-to-south TCRM on the Northern Intertie (the higher the TCRM score, the higher the risk of curtailment). Significant mitigation outside of the WOCN interface is needed in order for new line projects studied to yield entire reported increases in the WOCN interface ATC. The WOCN load-increase study method causes Monroe Substation to be the critical bus, which is where voltage instability first occurs. The west-side generation displacement method caused Raver to be the critical bus, with Monroe close behind. The outage contingencies that have the most impact on voltage stability are those that occur on the 500kV WOCN main grid. Maintaining the transfer limit on the path is dependent on reactive compensation, such as shunt capacitors, located close to the reactive losses in the 500kV WOCN main grid. The study team recommends (1) a study of conditions that could trigger the need for a project to increase ATC on the WOCN, such as load, generation, and interchange patterns and (2) pre-requisite mitigation projects outside of the WOCN interface be addressed. This will include improvements to simulation models, in particular load representation for the mid-term dynamics simulation. 2

Introduction The West of Cascades North (WOCN) interface is one of two major transmission paths that deliver remote resources from east of the Cascade Mountains to west-side loads. WOCN is particularly important to serving the Puget Sound area in Washington and for transfers to British Columbia. The WOCN typically loads most heavily during unusually cold winter weather events. Otherwise, the path tends to be loaded well below its transfer limit of about 10,200 megawatts (MW). Beginning in 2009, ColumbiaGrid s annual system assessment raised a possible reliability concern on the WOCN path. The 2010 and 2011 system assessments again documented the possibility that a critical outage on WOCN during the winter could jeopardize voltage stability. When remote renewable generation (wind) is increased in the planning models to displace west-side thermal generation (Centralia and others), heavy loading on the path occurs more frequently even during normal winter weather. ColumbiaGrid formed a Cross Cascades Study Team in 2010 as a result of the WOCN reliability concerns and the length of time it could potentially take to address them. The purpose of the team was to study and calculate how the path-loading limit could be increased using various transmission projects. The calculation methods used, which are explained in this report, incorporate both load growth and westside generation displacement, including the permanent closure of the coal-fired units at Centralia, Washington. The WOCN path consists of the Chief Joe-Monroe, Schultz-Raver #1, #3 and #4 and the Schultz-Echo lake 500 kv lines; the Chief Joe-Snohomish #3 and #4 and Rocky Reach-Maple Valley 345 kv lines; the Coulee-Olympia 300 kv line; and the Rocky Reach-White River and Bettas Road-Covington 230 kv lines. The transfer limit on this path is about 10,200 megawatts (MW) with the limit due to voltage stability limitations in the Puget Sound area. 3

In the 10-year planning case modeled in the 2011 System Assessment, there were three conditions that exceeded the voltage stability limit during transmission outage contingencies on the WOCN: Peak 10-year winter loads West-side thermal generation displaced by remote east-side wind or hydro generation Limited post-contingency reactive response 1. The Cross Cascades Study Team s work was limited to comparing the incremental Available Transfer Capability (ATC) benefits of alternative approaches to increasing the WOCN interface capacity. The team initially identified projects with short (two to five-year) implementation times. Alternatives that involved building new lines entail a longer implementation time (more than five years) and were studied with the assumption the initial projects were completed. In its analysis, the study team assumed a suite of projects that is planned by the Puget Sound Area Study Team (PSAST) will have been completed. These projects included the following: Raver 500/230kV transformer with a Raver-Covington 230kV line; SCL 6 ohm series inductors at Broad Street; Lakeside 230kV addition; Bothell-Snoking 230kV high temperature wire reconductor; and Delridge-Duwamish 230kV upgrade. In addition, the Portal Way 230/115kV transformer #2 was modeled in the generation-displacement method, but excluded in the load-increase method and the TCRM calculations because the 2nd Portal Way bank is not in the final suite of planned PSAST south-to north-reinforcement projects. The team also assumed several other thermal or voltage mitigation projects would be operational prior to the WOCN improvements. These projects were: PGE Murrayhill-St.Marys 230kV overload (reconductor assumed; not needed if Castle Rock-Troutdale is done); Bellingham area 230kV voltage stability limit (post-contingency dynamic shunt reactive assumed); and Olympia 230kV voltage stability limit (post-contingency dynamic reactive assumed). In addition to analyzing the ATC benefits of the projects, the study team used a Transmission Curtailment Risk Measure (TCRM) to evaluate the effect on the Northern Intertie of the new line options during a wide range of operating conditions. The TCRM score indicated how the new lines would interact with the existing system, i.e., the higher the TCRM score, the higher the risk of curtailment. Additional results of the TCRM analysis are included in the Study Results section. A complete list of the project alternatives is included on Table 1. The table also includes the incremental ATC benefits and the associated TCRM results for the Puget Sound area. 1 The post-contingency shunt reactive response was limited to the Maple Valley and Keeler static var compensators (SVCs) and the SVC-controlled 230kV shunt capacitors. All other switched shunt capacitors were assumed to be fixed at pre-contingency levels. 4

Table 1. Project Case and Results of ATC and TCRM Project Case Incremental Westside gen displaced or load increased [WOCN ATC increase above REFERENCE] Puget Sound Area TCRM GEN Displaced Load Increased N-S S-N PSAST Projects In-REFERENCE CASE 0 [0] 0 [0] 45,525 20,863 Schultz-Raver 3&4 series capacitor project 1338 [617] 395 [232] Schultz-Raver 3&4 series capacitor project & Schultz-Raver #4 reconductor 1978 [904] 885 [518] 43,880 15,291 All projects below have new Schultz series caps & Schultz-Raver #4 reconductor Chief Joe-Monroe #2 500 kv, 2 EchoLake Caps 3038 [1480} 2893 [1830] 59,326* 12,014 Chief Joe-Monroe #2 500 kv & Remove one 345 kv, Monroe-Snoh 230 kv added 2232 [1056] 2444 [1533] 62,608* 16,769 Chief Joe - Monroe #2&3 500 kv & Remove both 345 kv, Monroe-Snoh 230 kv added 3038 [1480] 2462 [1574] 370,694* 23,936 Sickler-Monroe 500 kv 2303 [1080] 2524 [1565] 53,823* 5,189 Sickler-Monroe 500 kv, convert Coulee-Schultz section of Coulee-Oly 287 kv to 500 kv 3093 [1500] 2451 [1707] 51,487* 5,189 Sickler-Monroe 500 kv with Chief Joe-Monroe #1 series caps 2638 [1300] 2849 [1753] Sickler-Monroe 500 kv, convert Coulee-Schultz section of Coulee-Oly 287 kv to 500 kv with Chief Joe-Monroe #1 series caps 2918 [1426] 3039 [1866] Upgrade Rocky Reach-Maple Valley 345 kv to 500 kv, 2 EchoLake Caps 2657 [1304] 1531 [990] 47,579 69,630 Upgrade Rocky Reach-Maple Valley 345 kv to 500 kv, convert Coulee-Schultz section of Coulee-Oly 287 kv to 500 kv 3145 [1524] 2224 [1094] 46,590 136,934* Upgrade Coulee-Olympia 287 kv to 500 kv 2412 [1153] 1374 [905] 40,504 19,651 Sickler-Raver 500 kv 2606 [1280] 1815 [874] 39,691 41,909* Sickler-Raver 500 kv, convert Coulee - Schultz section of Coulee-Oly 287 kv to 500 kv 2835 [1380] 1526 [982] 37,419 48,974* 5 *TCRM value could be reduced with additional prerequisite projects.

Table 2 provides a summary description of the length, size, configuration, and electrical parameters assumed for each project included in the study. Table 2. Summary of Assumed Project Study Parameters Project Schultz Series Capacitors Description 24 ohms on Schultz-Raver #3 Resistance (p.u.) Reactive Susceptance Winter MVA contingency rating -0.0096 4676 Schultz Series Capacitors Schultz Series Capacitor Increase and Schultz-Raver #4 reconductor Schultz Series Capacitor Increase and Schultz-Raver #4 reconductor Chief Jo-Monroe #2 16 ohms on Schultz-Raver #4-0.0063 4676 line reconductor 0.00094 0.0185 1.356 3897 24 ohm series capacitor on Schultz-Raver #3 120 miles of 3 bundle Deshutes conductor -0.0096 4676 0.00122 0.02733 2.257 5152 Chief Jo-Monroe #2 Chief Jo-Monroe #2 (also Sickleronroe with CJ series caps) Sickler - Monroe 24 ohm series capacitor on #2 line 24 ohm series capacitor on #1 line 80 miles of 3 bundle Deschutes conductor -0.0096 4676-0.0096 4676 0.00087 0.0195 1.648 5152 Sickler - Monroe 24 ohm series capacitor -0.0096 4676 Sickler - Echo Lake 56.5 miles section 1 0.00056 0.0126 1.05698 4415 Sickler - Echo Lake 56.5 miles section 2 0.00056 0.0126 1.05698 4415 Sickler - Echo Lake Sickler series capacitor -0.01764 4415 Sickler -Raver Sickler - Raver Sickler - Raver Sickler -Raver Sickler - Raver Sickler - Raver 57 miles section 1 with 3 Deschutes conductor 58 miles section 2 with 3 Deschutes conductor 44 ohm 34% comp at Cascade One side of Dbl Ck at 230 kv section 1 One side of Dbl Ck at 230 kv section 2 One side of Dbl Ck at 230 kv section 3 0.0012 0.0256 2.1471 5153 0.0012 0.0256 2.1471 5153-0.01764 5153 0.0027 0.0601 0.2252 419 0.002 0.0453 0.1699 419 0.0012 0.027 0.1011 419 Additional detail about the study methodology and results is presented in the follwoing sections of the report. 6

Study Methodology To compare the incremental ATC benefits for short and long lead-time WOCN projects, the team developed a study plan, which is included as Appendix A. The team used two methods to test the system model. One method displaced westside generation with remote wind generation to serve load, and the second method scaled the load upward, with the load increase served by remote hydroelectric generation. The performance metric for comparing the projects was the ATC increase provided on the WOCN interface. For each WOCN project a generation/load pattern was determined when the WOCN voltage stability limit was reached with that project included in the path. The generation/ load pattern was then duplicated on the reference topology and the interface flow was measured (see Appendix A section on Performance Metrics). The study team had planned to use the latest available model for the 10-year case. However, the 10-year winter peak general planning case (22HW1) was not yet ready, and the 5-year 2015-16 Heavy Winter Western Electricity Coordinating Council (WECC) case (16HW2) that was approved January 2011 was used instead, with study team revisions. In the 5-year model, the forecast load level was too low to compare the WOCN voltage stability limits for new line alternatives. To address this for the generation displacement method, the team increased west-side loads to approximate a 2024-25 winter level 2 served by increasing Columbia and Snake River hydro generation. The simulation stressed the transmission system to the WOCN voltage stability limit for the two methods using the PowerWorld Simulator version 16 PV (i.e., power versus voltage) tool. The limiting transfer calculated by the PV tool was confirmed to be at the WOCN voltage stability limit by using the QV (i.e., reactive versus voltage) tool at critical buses. The team simulated the post-contingency shunt capacitor switching, which raised the critical voltage at Raver, Echo Lake, and Monroe substations above 525kV. As a result, the final power flow solutions calculated during PV simulations showed the maximum transfer capacity exceeding actual limits because of lowvoltage solutions 3. After analyzing the monitored bus self dv/dq 4 quantities from the PV simulation, the team used QV simulations to confirm the too-high PV calculations. If the calculated maximum transfer was too high, the valid maximum transfer was found by trial and error, i.e., estimate a maximum transfer, model the outage, and then perform the QV calculation to determine if the estimate is correct. For all of the new line options under the generation displacement method, a potential new third Echo Lake 500kV shunt capacitor was assumed to be available if it could be used post contingency. If the simulation showed this new capacitor (nine total shunt capacitors on at Raver, Echo Lake, and Monroe substations) was available in the critical 7 2 This was done by scaling known non-fixed loads 13.2 percent higher, excluding the Olympic and Kitsap peninsulas. These loads were not scaled due to local voltage instability limits that occur ahead of main grid limits. 3 A low-voltage solution is an unstable operating point and deemed an unacceptable solution. The valid solution would result in over-voltages with 500kV voltages above 1.1 per unit at the higher transfer. 4 A dv/dq quantity is the amount of a small change in voltage divided by the corresponding change in reactive at a bus.

post-contingency condition, the simulation was repeated without it to quantify the incremental benefit. A Northwest-wide contingency analysis (using over 4,000 contingency definitions) was performed to determine whether transmission line or transformer ratings could cause a transfer to be limited ahead of the WOCN voltage stability limit. The contingency analysis was run on the base cases before the system was stressed and on the stressed base case at the WOCN voltage stability limit. The two results were compared to identify overloads caused by the transfer. All of the overloads identified in the comparison occurred outside the WOCN interface. Projects that would mitigate the overloads were modeled However, mitigation projects outside of WOCN were planning assumptions and not studied for adequacy. The study team discussed modeling assumptions that could affect the findings. To eliminate uncertainty, the team developed and conducted a list of sensitivity studies. These studies were performed on the base case model with the PSAST projects included. The effect of each sensitivity study on the WOCN Total Transfer Capability (TTC) was subsequently calculated, and the results are provided toward the bottom of Table 3 in the following section, Study Results. The study plan also included simulation of an approximate mid-term dynamics scenario with new WECC load model dynamic data and recently developed modeling tools 5. The Time Step Simulation tool was used to approximate the postcontingency voltage during the time scenario, with voltage and time responses of loads, shunt capacitors, and load voltage regulators over a period up to 20 minutes. The mid-term dynamics associated with the WOCN TTC calculation was initially addressed in the Centralia Closure Study in early 2011. That study quantified the effect of the WOCN TTC calculation with the planning assumption that 500kV shunt capacitors would reliably switch post contingency. It provided a preliminary simulation that showed the shunt capacitors could likely be planned to switch on after the contingency and switch off after system restoration. The Centralia closure report recommended the mid-term dynamics associated with this planning assumption for 500kV shunt capacitor switching be updated with the WOCN study. Appendix B contains the details of the simulation update. The simulation continued to show that the shunt capacitors will likely switch post contingency, but the team recommends repeating the study as load and other data corrections and updates are obtained. The supporting data for the generation displacement method with PV contingency results is contained in Appendix C. Full supporting data with additional simulation monitoring and setup files for reproducing results is posted on the Cross Cascades Study Team website 6. The Cross Cascades study plan also included TCRM analysis to determine the impact of the proposed projects on the Puget Sound area and the Northern Intertie. Details of the analysis and results can be found in the following section, Study Results, and Appendix D. The studies found that projects with line additions on WOCN would have an impact on the Northern Intertie and identified prerequisite projects to mitigate impacts. 5 One of these tools is the PowerWorld Simulator version 16 feature that converted the composite load model dynamic data to the power flow model with load voltage regulators, distribution feeder impedances, and static load models (constant MVA, constant current, and constant impedance). 6 The data and setup are available in large compressed folders. The content is in PowerWorld Simulator version 16 pwb and aux file format. 8

Study Results An overview of the results shows the study team found two of the initial projects with short implementation times would increase the WOCN transfer limit. Adding 24-ohm series capacitors on the Schultz-Raver 500kV #3 and #4 lines. Reconductoring 26 miles of 2.5 conductor section on the Schultz-Raver #4 line. The longer-term projects posed a more complex challenge. To fully realize the benefits of new cross Cascades line alternatives, like the Chief Joe-Monroe #2 500 kv line, it will take additional mitigation outside of the WOCN interface to remedy the following: Additional Portland area 115kV line overloads (reconductor assumed in study). I-5 voltage stability limits for the Category B Raver-Paul 500kV outage (mitigation not explored in study). I-5 Category C Paul-Allston & Napavine- Allston 500kV outage (Castle Rock-Troutdale project would mitigate). Category C John Day-Marion/Buckley- Marion outage (Cascade Crossing project would mitigate). The detailed study results for the generation displacement method are shown in Table 3. The table also includes the sensitivity study results. Key to Table 3 Columns 1. PROJECT - brief description of study option. More study options are listed than shown in the executive summary 2. WOCN Flow After Project Removed - power flow study result with the maximum westside generation displacement calculated with the study option on the present system topology. This calculation is needed to determine the project performance metric described in the study plan, which is ATC Increase column in the next column. 3. ATC Increase - reference WOCN flow (10,718 MW) subtracted from the WOCN Flow After Project Removed. This is the project performance metric described in the study plan. 4. Gen Shift the magnitude of wind generation increased to displace the westside thermal generation in the PV simulation that determined the voltage stability limit for the study option described under PROJECT. The Gen Shift was duplicated in the reference case to calculate the WOCN Flow After Project Removed. 5.Incremental Gen Shift reference Gen Shift subtracted from Gen Shift for the study option under PROJECT. The Incremental Gen Shift can be interpreted as an alternative project performance metric because it quantifies the system benefit (generation displacement performance) rather than the specific interface ATC benefit. 6. Caps-Raver number of post contingency capacitor banks on at the voltage stability limit from the PV simulation. 7. Caps-EchoLake number of post contingency capacitor banks on at the voltage stability limit from the PV simulation. The existing number of capacitor banks at EchoLake is two. If three are shown, then the project benefit would include adding another capacitor bank for the study option. 9

8. Caps-Monroe number of post contingency capacitor banks on at the voltage stability limit from the PV simulation. Three capacitor banks exist at Monroe. 9. Limiting Contingency the WOCN interface contingency, or adjacent contingency involving lines connected to Schultz, that determined the voltage stability limit. 10. V at Qmin the critical voltage at Raver (which was always the critical bus) from the QV simulation performed on the PROJECT representing the limiting condition of Gen Shift and Limiting Contingency. The QV simulation was intended to confirm that the PV simulation reached the voltage stability limit. 11. Qmin the lowest reactive power injected at the critical bus in the QV simulation. Ideally, the number should be close to zero. A large negative number (e.g. beyond -200 Mvars) could indicate the project performance metric (ATC increase or Incremental Gen Shift) could be too low. For all study options, the Qmin was close enough to zero to validate the PV simulation result that determined the project performance metric. 12. Comment notes about study results, e.g. low voltage solution issues, limiting contingencies outside of WOCN, and the critical bus. 10

Table 3. Generation Displacement Method Study Results PROJECT WOCN Flow After Project Removed ATC Increase Gen Shift Incremental Gen Shift (see Note 7) Increase westside load to approximate 10 years load growth PSAST Projects Not In (16HW2 loads-no load increase) 10543 3 PSAST Projects Not In 10403 1143 3 PSAST Projects In - REFERENCE CASE 10718 1962 3 Schultz-Raver 3&4 series capacitor project 11350 617 3300 1338 3 Schultz series capacitor size increase & Schultz-Raver #4 reconductor 11645 904 3940 1978 3 All Projects Below have Schultz Series Caps & Schultz-Raver #4 upgrade Chief Joe-Monroe #2 500kV (with series compensation) 12335 1578 5262 3300 3 Caps- Raver Chief Joe-Monroe #2 500kV, 2 EchoLake Caps 12235 1480 5000 3038 3 Chief Joe-Monroe #2 500kV & Remove one 345kV,Monroe-Snoh 230kV added 11800 1056 4194 2232 3 Chief Joe-Monroe #2&3 500kV & Remove both 345kV,Monroe-Snoh 230kV added 12235 1480 5000 3038 3 Sickler-Monroe 500kV (with series compensation) 11825 1080 4265 2303 3 Sickler-Monroe 500kV, add CJ-Monroe#1 series compensation 12050 1300 4600 2638 3 Sickler-Monroe 500kV, convert to 500kV Coulee-Schultz section of Coulee-Oly 287kV 12255 1500 5055 3093 3 Sickler-Monroe 500kV, convert CouOly, add CJ-Monroe#1 series comp 12180 1426 4880 2918 3 Upgrade Rocky Reach-Maple Valley 345kV to 500kV 12110 1358 4740 2778 3 Upgrade Rocky Reach-Maple Valley 345kV to 500kV, 2 EchoLake Caps 12055 1304 4619 2657 3 Upgrade Rocky Reach-Maple Valley 345kV to 500kV, convert Coulee-Schultz section of CouOly 12280 1524 5107 3145 3 Upgrade Coulee-Olympia 287kV to 500kV 11900 1153 4374 2412 3 Sickler-Raver 500kV 12030 1280 4568 2606 3 Sickler-Raver 500kV, convert to 500kV Coulee-Schultz section of Coulee-Oly 287kV 12133 1380 4797 2835 3 Sensitivity Studies with PSAST Projects In as Reference Shunt capacitors OFF below 115kV westside (1052 Mvars), except OlyPen 10485-227 1292-670 3 Unity PF loads westside (4179 Mvars) 10950 226 2365 403 3 Reverse west side gen displacement (South to North) 11223 493 3300 1338 3 Columbia&Snake Hydro as source 10675-42 1775-187 3 CastleRk-Troutdale Project IN 10850 129 2162 200 3 Cascade Crossing Project IN 10805 85 2100 138 2 Cascade Crossing Project+CastleRk-Troutdale 10840 119 2150 188 3 Remove PSAST Portal Way project 10730 12 1977 15 3 Notes 1. TTC numbers include the Category C 2.5% power margin power margin specified in WECC criteria. PV results show Cat B was never more limiting than Cat C even with more stringent 5% margin. 2. ATC increase on WOCN Interface with present day topology is the performance metric for project comparison 3. An alternative system performance metric is to compare the Gen Shift which measures the remote generation that can displace westside generation and assurances constraint mitigation outside WOCN 4. The calculations only reflect the genreation displacement method and not the load scaling method 5. V at Qmin is the critical bus in QV which was always at RAver for WOCN outages and Coulee-Schultz 1&2 6. Critical bus in QV for Schultz-Wautoma & Schultz-Vantage was Allston 7. Incremental Gen Shift column is the amount of westside gen displacement the project enables over the reference case (which is 1962 MW) Appendix C contains PV simulation results and full contingency analysis results 11

Caps- EchoLake Caps- Monroe Limiting Contingency V at Qmin (see Note 5) Qmin Comment 2 2 N-2: Schultz-EL/Raver 500kV 1.0503-73 1 3 N-2: Schultz-EL/Raver 500kV 1.0038-431 1 3 N-2: Schultz-EL/Raver 500kV 1.024-196 1 3 N-2: Schultz-EL/Raver 500kV 1.0417-55 2 3 N-2: Schultz-EL/Raver 500kV 1.06 0 3 3 N-2: Schultz-EL/Raver 500kV 1.07-20 Minor LV solution,raver-paul outage limiting 4880 Allston critical, also WOCS limit 4880 shift 2 3 N-2: Schultz-EL/Raver 500kV 1.065 0 3 3 N-2: Schultz-EL/Raver 500kV 1.065-100 Also at limit for SchultzWautoma-SchultVantage and WOCS voltage stability limits 2 3 N-2: Schultz-EL/Raver 500kV 1.07 0 Major LV solution on original PV. Coincides with JDMarion- BuckleyMarion outage voltage stability limit 2 2 N-2: COULEE-SCHULTZ 1&2 1.06-15 2 2 N-2: Schultz-Vantage/Wautoma 1.06 0 Monroe voltage at critical state was 1.095 pu 3 2 N-2: Schultz-EL/Raver 500kV 1.07-10 Major LV solution on original PV. Coincides with or exceeds voltage stability limits outside WOCN 3 2 N-2: Schultz-EL/Raver 500kV 1.07-40 LV solution on original PV 3 3 N-2: C.JO-MON & C.JO-SNO 345 3or4 1.06-5 2 3 N-2: C.JO-MON & C.JO-SNO 345 3or4 1.06-2 3 3 N-2: Schultz-EL/Raver 500kV 1.07-2 2 3 N-2: C.JO-MON & C.JO-SNO 345 3or4 1.06-1 3 3 N-2: C.JO-MON & C.JO-SNO 345 3or4 1.06 0 3 3 N-2: C.JO-MON & C.JO-SNO 345 3or4 1.06-10 Major LV solution on original PV 2 3 N-2: Schultz-EL/Raver 500kV (128Mvars OlyPen 115 caps on) 1.05 0 2 2 N-2: Schultz-EL/Raver 500kV 1.02-280 2 3 N-2: Schultz-EL/Raver 500kV 1.07 0 1 3 N-2: Schultz-EL/Raver 500kV 1.02-160 2 2 N-2: Schultz-EL/Raver 500kV 1.04-150 2 3 N-2: Schultz-EL/Raver 500kV 1.04-24 2 3 N-2: Schultz-EL/Raver 500kV 1.03-220 1 3 N-2: Schultz-EL/Raver 500kV 1.03-210 12

Table 4. Load-Increase Method Study Results Project nose Limit w/o Proj REFERENCE CASE (with 3rd Monroe Shunt Cap) 11049 10779 11049 add Schultz-Raver 3&4 series caps 11358 11081 11289 add Schultz-Raver 3&4 series caps AND 3rd Echo Lake shunt cap 11551 11269 11452 add Schultz-Raver 3&4 series caps (upgraded Schultz-Raver #4 line and series cap) 11691 11406 11583 All projects below include the Schultz-Raver 3&4 series capacitors and upgraded Schultz-Raver #4 500-kV line Sickler-Monroe 500kV 12721 12411 12655 Sickler-Monroe 500kV (+) Coulee-Schultz #3 500kV 12872 12558 12800 Sickler-Monroe 500kV with series caps on CHJ-MON #1 12942 12626 12848 Sickler-Monroe 500kV (+) Coulee-Schultz #3 500kV with series caps on CHJ-MON #1 13068 12750 12964 Coulee-Schultz-Olympia 500kV 12183 11886 11982 Chief Joe-Monroe #2 500kV 12971 12654 12926 Chief Joe-Monroe #2 500kV also remove one 345kV CHJ-SNH line 12638 12329 12621 Chief Joe-Monroe #2 500kV also remove one 345kV CHJ-SNH line and add one MON-SNH 230 12674 12365 12651 Chief Joe-Monroe 2&3 500kV also remove both 345kV CHJ-SNH lines 12589 12282 12561 Chief Joe-Monroe 2&3 500kV also remove both 345kV CHJ-SNH lines and add two MON-SNH 230 12686 12376 12663 Sickler-Echo Lake 500kV 12137 11841 12065 Sickler-Echo Lake 500kV (+) Coulee-Schultz #3 500kV 12629 12028 12475 Sickler-Raver 500kV 12300 11715 12239 Sickler-Raver 500kV (+) Coulee-Schultz #3 500kV 12153 11857 12058 Sensitivity Cases Reactive part of load is not scaled 11082 10812 11082 The detailed study results for the load-increase method are shown in Table 4. The table also shows the one sensitivity study result. Key to Table 4 Columns 6. TTC increase in the WOCN limit from the reference base case ( TTC = project Limit reference Limit ). 7. ATC increase in available transfer capability (DATC = TTC - Flow). 1. PROJECT - brief description of study option. More study options are listed than shown in the executive summary. 8. Load increase in westside load from the reference case to the PV curve nose point. 2. nose WOCN flow at the final solvable point during the PV curve process and as verified by a QV curve simulation. 3. Limit - WOCN flow after the power margin is removed (2.5% for Category C and 5.0% for Category B). 4. w/o Project WOCN flow at the Project nose point with the project removed from service. 5. Flow difference in WOCN flow that the project creates ( Flow = nose w/o project ). 9. MON number of post contingency capacitor banks on at the voltage stability limit from the PV simulation. Three capacitor banks exist at Monroe 500-kV. 10. ECH number of post contingency capacitor banks on at the voltage stability limit from the PV simulation. Two capacitor banks exist at Echo Lake 500-kV. 11. RAV number of post contingency capacitor banks on at the voltage stability limit from the PV simulation. Three capacitor banks exist at Raver 500-kV. 13

DFlow DTTC DATC DLoad MON ECH RAV V-crit Q-mar Limiting Contingency 0 0 0 2432 3 2 3 1.039-34.69 N-2: Schultz-Echo Lake 500kV & Schultz-Raver #1 500kV 69 301 232 2827 3 2 3 1.030-200.00 N-2: Chief Jo-Monroe #1 500kV & Chief Jo-Snohomish #3 345k 99 490 390 3141 3 3 3 1.034-135.28 N-2: Chief Jo-Monroe #1 500kV & Chief Jo-Snohomish #3 345k 109 627 518 3317 3 2 3 1.027-161.13 N-2: Chief Jo-Monroe #1 500kV & Chief Jo-Snohomish #3 345k 66 1631 1565 4956 3 2 3 1.021-142.90 N-2: Sickler-Monroe 500kV & Chief Jo-Snohomish #3 345kV 72 1779 1707 4883 3 2 3 1.019-132.30 N-2: Sickler-Monroe 500kV & Chief Jo-Snohomish #3 345kV 93 1847 1753 5281 3 2 3 N-2: Custer-Monroe #1 500kV & Chief Jo-Snohomish #4 345kV 104 1970 1866 5471 3 2 3 N-2: Custer-Monroe #1 500kV & Chief Jo-Snohomish #4 345kV 201 1106 905 3806 3 2 3 1.028-67.34 N-2: Chief Jo-Monroe #1 500kV & Chief Jo-Snohomish #3 345k 45 1875 1830 5325 3 2 3 1.015-144.74 N-2: Schultz-Echo Lake 500kV & Schultz-Raver #1 500kV 17 1550 1533 4727 3 2 3 1.049-135.91 N-2: Custer-Monroe #1 500kV & Chief Jo-Snohomish #4 345kV 23 1586 1562 4876 3 2 3 1.011-155.78 N-2: Chief Joe-Monroe #2 500kV & Chief Jo-Snohomish #4 345 28 1502 1474 4815 3 2 3 1.010-67.23 N-2: Chief Joe-Monroe 2&3 500kV 23 1597 1574 4894 3 2 3 1.016-1.88 N-2: Chief Joe-Monroe 2&3 500kV 71 1061 990 3963 3 2 3 1.022-133.73 N-2: Chief Jo-Monroe #1 500kV & Chief Jo-Snohomish #3 345k 155 1249 1094 4656 3 2 3 1.045-1.55 N-1: 3TM Monroe-Echo Lake-SnoKing 500kV 61 935 874 4247 3 2 3 1.025-184.39 N-1: 3TM Monroe-Echo Lake-SnoKing 500kV 95 1077 982 3958 3 2 3 1.023-251.17 N-2: Chief Jo-Monroe #1 500kV & Chief Jo-Snohomish #3 345k 0 32 32 2493 3 2 3 N-2: Schultz-Echo Lake 500kV & Schultz-Raver #1 500kV 12. V-crit the critical voltage at Monroe 500-kV (which was always the critical bus) from the QV simulation performed on the PROJECT representing the limiting condition of load increase and Limiting Contingency. The QV simulation was intended to confirm that the PV simulation reached the voltage stability limit. 13. Q-mar the lowest reactive power injected at the critical bus in the QV simulation. Ideally, the number should be close to zero. A large negative number (e.g. beyond -300 Mvars) could indicate the project performance metric (ATC increase or Incremental Gen Shift) could be too low. For all study options, the Q-margin was close enough to zero to validate the PV simulation result that determined the project performance metric. 14. Limiting Contingency the WOCN interface contingency that determined the voltage stability limit. 14

Transmission Curtailment Risk As mentioned earlier, the study team used TCRM to evaluate the effect of the new line project options on the Northern Intertie. The TRCM analysis showed that new lines connecting into Monroe Substation generally increase the northto-south TCRM (increasing Northern Intertie curtailment risk) and decrease the south-to-north TCRM compared to the pre-project base case. On the other hand, lines connecting into Raver or Echo Lake Substation generally decrease the north-to-south TCRM and increase the south-tonorth TCRM 7. The analysis also indicated that individual TCRM values could be improved with additional mitigating projects. For example, the impact to WOCN projects to north-to-south TCRM could be significantly reduced with the addition of a second Portal Way transformer. For each proposed WOCN line project, additional projects would be needed to reduce the Northern Intertie TCRM for north to south and south to north flows in the Puget Sound area. The team identified the following potential projects to mitigate the impact of the WOCN projects on the Northern Intertie. a) Chief Joe-Monroe #2: The impact on northto-south TCRM can be mitigated by projects including a second Portal Way transformer. A sensitivity study showed that the resultant TCRM with Portal Way for the base case was 15,060 and for the Chief Joe-Monroe #2 was 19,137. c) Sickler-Monroe 500 kv with series compensation: High TCRM can be mitigated by projects including a second Portal Way transformer. d) Upgrade of Rocky Reach-Maple Valley to 500kV: This proposed line addition would either need to have a breaker failure contingency resolved, or a second Maple Valley 500/230kV transformer installed to reduce South to North TCRM. e) Sickler-Raver 500 kv addition: The impact on South to North TCRM can be mitigated by projects including a Sammamish-Sammamish E 230kV switch upgrade. Conclusions The Cross Cascades Study Team reached several conclusions based on the analysis of projects to increase the transfer limits on the WOCN path and the sensitivity studies. These include: 1. The Chief Joe-Monroe #2 option showed more benefit than other study options for adding a new cross Cascades transmission line. However, the Sickler-Monroe 500kV option was close. 2. New transmission line options that terminate at Monroe Substation (Chief Joe-Monroe #2 or Sickler-Monroe) reduced the west-side Northern Intertie south-to-north TCRM and increased the north-to-south TCRM. b) Chief Joe-Monroe #2&3 500kV option that removes both 345kV lines and adds 230kV between Monroe and Snohomish: This project will likely need a second Monroe 500/230kV transformer. 3. Mitigation outside of the WOCN interface is needed in order for the projects studied to yield increases in the WOCN interface ATC. These prerequisite projects include: 15 7 The PSAST is recommending a north-to-south plan to address Puget Sound reliability issues. For the most part, only the approved south to-north PSAST reinforcement projects were included in the TCRM base case analysis for the WOCN. The baseline TCRM analysis did not include a second Portal Way transformer. A low-cost north-to-south project, the Monroe-Novelty 230 kv line upgrade to 100C, was modeled in the TCRM base cases.

a. The PSAST suite of projects to mitigate overloads caused by south-to-north flow on the west side of the Northwest-British Columbia Path (Path 3). b. Reinforcement projects for north-tosouth flow on the west side of Path 3 will be needed to mitigate the impact of the cross Cascades reinforcement options on north-to-south transfer capacity and TCRM. Several projects have already been identified in the Bonneville Power Administration s NOS2010 studies and in other studies done by the PSAST. c. In the Portland area, the Category B Keeler-Pearl 500kV outage overloads the Murrayhill-St Marys 230kV line prior to realizing any benefit from initial WOCN projects. Subsequent new line WOCN projects increase the number of line overloads in the Portland area and need mitigation to obtain the calculated WOCN benefits. The I-5 Corridor Reinforcement project or Cascade Crossing project may be suitable mitigation. 4. The WOCN load-increase study method causes Monroe Substation to be the critical bus, which is where voltage instability first occurs. The westside generation displacement method caused Raver to be the critical bus, however, with Monroe close behind. 5. The voltage stability limits associated with the main grid outage contingencies occur on the 500kV WOCN main grid. Maintaining the transfer limit is dependent on reactive compensation, such as shunt capacitors, located close to the reactive losses in the 500kV WOCN main grid. The WOCN voltage stability limit is not sensitive to lower voltage shunt capacitors or power factor of loads connected at 230kV or below. Next Steps The study objective of this team was to compare incremental WOCN ATC increases for different alternative projects. The study objective did not include addressing the conditions (e.g. load level, Puget Sound generation status, Upper Columbia generation level, post contingency capacitor switching) that would trigger the need for a project. The team suggests (1) that a study be performed with the objective to identify conditions (load, generation, and interchange patterns) that trigger project need and (2) pre-requisite mitigation projects outside of the WOCN interface be addressed. Associated with conditions that trigger project need, the team identified improvements that are needed with the mid-term dynamics simulation that are expected to reconfirm the reliable post contingency 500kV shunt capacitor switching assumption. The needed improvements are: (1) The composite load model for the motor loads in the powerflow simulation need to be changed by determining a second order polynomial curve fit equation based on transient stability output of voltage versus load P and Q, (2) Update the powerflow basecase load Long_ID used for the composite load model (Climate zone and load composition, or industrial load type). This data determines the default load composition and component data for the CMPLDW records in the dynamics file. 16

APPENDIX A- Study Plan Study Purpose Cross Cascades North Study Plan 6/27/11, revised 10/05/11 The purpose of this study is to investigate and compare the alternatives to increase the West of Cascades North (WOCN) path capacity. This is primarily a concern during heavy winter loading conditions. Base System Model Base Model: WECC Approved 16HW2a Major Revisions to 16HW2a model: 1. Remove Cascade Crossing project. 2. Remove I-5 Corridor Reinforcement project. 3. Remove Boardman-Hemingway project. 4. Set the westside Northern Intertie to 1500MW south-to-north. 5. Set Puget Sound Area Generation to 680 MW 6. Set Frederickson peaker CTs off 7. Set one unit off at the Centralia 8. Other changes listed in Appendix B Study Methodology General Strategy: CG and BPA will independently perform the system studies starting with the same base conditions and stressing the system model using different accepted methodologies. This approach is favored by the team to divide the labor, provide a cross check for accuracy, and see if the conclusions will be the same. All alternatives prior to a new line will be identified with their corresponding transfer capability increase. Pre-contingency voltage tuning 1. Maximize the pre-contingency switched shunt at 230kV and below on the westside. Change the 500/230kV transformer taps to allow 230kV switched shunt support of pre-contingency 500kV voltage (i.e. maximize off nominal tap ratio) and adjust the 500kV switched shunt to prevent pre-contingency 17

over voltages. Note: the effect of this planning assumption to optimize shunt reactive will be addressed in a sensitivity study. 2. Minimize the reactive generation so the dynamic reactive reserve is available post-contingency. This could involve reducing the voltage schedule at the sending end (e.g. Coulee) which will also help the Schultz series capacitors from exceeding Vmax. Note: the effect of this planning assumption to optimize generator reactive will be addressed in a sensitivity study. 3. Maximize the SVC dynamic reactive reserve and controlled switched shunt availability to ensure maximum post contingency dynamic reactive capability. Post Contingency 1. Shunt capacitor switching: assume reliable post contingency capacitor/reactor switching at 500kV connected shunts. 2. CG will use the generator parameter Xcomp in the WECC Master Dynamics File for setting post contingency LDC_RCC. BPA will assume generators control the voltage on the generator terminal bus in the model (i.e. LDC_RCC = 0.0). 3. Allow system design to result in a very high post contingency critical voltage (above 540kV). Preliminary mid-term dynamics simulations with rough load model approximations suggested this could be reliably achieved. A sensitivity study using updated models and data (the WECC composite load model) is included in this study plan. Methodologies to Stress The Model to the Limiting Condition The study team is interested in two approaches to stress the model to the limiting conditions. Any system limits outside the WOCN interface that are found by the modeled transfer (generation shift or load increase served by remote generation) will have an assumed mitigation project put in the model. For limits found that can be mitigated by the suite of projects identified by the Puget Sound Area Study Team (PSAST), the complete list of PSAST projects will be added to the model. As other constraints are found, system owners participating in this study team will identify assumed mitigation projects. Appendix B will be maintained with a list of files that modify the model with the assumed mitigation projects. 1. Sequentially displace westside generation with wind generation and keep the load constant. CG will use this method. The specific order of west side generation displacement is found in Attachment D. It is generally based on the publicly available TEPPC heat rate data for the gas generation (average for each generation project rather than TEPPC incremental data). For system expansion alternatives that increase the voltage stability limit beyond what can be found with all westside generation displaced, the system will be further stressed by scaling up non-fixed loads (except Olympic/Kitsap Peninsula loads) until the voltage stability limit is found. 18

2. Increase the area Northwest load (except Olympic/Kitsap Peninsula loads) at non-fixed load buses by a uniform percentage and serve by remote generation. All fixed loads are assumed to have a load ID starting with F, I, or S, except a separate list of fixed Pacificorp load buses will be used because Pacificorp does not follow the F and I load ID convention. The Puget Sound Energy, Snohomish PUD, and Seattle City Light generation total is fixed at 680 MW to model historical low values for peak winter conditions. The PSE Frederickson peaker CT s will be off. Hydro generation on the Upper, Mid, and Lower Columbia rivers as well as on the Snake River will be set proportionally based on historical levels and scaled by this proportion to displace any generation changes needed to reach 680 MW in the Puget Sound Area or to accommodate any load changes when finding the WOCN path limit. Performance Metrics Each of the two methodologies described above will use the net ATC increase in WOCN path capacity as a basis for comparison. This is calculated as follows: Net ATC increase = (TTC increase) (Flow increase) TTC increase = (WOCN limit with project) (WOCN limit without project) Flow increase = (WOCN basecase flow with project) (WOCN basecase flow without project) For methodology #1 the WOCN limit will be the path flow at the nose of the PV curve. For Methodology #2 the WOCN limit will be where the QV curve shows no margin. Sensitivity Studies 1. Load powerfactor: change loads to unity power factor to simulate full compensation at the loads. This test will determine if the location of shunt compensation should occur at the loads or rather out on the transmission system 2. Less than optimal shunt reactive at 230kV and below. Switch off about 50% of the 115kV shunt reactive in the Puget Sound area. This will simulate less than optimal shunt compensation at the loads. 3. Generator reactive: Increase pre-contingency generator reactive output at Coulee, Chief Jo, The Dalles, and John Day to reduce post contingency dynamic reactive reserve. This will show the impact of having less dynamic reactive reserves available. 4. For the method #1 (westside generation displaced by wind generation), units will be displaced from south to north. The purpose is to determine if any interaction exists between the WOCN and WOCS transfer paths. 5. for method #2, lock all 500kV switched shunts remote from westside loads. 6. For method #1, units will be displaced with Columbia and Snake River hydro rather than with wind generation (same proportions as method #2). 19

7. Identify TCRM (Transmission Curtailment Risk Measure) impacts of WOCN transfer level on Northern Intertie. 8. Identify impacts with other proposed projects outside the WOCN path (e.g. I-5, Cascade Crossing). Mid-term Dynamics WECC is obtaining data for the new composite load model. After the data accumulation is finished, the preliminary mid-term dynamics simulation will be repeated (the preliminary study assumed 100% constant current load). The preliminary mid-term dynamics investigation led the team to allow design with post contingency capacitor switching and very high critical voltage. The new study will still be considered preliminary, but with updated information. The estimated date for the data and the tools to update the simulation is February 2012. Additional Background: The WECC post transient methodology specifies contingency powerflow simulations are to be performed with (a) constant MVA loads in the absence of the distribution model to represent load conditions several minutes after the contingency and (b) model only post contingency automatic actions that will occur within 3 minutes of the contingency. This CCN study assumes 500kV shunt capacitors will reliably switch on in the post transient period (without loss of load and without hunting). This is a major assumption. It provides over 1000 MW s of additional westside generator displacement capability before a voltage stability limit is reached. However, the scenario with capacitors switching during the post transient time frame has been studied only with approximate and preliminary data, which indicated that reliable post contingency capacitor switching could likely be designed. The new composite load model information will provide the added detail to reconfirm the post contingency capacitor switching assumption. Potential projects for study Initial pre-new line alternatives The study procedure will model all the following projects and remove individual projects to determine if projects do not provide a benefit. 1. Schultz-Raver 3&4 series capacitor project 2. Shunt capacitors near sending end generators (Chief Joe and Coulee) if dynamic generator reactive reserve is fully used post contingency 3. Echo Lake 500kV shunt capacitors if the critical voltage at the voltage stability limit allows more post contingency shunt capacitor switching without exceeding Vmax. 4. Schultz series capacitor size increase with mitigation of thermal limits (Schultz-Raver #4 reconductor) 20

5. Increase size of Schultz-Echo Lake & Schultz-Raver series capacitors 6. Chief Jo-Monroe series capacitors 7. Switched shunt with 50% of standard step size, with more steps Subsequent new line alternatives 1. Chief Joe-Monroe #2 500kV (with series compensation) 2. Sickler-Monroe 500kV (with series compensation) 3. Upgrade one, or both, Chief Joe-Snohomish 345kV to 500kV 4. Upgrade Rocky Reach-Maple Valley 345kV to 500kV 5. Upgrade Coulee-Olympia 300kV to 500kV Thermal Studies For the thermal studies the system will be stressed with the same two methods described in the voltage stability section. The goal will be to find and mitigate limits in the system model where thermal overloads occur due to the modeled transfer using the full contingency definitions list. Study Schedule June 17 - post draft study plan for comment June 21 - post base case changes June 27 - Final Study Plan Aug 2 - Post draft Team Report Outline Aug 3 - All day team meeting at CG to review draft report outline and study results to date Sept 7-2 hour meeting Oct 5-4 hour meeting at ColumbiaGrid Nov 9 - All Day Meeting at ColumbiaGrid Nov 30 - Draft Report Feb 28 - Final Report 21

Attachment A - Interface Definitions The West of Cascades North (WOCN) path will be defined as listed in the WECC path rating catalog and is given below. New lines, under evaluation for WOCN capacity increases, will be added to the path definition as well. Chief Joe Monroe #1 500kV (BPA) Schultz Echo Lake 500kV (BPA) Schultz Raver #1 500kV (BPA) Schultz Raver #3 500kV (BPA) Schultz Raver #4 500kV (BPA) Chief Joe Snohomish #3 345kV (BPA) Chief Joe Snohomish #4 345kV (BPA) Rocky Reach Maple Valley 345kV (BPA) Grand Coulee Olympia 300kV (BPA) Bettas Road Covington 230kV (BPA) Rocky Reach White River 230kV (PSE) The West of Cascades South (WOCS) path will be defined as listed in the WECC path rating catalog and is given below. Big Eddy-Ostrander 500kV (BPA) Ashe-Marion 500kV (BPA) Buckley-Marion 500kV (BPA) Knight-Ostrander 500kV (BPA) John Day-Marion 500kV (BPA) McNary-Ross 345kV (BPA) Big Eddy-McLoughlin 230kV (BPA) Big Eddy-Chemawa 230kV (BPA) Midway-N.Bonneville 230kV (BPA) Jones Canyon-Santiam 230kV (BPA) Big Eddy-Troutdale 230kV (BPA) Round Butte-Bethel 230kV (PGE) Attachment B- Changes to WECC Approved 16HW2a For Base Case: 1_BPA-Covington230Bus.aux 2_BPA-BettasRoadUpgrades.aux Avista-Moscow&WestsideTransformers.aux BPA&PAC-Facebook40MVAload.aux BPA-Albany-Eugene Uprate.aux BPA-AllSwitchedShuntCorrections.aux BPA-BandonRogueReconductor.aux BPA-BreakersRevised.aux BPA-BridgeCaps.aux BPA-C_OR_fixes_2016hw.aux BPA-CreateMonroe500kV3rdCap.aux BPA-Elwha_Dam_Removal.aux BPA-Fairmount-HappyV-RatingChange.aux BPA-Garrison230kVLineRatings.aux BPA-Harney9MVarReactorAddition.aux BPA-IdahoFixes_16hw.aux BPA-Intalco_ShuntLoadCorrections.aux BPA-JonesCyn_EOmak-Caps.aux BPA-Longview-LexingtonRetermination.aux BPA-PilotButte20MvarCap.aux BPA-PonderosaPrinville115kVCorrectRating.aux BPA-RemoveAllstonCaps.aux BPA-Snohomish230-115TX.aux BPA-TransformerLineRatings.aux BPA-TransformerTaps.aux BPA-TriCitiesFixes_16hw_v2.aux BPA-Mpv-Klh_70C.aux BPA-Mon-Snoh1-2_100C.aux BPA-Mon-Nov_70C.aux MapleValleyKeelerSVCs.aux MerwinViewTapRating.aux PDCI_Setup.aux PSEBakerSedroUpgrades-16HW2.aux PSESedroBellingham4Upgrade-16HW2.aux 22

SCL-Corrections-16HW2-08Jun11.aux SNPD-Corrections-16HW2-08Jun11.aux TPWR-16HW2.aux User customized voltage tuning For Base Case Modifications to Mitigate PSAST overloads: 1b_Del-Duw_Upgrade.aux 3_PortalWayTx2.aux 1c_BothSnokhightempwire_withrxb.aux PSAST_Raver500-230_TX_option.aux PSAST_SCL- 1d_LakeSide_One230kVLine.aux CableSeriesReactors6ohm.aux For Base Case Modifications to Mitigate PGE and PACW overloads: Assume reconductor with no parameter changes unless supplied by PGE or PACW Attachment C Contingency List The full Northwest Category B and Category C contingency definition list used for ColumbiaGrid system assessments total over 4000. The full list is applied to flag thermal overloads. For voltage stability simulations, the list was reduced to 45 contingencies that resulted in new failed solutions when westside generation was displaced with wind. These contingency names are listed below. Category B: N-1: Raver Shunt Capacitors N-1: EchoLake Shunt Capacitors L_CHIEFJO500-MONROE500C1-MS N-1: 3TM Monroe-Echo LK-SnoK 500kV N-1: ALLSTON500-KEELER500C L_ECHOLAKE500-SCHULTZ500C1-MS 23

Category C: BF 5072 Echo Lk-Maple Vl & Echo Lk Caps N-2: Big Eddy-Ostrander 500kV/Big Eddy-Troutdale 230kV N-2: CUSTER-MONROE 1&2 BF 4510 PEARL-MARION & PEARL 500/230 BK1 BF 4598 Chief Joe-Monroe #1 500kV & Chief Joe PH 6&7 (21-27) BF 5111 MON-ECHO L-ECHO L CAPS N-2: Pearl-Ostrander 500kV/Pearl-Marion 500kV BF 4519 Cust-Mon #1 500kV & Mon Caps N-2: PAUL-SATSOP 500 & OLYMPIA-SATSOP 230 N-2: CHIEF JO-MONROE 500 & CHIEF JO- SNOHOMISH 345 #4 N-2: PAUL-STSOP, PAUL-OLY N-2: Ashe-Marion 500kV/Buckley-Marion 500kV N-2: MONROE-CUSTER #1 500 & CHIEF JO- SNOHOMISH #4 BF 4548 ALLSTON-PAUL-SATSOP BF 4672 MON-C JOE 500 W/MON MSC N-2: JOHN DAY-MARION & BUCKLEY-MARION 500 N-2: PAUL-ALLSTON 2 & ALLSTON-NAPAVINE 500 BF KEELER 230 - FAULT ANY LINE BF 5148 COULEE-SCHULTZ-ECHO LAKE BF 4272 INGLEDOW-CUSTER-MON BF 5078 SCHULTZ-ECHO LAKE-MAPLE VALLEY BF 4526 ECHO LAKE-MONROE-CUSTER BF 5157 COULEE-SCHULTZ-RAVER BF 4522 Monroe-Echo Lake-SnoKing 500kV & Monroe Cap Grp #3 N-2: CHIEF JO-MONROE & CHIEF JO-SICKLER 500 BF Chief Joe 1 & 2 Bus Section 230 N-2: MON-SK-EL 5 & MON-SAMM2 (198) BF 4394 ALLSTON-KEELER-PEARL BF OLYMPIA WEST BUS BF 4530 RAVER-PAUL-SATSOP BF 4322,24,94 Clear Keeler 500kV N-2: Schultz-Echo Lake 500kV / Schultz-Raver #4 500kV BF 5075 SCHULTZ-ECHO LAKE-ECHO LAKE CAPS BF 4540 NAPAVINE-PAUL-SATSOP BF 4554 OLYMPIA-PAUL-TONO BF 5121 MAPLE VALLEY-ECHO LAKE-SCHULTZ-ECHO LAKE CAPS N-2: CHIEF JO-MONROE 500 & CHIEF JO- SNOHOMISH 345 #3 N-2: Schultz-EchoLake & Schultz-Raver 500kV N-2: COULEE-SCHULTZ 1&2 N-2: SCHULTZ-WAUTOMA 500 & VANTAGE-SCHULTZ 500 BF 4502 NAPAVINE-ALLSTON-KEELER 24

Attachment D Westside Generation Displacement Order Number Name ID 42042 WHITHRN2 2 42721 FREDERC1 1 42722 FREDERC2 2 42112 FREDONA2 2 42111 FREDONA1 1 42043 WHITHRN3 3 42115 FREDONA4 4 42114 FREDONA3 3 43019 BEAVER 1 43017 BEAVER 5 43017 BEAVER 6 43017 BEAVER 4 43017 BEAVER 3 43017 BEAVER 2 43017 BEAVER 1 42014 ENSERCHL L 42013 ENSERCH3 3 42012 ENSERCH2 2 42011 ENSERCH1 1 42133 MRPTGEN3 3 42022 SUMAS L L 42021 SUMAS 1 1 42032 TENASKA2 2 42033 TENASKAL L 42031 TENASKA1 1 47734 BHP 30 30 47735 BHP 40 40 47736 BHP 5060 50 47736 BHP 5060 60 47737 BHP 70 70 47598 GRYHB S1 1 47597 GRYHB G2 1 47596 GRYHB G1 1 47676 MNTFRM S 1 47675 MNTFRM G 1 47216 RVR RD C 1 47576 FREDST S 1 47577 FREDST G 1 43907 PORTW S1 1 43905 PORTW G1 1 47590 CHEH S1 1 47588 CHEH G1 1 47589 CHEH G2 2 47740 CENTR G1 1 46439 ROSS 42 1 46439 ROSS 42 2 46441 ROSS 44 3 46441 ROSS 44 4 42124 UP BAKER 1 42124 UP BAKER 2 42121 LO BAKER 1 46429 GORGE 3 46429 GORGE 2 46429 GORGE 1 46430 GORGE 4 4 46419 DIABLO31 1 46420 DIABLO32 2 45689 JACKSN2 1 45687 JACKSN1 1 25

APPENDIX B Investigation of Post Contingency Shunt Capacitor Switching Assumption The Powerworld Simulator version 16 includes a feature that converts the WECC Composite Load Model dynamic data (CMPLDW) into the powerflow model with static ZIP loads. The CMPLDW data submitted to WECC as of January 12, 2012 was used to modify the 16HW2 loads used in this study. Each load object in the power flow model was modified to represent a distribution system equivalent (step-down transformer, voltage regulator, and distribution feeder) and six load objects representing the four motor types, one electronic, and one static load. The motor and electronic loads were modeled as static constant MVA in the powerflow. The westside load area 500kV shunt capacitor cut-in and cut-out settings (time and voltage) and the distribution voltage regulator setpoint, deadband, and tap changing time delays were modeled in the PowerWorld Simulator version 16 Time Step Simulation (TSS) tool. The project option selected for the investigation was the Schultz series capacitor addition with the Schultz-Raver #4 reconductor. The powerflow basecase selected was the voltage stability limited basecase from the PV application for the most severe WOCN contingency (Schultz-EchoLake & Schultz- Raver1). The voltage stability limit was confirmed with the QV Tool and resulted in near zero Mvar margin with a high (530kV) critical voltage. Maximum allowable westside thermal generation was off and the only thermal generation on in the case was Centralia (114 MW) and March Point (101 MW). The SCL Skagit generation totaled 220 MW and Jackson hydro was 40 MW. Shunt capacitors switched in post contingency at Echo Lake (316 Mvars) and Monroe (633 Mvars) in addition to SVC response at Maple Valley and Keeler. The Time Step Simulation tool failed to solve after opening the second line. The generation at Centralia, which started at 114 MW, was increased until a successful solution was achieved after opening the second line. That generation level (found by trial and error) was 530 MW. The subsequent simulation appeared to show the shunt capacitors switching in time to avoid load loss and the voltage stabilized after a few minutes. To ensure the simulation was at, or slightly beyond, the voltage stability limit, the 530 MW of Centralia generation was opened at 5 minutes. The simulation showed the voltage stabilized after another several minutes. To examine potential overvoltage effects after system restoration, the opened lines were closed near the end of the simulation. The load voltages did not appear too high. Conclusion: reliable post contingency capacitor switching should be feasible to be planned. However, the simulation needs to be repeated as new information becomes available. Errors were found in the submitted load climate zone data for the approved 12HS4 case. Also, in the coming months and years, new information should become available on the load mix and composition at the modeled load objects. The composite load model for power flow purposes will also be improved. 26

Q_sync (Mvar) The following annotated plots from the simulation provide additional detail. QV Curves 350 Raver is critical bus 300 250 200 Critical voltage is 530kV & reactive margin is zero 150 100 50 0 1.02 1.03 1.04 1.05 1.06 Voltage (pu) 1.07 1.08 1.09 1.1 gfedcb gfedcb gfedcb gfedcb gfedcb **BUS** 40045 (ALLSTON_500.0),**CASE** BASECASE **BUS** 40323 (CUSTER W_500.0),**CASE** BASECASE **BUS** 40749 (MONROE_500.0),**CASE** BASECASE **BUS** 40821 (PAUL_500.0),**CASE** BASECASE **BUS** 40869 (RAVER_500.0),**CASE** BASECASE Build Date: February 7, 2012 The Northwest load objects above 1 MW (over 1900 loads) were converted to the WECC composite load model (CMPLDW) for powerflow. The model included a distribution system equivalent, static constant MVA loads for motor loads, static constant MVA for electronic loads, and constant current constant impedance components specified in the dynamic data file. The dynamic data was derived from the default data set for climate zone and load composition specified in the PSLF 12HS4 case load Long_ID data. The example of the conversion for one load at Poulsbo 115kV is shown below. 27

POULSBO Bus: POULSBO (42961) Nom kv: 115.00 Area: NORTHWEST (40) Zone: KITSAP (434) 21.614 MW 2.226 Mvar ID 1 Original basecase load 1.0037 pu 115.42 KV -70.22 Deg 0.00 $/MWh C 53.8 MW 10.1 Mvar 54.8 MVA 75.4 MW -12.3 Mvar 76.4 MVA Amps CKT 1 FOSS CNR 42904 1.0001 pu 115.01 KV C Amps BANGOR 40077 CHRISCNR 42935 KNGSTNTP 42945 SERWOLD 42970 CKT 1 VALLYJCT 42907 1.0141 pu 116.62 KV BUCKLIN 42932 C KITSAP POULSBO Bus: POULSBO (42961) Nom kv: 115.00 Area: NORTHW EST (40) Zone: KITSAP (434) 1.0134 pu 116.54 KV -68.18 Deg 0.00 $/MW h D 55.7 MW 9.9 Mvar 56.6 MVA CMPLDW load Step down transformer equiv w/ltc 77.9 MW -12.0 Mvar 78.8 MVA 22.2 MW 2.0 Mvar 22.3 MVA 0.00 MW 0.00 Mvar Amps D CKT 1 FOSS CNR 42904 1.0099 pu 116.13 KV Amps 0.9750 tap BANGOR 40077 CHRISCNR 42935 KNGSTNTP 42945 SERWOLD 42970 CKT 1 VALLYJCT 42907 1.0238 pu 117.74 KV BUCKLIN 42932 C KITSAP 42933 TRACYTON 42981 842961 942961 942961 0.9998 pu 114.98 KV CKT 1 842961 System State 28

942961 Bus: 942961 (942961) Nom kv: 115.00 Area: NORTHWEST (40) Zone: KITSAP (434) 0.9998 pu 114.98 KV -74.56 Deg 0.00 $/MWh CMPLDW loads: static ZIP 2.116 MW 0.000 Mvar 1.006 MW 0.752 Mvar 1.969 MW 1.601 Mvar ID C1 ID C2 ID C3 ID C4 ID C5 ID C6 21.5 MW -2.8 Mvar 21.7 MVA Line Shunts 0.377 MW 0.307 Mvar 2.409 MW 0.489 Mvar 13.650 MW -0.305 Mvar Distribution feeder equivalent 842961 CKT 1 842961 POULSBO 42961 1.0134 pu 116.54 KV FOSS CNR After load conversion, the powerflow would not solve the contingency with the post contingency generator voltage control using the Line Drop Compensation and Reactive Current Compensation (LDC_RCC) settings prior to 500kV capacitor switching. However, the contingency solved if the LDC_RCC was not applied and the generator high side voltage control was retained. The condition that would solve with LDC_RCC was determined by trial and error. The Centralia generation in the critical basecase was 114 MW. The Centralia generation was incrementally increased, displacing wind generation, until the powerflow solved. With LDC_RCC set to control the generator terminal (LDC_RCC=0) for all generators, the contingency solved with 440 MW at Centralia. With LDC_RCC set to the data in the WECC Master Dynamics File, the contingency solved with 530 MW at Centralia. The basecase with the composite load model was then modified to 530 MW at Centralia for subsequent Time Step Simulation application. 29

Transient stability in PSLF v18, open Schultz-Echo Lake and Schultz-Raver #1 A transient stability simulation was performed with the dynamic CMPLDW data at a slightly higher stress level. The Centralia generation was opened. The lines were switched open and the initial seconds monitored for voltage dip levels. The Raver 500kV and Poulsbo 115kV bus voltage levels are shown below. These levels are higher than what is later shown in the initial seconds of the powerflow Time Step Simulation. 30

42961 (POULSBO) TO 842961 (842961) CKT 1 XF Tap Pos. Values Time Step Simulation in PowerWorld Simulator v16 1.09 1.08 1.07 Outage, SVC response, EchoLake cap Time versus bus per unit voltage over 20 min Bus Timepoint Custom Results Variables 1.06 1.05 1.04 Test w/trip of 500MW at Centralia Outage lines restored 1.03 1.02 1.01 1 0.99 0.98 0.97 0.96 0.95 0.94 0.93 0.92 0.91 0.9 Monroe cap Voltage stability limit exceeded, Keeler cap switched restored to voltage stable condition 0:00:00 0:01:00 0:02:00 0:03:00 0:04:00 0:05:00 0:06:00 0:07:00 0:08:00 0:09:00 DateTime 0:10:00 0:11:00 0:12:00 0:13:00 0:14:00 0:15:00 40869 (RAVER) PU Volt 42961 (POULSBO) PU Volt 942961 (942961) PU Volt 0:16:00 0:17:00 0:18:00 0:19:00 0:20:00-4 -5-6 -7-8 -9-10 -11-12 -13-14 -15-16 -17-18 -19-20 0:00:00 42961 (POULSBO) TO 842961 (842961) CKT 1 XF Tap Pos. Time versus Poulsbo step down transformer LTC tap position over 20 min 0:05:00 DateTime 0:10:00 0:15:00 42961 (POULSBO) TO 842961 (842961) CKT 1 XF Tap Pos. 0:20:00 31

Values Sw itched Shunt Tim epoint Custom Results Variables 600 550 500 450 400 350 300 250 200 150 100 50 0 500kV shunt capacitors switched and SVC response over 20 min 0:00:00 0:05:00 DateTime 0:10:00 0:15:00 0:20:00 40381 (ECHOLAKE) #s Nominal Mvar 40595 (KEEL-SVC) #1 Nominal Mvar 40601 (KEELER) #s Nominal Mvar 40749 (MONROE) #s Nominal Mvar 40771 (MV-SVC) #1 Nominal Mvar 32

System Load MW Values Injection Group Timepoint Custom Results Variables 2,000 1,900 1,800 1,700 1,600 1,500 1,400 1,300 1,200 1,100 Generator reactive aggregated at the Upper Columbia (Chief Joe and Coulee) and Lower Columbia (Bonneville, The Dalles, John Day, and McNary) over 20 min 1,000 900 800 0:00:00 0:05:00 DateTime 0:10:00 0:15:00 0:20:00 AGGREGATE-Low er Columbia Total Mvar Injection AGGREGATE-Upper Columbia gen Total Mvar Injection 150,600 150,400 150,200 150,000 149,800 149,600 149,400 149,200 149,000 148,800 System Load MW WECC wide system load over 20 minutes (excludes distribution losses associated with composite load model) 0:00:00 0:05:00 DateTime 0:10:00 0:15:00 0:20:00 System Load MW 33

Values System Loss MW 6,450 6,400 6,350 6,300 6,250 6,200 6,150 6,100 0:00:00 0:05:00 System Loss MW WECC wide system losses over 20 minutes (includes distribution losses associated with composite load model) DateTime 0:10:00 System Loss MW 0:15:00 0:20:00 The following plots show expanded views of the above 20 minute plots with added analytical details Time zero to 1 minute with one load example (Poulsbo) out of 1900 loads converted to composite load powerflow model 1:Raver-Schultz opens Time vs per unit voltage: Raver 500kV, Poulsbo 115kV, Poulsbo load voltage Bus Timepoint Custom Results Variables 1.08 1.07 2:EchoLake-Schultz opens 1.06 1.05 1.04 1.03 1.02 1.01 1 0.99 4:EchoLake shunt cap inserts 3:SVC response 6:Dist voltage restored, taps stop changing, timer resets rresresponds 1:Dist voltage regulator timer starts 0.98 0.97 0.96 0.95 0.94 5:Dist voltage reg taps start changing 0.93 0.92 0.91 0:00:00 0:00:05 0:00:10 0:00:15 0:00:20 0:00:25 0:00:30 DateTime 0:00:35 0:00:40 0:00:45 0:00:50 0:00:55 0:01:00 40869 (RAVER) PU Volt 42961 (POULSBO) PU Volt 942961 (942961) PU Volt 34

Values 42961 (POULSBO) TO 842961 (842961) CKT 1 X -4-6 -8-10 -12-14 42961 (POULSBO) TO 842961 (842961) CKT 1 XF Tap Pos. 0:00:00 0:00:05 0:00:10 0:00:15 0:00:20 0:00:25 0:00:30 DateTime 0:00:35 0:00:40 0:00:45 0:00:50 0:00:55 0:01:00 42961 (POULSBO) TO 842961 (842961) CKT 1 XF Tap Pos. Time 1 minute to 2 minutes Bus Timepoint Custom Results Variables 1.04 1.035 1.03 1.025 1.02 1.015 1.01 1.005 1 0.995 0.99 0.985 0.98 0.975 0.97 0.965 0.96 0.955 0.95 0.945 0.94 0.935 0.93 0.925 0.92 0.915 0.91 0.905 0:01:00 0:01:05 7:Dist voltage regulator timer started 5 sec ago 0:01:10 0:01:15 0:01:20 0:01:25 DateTime 0:01:30 0:01:35 9:Monroe shunt cap inserts 9:Dist voltage reg above deadband, timer starts 8:Dist voltage reg taps start 0:01:40 0:01:45 40869 (RAVER) PU Volt 42961 (POULSBO) PU Volt 942961 (942961) PU Volt 0:01:50 0:01:55 0:02:00 35

42961 (POULSBO) TO 842961 (842961) CKT 1 X Values 42961 (POULSBO) TO 842961 (842961) CKT 1 X -15 42961 (POULSBO) TO 842961 (842961) CKT 1 XF Tap Pos. -16-17 -18-19 0:01:00 0:01:05 0:01:10 0:01:15 0:01:20 0:01:25DateTime 0:01:30 0:01:35 0:01:40 0:01:45 42961 (POULSBO) TO 842961 (842961) CKT 1 XF Tap Pos. 0:01:50 0:01:55 0:02:00 Time 2 minutes to 3 minutes 1.06 1.055 1.05 1.045 1.04 1.035 1.03 1.025 1.02 1.015 1.01 1.005 1 0.995 0.99 0.985 0.98 0.975 0.97 0.965 0.96 0.955 0.95 0.945 0.94 0.935 0.93 0.925 Bus Timepoint Custom Results Variables 10:Dist voltage reg taps start reverse tap changing 12:Above deadband again, timer starts 11:Dist voltage within deadband, taps stop, timer resets 13:Tap changes bring voltage back to deadband 0:02:00 0:02:05 0:02:10 0:02:15 0:02:20 0:02:25 0:02:30 DateTime 0:02:35 0:02:40 0:02:45 0:02:50 0:02:55 0:03:00 40869 (RAVER) PU Volt 42961 (POULSBO) PU Volt 942961 (942961) PU Volt -10 42961 (POULSBO) TO 842961 (842961) CKT 1 XF Tap Pos. -12-14 -16-18 0:02:00 0:02:05 0:02:10 0:02:15 0:02:20 0:02:25 0:02:30 DateTime 0:02:35 0:02:40 0:02:45 0:02:50 0:02:55 0:03:00 42961 (POULSBO) TO 842961 (842961) CKT 1 XF Tap Pos. 36

OULSBO) TO 842961 (842961 Values Time 3 minutes to 5 minutes Bus Timepoint Custom Results Variables 1.06 1.055 1.05 1.045 1.04 1.035 1.03 1.025 1.02 1.015 1.01 1.005 1 0.995 0.99 0.985 0.98 0.975 0.97 0.965 0.96 0.955 0.95 0.945 0.94 14:Above deadband again, timer starts 16:Test with open 500MW Centralia gen 15:Tap change brings voltage back to deadband 0:03:00 0:03:10 0:03:20 0:03:30 0:03:40 0:03:50 0:04:00 DateTime 0:04:10 0:04:20 0:04:30 0:04:40 0:04:50 0:05:00 40869 (RAVER) PU Volt 42961 (POULSBO) PU Volt 942961 (942961) PU Volt -10 42961 (POULSBO) TO 842961 (842961) CKT 1 XF Tap Pos. -11 0:03:00 0:03:10 0:03:20 0:03:30 0:03:40 0:03:50 DateTime 0:04:00 0:04:10 0:04:20 0:04:30 42961 (POULSBO) TO 842961 (842961) CKT 1 XF Tap Pos. 0:04:40 0:04:50 0:05:00 Time 5 minutes to 8 minutes 37

42961 (POULSBO) TO 842961 (842961) CKT 1 XF T Values Bus Timepoint Custom Results Variables 1.06 1.05 1.04 16:Test with opening 500MW Centralia gen gen 18:Keeler shunt cap inserts 1.03 1.02 1.01 1 16:Below deadband, timer starts 19:within deadband,timer resets 21:taps start changing 0.99 0.98 0.97 0.96 0.95 17:taps start changing 20:Below deadband, timer starts Keeler cap restores voltage stability, tap changes raise load voltage 0.94 0.93 0.92 0.91 Beyond voltage stability limit, tap changes cause further decrease to load voltage 0.9 0:05:00 0:05:10 0:05:20 0:05:30 0:05:40 0:05:50 0:06:00 0:06:10 0:06:20 DateTime 0:06:30 0:06:40 0:06:50 40869 (RAVER) PU Volt 42961 (POULSBO) PU Volt 942961 (942961) PU Volt 0:07:00 0:07:10 0:07:20 0:07:30 0:07:40 0:07:50 0:08:00-10 -11-12 -13-14 -15-16 -17-18 -19-20 0:05:00 0:05:15 0:05:30 42961 (POULSBO) TO 842961 (842961) CKT 1 XF Tap Pos. 0:05:45 0:06:00 0:06:15 DateTime 0:06:30 0:06:45 0:07:00 0:07:15 0:07:30 42961 (POULSBO) TO 842961 (842961) CKT 1 XF Tap Pos. 0:07:45 0:08:00 Time 18 minutes to 20 minutes 38

42961 (POULSBO) TO 842961 (842961) CKT 1 XF Tap Pos. Values 24:close EchoLake-Schultz 25:SVC response Bus Timepoint Custom Results Variables 1.08 1.07 1.06 1.05 1.04 1.03 22:close Raver-Schultz#1 26:taps start changing 29:Tap change brings voltage back to deadband 1.02 1.01 1 23:above dead band, timer starts 0.99 0.98 0.97 0.96 0.95 0.94 27:Tap changes bring voltage back to deadband 28:Above deadband, timer starts again 0.93 0.92 0.91 0:17:50 0:18:00 0:18:10 0:18:20 0:18:30 0:18:40 0:18:50 DateTime 0:19:00 0:19:10 0:19:20 0:19:30 0:19:40 0:19:50 0:20:00 40869 (RAVER) PU Volt 42961 (POULSBO) PU Volt 942961 (942961) PU Volt -4-5 -6-7 -8-9 -10-11 -12-13 -14-15 -16-17 -18-19 42961 (POULSBO) TO 842961 (842961) CKT 1 XF Tap Pos. 0:18:00 0:18:10 0:18:20 0:18:30 0:18:40 0:18:50 DateTime 0:19:00 0:19:10 0:19:20 0:19:30 0:19:40 0:19:50 0:20:00 42961 (POULSBO) TO 842961 (842961) CKT 1 XF Tap Pos. 39

APPENDIX C- Generation Displacement Method PV Contingency Results and Contingency Analysis with Critical WOCN PV Basecase This appendix provides examples of available supporting data. The tables and graphs are copied from PowerWorld Simulator output. If a reviewer wants to examine the full range of calculated output from the simulations (e.g. interface and branch flows, bus voltages, bus self dv/dq, real and reactive losses, generator reactive output), the quantities are archived in PowerWorld Simulator version 16 pwb and aux files. C1: Base Condition Prior to WOCN Projects and Prior to PSAST Projects (16HW2 load level) The table below shows the PV simulation summary on the modified WECC 2015-16 Heavy Winter case (16HW2) prior to WOCN project additions. The column PV Scenario is the contingency label. The contingency labels are listed in ascending order of the PV calculated voltage stability limit. Remote wind generation is increased sequentially displacing the west side thermal generation in the order listed in the Study Plan. This incremental generation shift gradually increases the WOCN interface flow until the individual contingencies fail to solve beyond the Max Shift level. Reached Nose in the Critical Reason column indicates the last successful powerflow solution is at the Max Shift column (the amount of remote wind generation increased at that final condition). 40

PV Scenarios Overview PV Scenario Critical Reason Max Shift N-2: PAUL-STSOP, PAUL-OLY Reached Nose 262.5 N-2: Schultz-EchoLake & Schultz-Raver 500kV Reached Nose 1356.25 BF 4554 OLYMPIA-PAUL-TONO Reached Nose 2059.38 BF 5121 MAPLE VALLEY-ECHO LAKE-SCHULTZ-ECHO LAKE CAPS Reached Nose 2440.63 N-2: CHIEF JO-MONROE 500 & CHIEF JO-SNOHOMISH 345 #4 Reached Nose 2445.75 N-2: CHIEF JO-MONROE 500 & CHIEF JO-SNOHOMISH 345 #3 Reached Nose 2446.88 N-2: Pearl-Ostrander 500kV/Pearl-Marion 500kV Reached Nose 2503.13 N-2: Schultz-Echo Lake 500kV / Schultz-Raver #4 500kV Reached Nose 2533.25 BF 4598 Chief Joe-Monroe #1 500kV & Chief Joe PH 6&7 (21-27) Reached Nose 2681.25 BF 5157 COULEE-SCHULTZ-RAVER Reached Nose 2731.25 BF 5148 COULEE-SCHULTZ-ECHO LAKE Reached Nose 2740.63 N-2: COULEE-SCHULTZ 1&2 Reached Nose 2746.88 BF 4672 MON-C JOE 500 W/MON MSC Reached Nose 2786.38 BF 5078 SCHULTZ-ECHO LAKE-MAPLE VALLEY Reached Nose 2846.88 BF 4526 ECHO LAKE-MONROE-CUSTER Reached Nose 2936.38 BF 4530 RAVER-PAUL-SATSOP Reached Nose 2989.5 BF 4519 Cust-Mon #1 500kV & Mon Caps Reached Nose 3008.25 N-2: CHIEF JO-MONROE & CHIEF JO-SICKLER 500 Reached Nose 3068.75 BF Chief Joe 1 & 2 Bus Section 230 Reached Nose 3068.75 BF 5111 MON-ECHO L-ECHO L CAPS Reached Nose 3150 BF 5075 SCHULTZ-ECHO LAKE-ECHO LAKE CAPS Reached Nose 3250.88 BF 4522 Monroe-Echo Lake-SnoKing 500kV & Monroe Cap Grp #3 Reached Nose 3286.38 L_CHIEFJO500-MONROE500C1-MS Reached Nose 3318.75 N-2: CUSTER-MONROE 1&2 Reached Nose 3343.75 L_ECHOLAKE500-SCHULTZ500C1-MS Reached Nose 3365.63 BF 4510 PEARL-MARION & PEARL 500/230 BK1 Reached Nose 3598.88 BF OLYMPIA WEST BUS Reached Nose 3725 N-2: PAUL-SATSOP 500 & OLYMPIA-SATSOP 230 Reached Nose 3725 N-2: PAUL-ALLSTON 2 & ALLSTON-NAPAVINE 500 Reached Nose 3725 BF 4548 ALLSTON-PAUL-SATSOP Reached Nose 3970.75 BF 4502 NAPAVINE-ALLSTON-KEELER Reached Nose 4045.75 Contingencies from the PV summary table can be further investigated by accessing archived simulation quantities. For this WOCN voltage stability limit study, further investigation is needed because the very high critical voltage can cause the maximum gen shift to be overstated. This can occur because the powerflow solution during the PV simulation, instead of failing at the valid maximum shift, can successfully solve beyond the valid maximum shift with low voltage solutions. The Raver 500kV bus self dv/dq and per unit voltage (two of many monitored quantities) are plotted below. This is one method to initially screen for potential low voltage solutions. If the bus self dv/dq becomes negative and the solved voltage drops in the final solution points of the simulation, this could indicate an overstated maximum shift. QV simulations are needed to confirm the low voltage solution (all switched shunt and transformer LTC s changed to fixed). The plots below show the Max Shift is slightly overstated. 41

PU Volt dv/dq 0.0002 0-0.0002-0.0004-0.0006-0.0008-0.001 Low Voltage Solution with Schultz-Raver&Schultz- EchoLake slightly overstates maximum shift -0.0012 0 200 400 600 800 Nominal Shift 1,000 1,200 Build Date: February 15, 2012 1.085 1.08 1.075 1.07 1.065 Low Voltage Solution overstates maximum shift. At this shift magnitude, the correct voltage solution exceeds 1.1 per unit, but it solved at 1.064 per unit 0 100 200 300 400 500 600 700 800 Nominal Shift 900 1,000 1,100 1,200 1,300 N-2: Schultz-EchoLake & Schultz-Raver 500kV: RAVER_500.0 (40869) Build Date: February 15, 2012 At the valid maximum generation shift (voltage stability limit), a full contingency analysis is performed and compared to the contingency analysis prior to the generation shift. The objective is to determine branch overloads that occur prior to the voltage stability limit. Mitigation needs to be modeled and the PV simulation potentially re-run with the changed system model. In the table below, the branch loading after the generation shift is shown in column Max % Loading Cont.. The branch loading before the generation shift is shown in column Max % Ld Cont Compare. The difference between these two quantities is shown in column Worst Increased. The higher the number, the higher the impact caused by the generation shift to increase WOCN interface flow. The listed branch overloads need to be mitigated and the PV simulation performed again. 42

Contingency Analysis Results on critical Schultz-Raver&Schultz-EchoLake basecase Line Records Owner Name 1 From NumbFrom Name From NomTo NumbeTo Name To Nom kvcircuit Max % Loading Cont. Max % Ld Cont Comp Worst Increased ViolationsNew ViolaMax % Ld Cont Name Portland General Elec 43348 MURRAY H 230 43541 ST MARYS 230 1 107.21 85.06 24.55 4 1 N-1: KEELER500-PEARL500C1 PacifiCorp - West 45299 TROUTDAL 115 45303 TROUTPP2 230 1 103.92 94.15 11.64 3 0 N-2: Pearl-Ostrander 500kV/Pearl-Marion 500kV Puget Sound Energy 42448 SCENIC 115 42449 SKYKOMSH 115 1 105.44 97.77 7.67 3 0 N-2: Schultz-EchoLake & Schultz-Raver 500kV Puget Sound Energy 45608 BEVERLY 115 42399 GLDBRTIE 115 1 101.74 94.09 7.65 1 0 N-2: Schultz-EchoLake & Schultz-Raver 500kV Puget Sound Energy 42399 GLDBRTIE 115 42449 SKYKOMSH 115 1 102.51 94.91 7.6 1 0 N-2: Schultz-EchoLake & Schultz-Raver 500kV PacifiCorp - West 40049 ALVEY 230 46291 MCKEN TP 230 1 114.48 109.82 4.66 1 0 BF SANTIAM 230 (BUS SECTIONALZING BREAKER) Seattle City Light 46409 BROAD ST 115 46451 UNION T 115 1 122.63 120.97 4.46 12 0 BF 4522 Monroe-Echo Lake-SnoKing 500kV & Monroe Cap Grp #3 Seattle City Light 46433 MASS 115 46451 UNION T 115 1 122.58 120.97 4.32 12 0 BF 4522 Monroe-Echo Lake-SnoKing 500kV & Monroe Cap Grp #3 Seattle City Light 46435 MASS 230 46438 BK18MTP 110 2 101.92 101.43 2.54 7 0 N-1: 3TM Monroe-Echo LK-SnoK 500kV Seattle City Light 46435 MASS 230 46440 BK19MTP 110 3 101.53 101.04 2.53 7 0 N-1: 3TM Monroe-Echo LK-SnoK 500kV Bonneville Power Admin 40306 COVINGT5 500 40304 COVINGTE 230 5 123.4 121.35 2.05 1 0 N-2: RAVER-COVINGTON & RAVER-TACOMA 500 PacifiCorp - West 43215 GRESHAM 230 45303 TROUTPP2 230 1 107.82 106.66 1.81 4 0 BF BSBF MCLOUGHLIN TIE V668 230KV Seattle City Light 46403 BOTHELL 230 90150 BOTSNO21 230 2 106.74 105.96 1.67 5 0 L_BOTHELL230-SNOKS1230C1-MS Portland General Elec 40824 PEARL E 230 43773 PEARL # 230 1 107.07 105.66 1.41 1 0 L_PEARLW230-SHERWOOD230C1-MS New failed solution N-2: PAUL-STSOP, PAUL-OLY 43

C2: Base Condition after PSAST Project Suite (13.2% westside scalable load increase over 16HW2 loads) The overloads reported in C1 for the Puget Sound area are mitigated by adding the PSAST suite of projects to the model. The Portland area overloads (e.g. Murrayhill-St Marys) are mitigated by assuming reconductor with no topology change. The Portland area parameters are not changed, so the progression of the loading will be tracked as WOCN projects are added and generation shift increased. The westside load is increased because insufficient generation exists on the westside to displace with remote wind generation for later studies with projects that increase WOCN voltage stability limits. The table below shows the PV simulation results for this revised model. The first two most limiting contingencies are outside of the WOCN interface. Further investigation of BF 4272 INGLEDOW-CUSTER-MON shows the voltage stability limit is reached and the critical bus is in the Bellingham 230kV vicinity with a low critical voltage. Post contingency reactive switching is assumed as mitigation. Further investigation of N-2: PAUL-STSOP,PAUL-OLY shows the voltage stability limit is reached and the critical bus in in the vicinity of Olympia 230kV. Mitigation is assumed as a combination of post contingency reactive switching and controlled load tripping. The third contingency N-2: Shultz-EchoLake & Schultz-Raver 500kV is the WOCN contingency that establishes the limit for condition before WOCN projects. This becomes the reference case to compare subsequent WOCN project limits. PV Scenarios Overview PV Scenario Max Shift BF 4272 INGLEDOW-CUSTER-MON 939.5 N-2: PAUL-STSOP, PAUL-OLY 1050 N-2: Schultz-EchoLake & Schultz-Raver 500kV 1961.38 BF 4519 Cust-Mon #1 500kV & Mon Caps 2750 N-2: CHIEF JO-MONROE 500 & CHIEF JO-SNOHOMISH 345 #4 2884.38 N-2: CHIEF JO-MONROE 500 & CHIEF JO-SNOHOMISH 345 #3 2884.38 N-2: CUSTER-MONROE 1&2 2958.25 N-2: Pearl-Ostrander 500kV/Pearl-Marion 500kV 3103.13 BF 4530 RAVER-PAUL-SATSOP 3328.13 N-2: MONROE-CUSTER #1 500 & CHIEF JO-SNOHOMISH #4 3340.63 N-2: CHIEF JO-MONROE & CHIEF JO-SICKLER 500 3358.25 BF 4554 OLYMPIA-PAUL-TONO 3453.13 L_FAIRMONT230-SHELTON230C4 3562.5 L_FAIRMONT230-SHELTON230C3 3562.5 N-2: PAUL-SATSOP 500 & OLYMPIA-SATSOP 230 3562.5 BF A1610 FAIRMONT 230 3562.5 BF 4526 ECHO LAKE-MONROE-CUSTER 3584.38 N-2: Schultz-Echo Lake 500kV / Schultz-Raver #4 500kV 3603.13 BF 4522 Monroe-Echo Lake-SnoKing 500kV & Monroe Cap Grp #3 3668.75 BF OLYMPIA 230 E BUS - FLT ANY LN 3677 BF 5075 SCHULTZ-ECHO LAKE-ECHO LAKE CAPS 3775 BF 4672 MON-C JOE 500 W/MON MSC 3781.25 N-2: PAUL-ALLSTON 2 & ALLSTON-NAPAVINE 500 3833.25 BF 5111 MON-ECHO L-ECHO L CAPS 4227 BF OLYMPIA WEST BUS 4256 BF 4548 ALLSTON-PAUL-SATSOP 4369.81 Contingency analysis is performed before the generation shift and with the generation shift at the WOCN voltage stability limit. The results are compared and the overloads affected by the generation shift are identified. All overloads at this generation shift level occur in the Portland area and the assumed mitigation is reconductor with no topology or parameter change (so the PV simulation is not repeated). 44

Reference Condition at WOCN Voltage Stability Limit Owner Name 1 From NumberFrom Name From NomTo NumbeTo Name To Nom kvcircuit Max % Loading Cont. Max % Ld Cont Comp Worst Increased ViViolationsNew ViolaMax % Ld Cont Name Portland General Elec 43348 MURRAY H 230 43541 ST MARYS 230 1 122.85 85.23 40.14 5 1 N-1: KEELER500-PEARL500C1 PacifiCorp - West 45299 TROUTDAL 115 45303 TROUTPP2 230 1 115.51 95.08 22.96 10 0 N-2: Pearl-Ostrander 500kV/Pearl-Marion 500kV Bonneville Power Admin 40013 ACTON 115 40187 CASCD LK 115 1 109.84 87.89 21.95 9 0 N-2: Pearl-Ostrander 500kV/Pearl-Marion 500kV Bonneville Power Admin 40187 CASCD LK 115 40541 HOOD RVR 115 1 111.87 89.95 21.93 9 0 N-2: Pearl-Ostrander 500kV/Pearl-Marion 500kV PacifiCorp - West 45073 CULLY 115 45328 HOLLYW T 115 1 111.3 92.46 18.84 4 0 N-2: Pearl-Ostrander 500kV/Pearl-Marion 500kV PacifiCorp - West 45143 HLLYWOOD 115 45328 HOLLYW T 115 1 111.29 92.46 18.83 4 0 N-2: Pearl-Ostrander 500kV/Pearl-Marion 500kV Bonneville Power Admin 40217 CHENOWTH 115 47028 DISCV NW 115 1 112.7 103.12 9.58 3 0 L_CHEIGHTS115-SEUFERT115C1 Portland General Elec 40824 PEARL E 230 43773 PEARL # 230 1 112.37 105.08 7.3 1 0 L_PEARLW230-SHERWOOD230C1-MS PacifiCorp - West 43291 LINNEMAN 230 45301 TROUTPP1 230 1 132.46 126.14 6.33 4 0 L_TROUTDW230-TROUTPP2230C1 Bonneville Power Admin 40213 CHEMAWA 230 40211 CHEMAWA 115 1 104.92 99.86 5.06 3 0 T_SALEM230-SALEM115C1 PacifiCorp - West 43215 GRESHAM 230 45303 TROUTPP2 230 1 109.35 106.28 3.07 4 0 BF BSBF MCLOUGHLIN TIE V668 230KV note: violations only from N-2: Pearl-Ostrander 500kV/Pearl-Marion 500kV were removed from the list New failed solution N-2: PAUL-STSOP, PAUL-OLY 45

C3: Schultz Series Capacitors and Schultz-Raver #4 Reconductor The initial WOCN projects with short completion schedules are the addition of series capacitors on the Schultz-Raver #3&4 lines at Schultz and reconductor a section of the Raver-Schultz #4 line. PV Scenarios Overview PV Scenario Max Shift N-2: PAUL-STSOP, PAUL-OLY 1055.13 BF 4272 INGLEDOW-CUSTER-MON 1184.38 N-2: Schultz-EchoLake & Schultz-Raver 500kV 3193.75 N-2: Pearl-Ostrander 500kV/Pearl-Marion 500kV 3212.5 N-2: Schultz-Echo Lake 500kV / Schultz-Raver #4 500kV 3225 L_FAIRMONT230-SHELTON230C4 3343.75 L_FAIRMONT230-SHELTON230C3 3343.75 BF A1610 FAIRMONT 230 3343.75 BF 4530 RAVER-PAUL-SATSOP 3573.88 N-2: CHIEF JO-MONROE 500 & CHIEF JO-SNOHOMISH 345 #4 3609.38 N-2: CHIEF JO-MONROE 500 & CHIEF JO-SNOHOMISH 345 #3 3609.38 BF 4526 ECHO LAKE-MONROE-CUSTER 3750 N-2: OLY-SHLT #3&4 D. CKT 3900 BF 4540 NAPAVINE-PAUL-SATSOP 3911.38 BF OLYMPIA WEST BUS 3918.75 BF OLYMPIA 230 E BUS - FLT ANY LN 3937.5 N-2: PAUL-ALLSTON 2 & ALLSTON-NAPAVINE 500 3950 BF 4554 OLYMPIA-PAUL-TONO 3953.13 BF 4548 ALLSTON-PAUL-SATSOP 3953.13 N-2: PAUL-SATSOP 500 & OLYMPIA-SATSOP 230 3981.25 Contingency analysis is performed before and after the generation shift to the WOCN voltage stability limit and overloads caused by the generation shift are identified. Mitigation is assumed as reconductor with no topology changes. 46

Line Records Owner Name 1 From NumFrom Name From NomTo NumbeTo Name To Nom kvcircuit Max % Loading CMax % Ld Cont Comp Worst IncrViolationsNew ViolaMax % Ld Cont Name Portland General Elec 43348 MURRAY H 230 43541 ST MARYS 230 1 141.61 82.06 63.02 5 1 N-1: KEELER500-PEARL500C1 PacifiCorp - West 45299 TROUTDAL 115 45303 TROUTPP2 230 1 122.21 94.16 35.26 20 0 BF ROSS 230 EAST BUS, FAULT NORTH BONNEVILLE #2 PacifiCorp - West 45299 TROUTDAL 115 47214 RUNYAN 115 1 104.21 79.37 29.53 2 0 BF ROSS 230 EAST BUS, FAULT NORTH BONNEVILLE #2 Bonneville Power Admin 40013 ACTON 115 40187 CASCD LK 115 1 113.57 86.03 28.43 38 0 N-2: BIG EDDY-OSTRANDER 500 & BIG EDDY-TROUTDALE 230 Bonneville Power Admin 40187 CASCD LK 115 40541 HOOD RVR 115 1 115.56 88.07 28.4 48 0 N-2: BIG EDDY-OSTRANDER 500 & BIG EDDY-TROUTDALE 230 PacifiCorp - West 45143 HLLYWOOD 115 45328 HOLLYW T 115 1 107.17 89.34 26.47 5 0 BF 4283 KEELER-PEARL-OSTRANDER PacifiCorp - West 45073 CULLY 115 45328 HOLLYW T 115 1 107.18 89.33 26.47 5 0 BF 4283 KEELER-PEARL-OSTRANDER Portland General Elec 43761 GLENCOE# 115 43581 TABOR 115 1 104.92 83.24 21.68 3 0 BF 4283 KEELER-PEARL-OSTRANDER Portland General Elec 43313 MCLOUGLN 230 90126 BIGMCL13 230 1 102.61 86.63 15.98 2 0 N-2: BIG EDDY-OSTRANDER 500 & BIG EDDY-TROUTDALE 230 Northern Wasco PUD 47003 3MILE TP 115 47019 CHEIGHTS 115 1 101.12 85.67 15.45 2 0 N-1: The Dalles-Chenoweth Portland General Elec 43773 PEARL # 230 43527 SHERWOOD 230 1 101.86 87.36 14.49 1 0 L_PEARLW230-SHERWOOD230C1-MS Bonneville Power Admin 40217 CHENOWTH 115 47028 DISCV NW 115 1 116.48 102.66 13.83 5 0 L_CHEIGHTS115-SEUFERT115C1 PacifiCorp - West 43291 LINNEMAN 230 45301 TROUTPP1 230 1 136.98 125.44 11.53 5 0 L_TROUTDW230-TROUTPP2230C1 Portland General Elec 40824 PEARL E 230 43773 PEARL # 230 1 115.73 104.77 10.97 2 0 L_PEARLW230-SHERWOOD230C1-MS Bonneville Power Admin 40213 CHEMAWA 230 40211 CHEMAWA 115 1 108.1 99.42 8.68 3 0 T_SALEM230-SALEM115C1 Portland General Elec 43763 HEMLOCK# 115 43477 ROCKWD 2 115 1 99.49 91.69 7.8 1 0 N-2: GRESHAM-TROUTDALE-TROUTDALE,LINNEMAN-TROUTDALE Portland General Elec 43684 BLUELAKE 115 43685 BLUELAKE 230 1 105.27 97.9 7.36 1 0 N-2: GRESHAM-TROUTDALE-TROUTDALE,LINNEMAN-TROUTDALE Seattle City Light 90087 MAPSNO11 230 90088 MAPSNO12 230 1 101.67 98.21 5.37 6 0 BF 4526 ECHO LAKE-MONROE-CUSTER Seattle City Light 40689 MAPLE V3 230 90087 MAPSNO11 230 1 99.96 96.55 5.28 2 0 BF 4526 ECHO LAKE-MONROE-CUSTER Bonneville Power Admin 90088 MAPSNO12 230 41008 SNOK S3 230 1 99.85 96.51 5.28 2 0 BF 4526 ECHO LAKE-MONROE-CUSTER New Failed Solutions: BF 4272 INGLEDOW-CUSTER-MON N-2: PAUL-STSOP, PAUL-OLY 47

C4: Chief Jo-Monroe #2 option, 3 Echo Lake shunt capacitors The most robust WOCN project to increase the voltage stability limit appears to be a second Chief Jo- Monroe line with series compensation on both #1 and #2 lines. PV Scenarios Overview PV Scenario Max Shift N-2: PAUL-STSOP, PAUL-OLY 1071.88 N-2: Pearl-Ostrander 500kV/Pearl-Marion 500kV 3325 BF 4530 RAVER-PAUL-SATSOP 3586.38 L_FAIRMONT230-SHELTON230C3 3648.88 BF A1610 FAIRMONT 230 3648.88 L_FAIRMONT230-SHELTON230C4 3675 BF 4554 OLYMPIA-PAUL-TONO 3796.88 N-2: PAUL-SATSOP 500 & OLYMPIA-SATSOP 230 4205.13 N-2: Chief Jo-Monroe 1&2 4239.5 BF 4548 ALLSTON-PAUL-SATSOP 4284.38 BF OLYMPIA WEST BUS 4284.38 N-2: PAUL-ALLSTON 2 & ALLSTON-NAPAVINE 500 4284.38 BF 4272 INGLEDOW-CUSTER-MON 4483.25 BF 4540 NAPAVINE-PAUL-SATSOP 4631.25 BF 4322,24,94 Clear Keeler 500kV 4733.25 BF 4394 ALLSTON-KEELER-PEARL 4733.25 BF 4502 NAPAVINE-ALLSTON-KEELER 4756.25 N-1: Paul-Satsop & Satsop 500/230kV tx 4958.25 BF OLYMPIA 230 E BUS - FLT ANY LN 5008.25 BF KEELER 230 - FAULT ANY LINE 5056.25 N-2: OLY-SHLT #3&4 D. CKT 5062.5 N-2: CUSTER-MONROE 1&2 5254 N-2: Schultz-EchoLake & Schultz-Raver 500kV 5262.5 BF 4519 Cust-Mon #1 500kV & Mon Caps 5300 N-2: MONROE-CUSTER #1 500 & CHIEF JO-SNOHOMISH #4 5303.13 N-2: CHIEF JO-MONROE 500 & CHIEF JO-SNOHOMISH 345 #3 5318.75 N-2: CHIEF JO-MONROE 500 & CHIEF JO-SNOHOMISH 345 #4 5325 Reducing EchoLake shunt capacitors from three to two reduces max shift for Schultz-EchoLake&Schultz-Raver to 5193 MW. The limiting WOCN contingency from the PV summary is the N-2: Schultz-EchoLake&Schultz-Raver. Further investigation of archived monitored quantities shows the maximum shift is overstated due to low voltage solutions. The Raver bus self dv/dq and per unit voltage plots shown below demonstrate the screening method for low voltage solutions. With the robust WOCN project, the high generation shift causes main grid voltage stability limits for several contingencies outside of WOCN. The calculated WOCN maximum generation shift can t be achieved without major mitigation for contingencies for Raver-Paul, Paul-Allston&Napavine-Allston, and John Day-Marion&Buckley-Marion. 48

PU Volt dv/dq 0.00006 0.00004 0.00002 0-0.00002-0.00004-0.00006-0.00008-0.0001 Low Voltage Solution with Schultz-Raver&Schultz- EchoLake slightly overstates maximum shift 0 1,000 2,000 3,000 Nominal Shift 4,000 5,000 N-2: Schultz-EchoLake & Schultz-Raver 500kV: RAVER_500.0 (40869) Build Date: February 15, 2012 1.09 1.08 1.07 1.06 0 1,000 2,000 3,000 4,000 5,000 Nominal Shift Build Date: February 15, 2012 49

PU Volt dv/dq 0.00011 0.0001 0.00009 0.00008 0.00007 0.00006 0.00005 0.00004 0.00003 0.00002 0.00001 0-0.00001-0.00002-0.00003-0.00004-0.00005 JohnDay-Marion&BuckleyMarion also at limit, max shift overstated due to low voltage solution 0 500 1,000 1,500 2,000 2,500 3,000 Nominal Shift 3,500 4,000 4,500 5,000 5,500 N-2: JOHN DAY-MARION & BUCKLEY-MARION 500: ALLSTON_500.0 (40045) Build Date: February 15, 2012 1.09 1.08 1.07 1.06 1.05 1.04 1.03 1.02 1.01 1 0.99 0.98 0.97 0.96 0 500 1,000 1,500 2,000 2,500 3,000 Nominal Shift 3,500 4,000 4,500 5,000 5,500 N-2: JOHN DAY-MARION & BUCKLEY-MARION 500: ALLSTON_500.0 (40045) Build Date: February 15, 2012 50

PU Volt 1.09 1.085 1.08 1.075 1.07 1.065 1.06 1.055 1.05 Paul-Allston&Napavine-Allston reaches voltage stability limit prior to WOCN Schultz-Raver&Schultz-EchoLake 1.045 0 500 1,000 1,500 2,000 2,500 Nominal Shift 3,000 3,500 4,000 N-2: PAUL-ALLSTON 2 & ALLSTON-NAPAVINE 500: ALLSTON_500.0 (40045) N-2: PAUL-ALLSTON 2 & ALLSTON-NAPAVINE 500: PAUL_500.0 (40821) Build Date: February 15, 2012 51

Q_sync (Mvar) QV Curves 210 200 190 180 170 160 150 140 130 120 110 100 90 80 70 60 50 40 30 20 10 0-10 -20-30 -40-50 -60-70 -80-90 -100-110 Critical bus is at Paul 0.98 0.99 1 1.01 1.02 1.03 1.04 Voltage (pu) 1.05 1.06 1.07 1.08 1.09 1.1 gfedcb **BUS** 40045 (ALLSTON_500.0),**CASE** BASECASE gfedcb **BUS** 40821 (PAUL_500.0),**CASE** BASECASE gfedcb **BUS** 40869 (RAVER_500.0),**CASE** BASECASE Build Date: February 15, 2012 52

Q_sync (Mvar) QV Curves 270 260 250 240 230 220 210 200 190 180 170 160 150 140 130 120 110 100 90 80 70 60 50 40 30 20 10 0-10 -20-30 -40-50 -60 N-1 Raver-Paul is also at a voltage stability limit with same gen shift as Paul-Allston&Napavine- Allston limit. Critical bus is at Paul 0.97 0.98 0.99 1 1.01 1.02 1.03 Voltage (pu) 1.04 1.05 1.06 1.07 1.08 1.09 1.1 gfedcb **BUS** 40821 (PAUL_500.0),**CASE** BASECASE gfedcb **BUS** 40869 (RAVER_500.0),**CASE** BASECASE Build Date: February 15, 2012 The table below reports the branch overloads at the generation shift representing the WOCN voltage stability limit. 53

Line Records Owner Name 1 From NumFrom Name From NTo NumbeTo Name To NomCircuitMax % Loading Max % Ld Cont Comp Worst Increased ViolaViolationsNew ViolaMax % Ld Cont Name Portland General Elec 43348 MURRAY H 230 43541 ST MARYS 230 1 172.79 76.01 96.78 4 3 N-1: KEELER500-PEARL500C1 Tacoma Power 46607 COWLITZ 115 46689 COLLINS2 115 1 147.29 84.16 63.14 1 0 N-2: C3_32 CANYON BANK & CANYON-COWLITZ #1 PacifiCorp - West 45299 TROUTDAL 115 45303 TROUTPP2 230 1 138.72 93.32 55.41 93 55 BF 4283 KEELER-PEARL-OSTRANDER Bonneville Power Admin 40187 CASCD LK 115 40541 HOOD RVR 115 1 132.6 86.64 48.69 3934 3838 N-2: BIG EDDY-OSTRANDER 500 & BIG EDDY-TROUTDALE 230 Bonneville Power Admin 40013 ACTON 115 40187 CASCD LK 115 1 130.59 84.6 48.6 3917 3859 N-2: BIG EDDY-OSTRANDER 500 & BIG EDDY-TROUTDALE 230 PacifiCorp - West 45073 CULLY 115 45328 HOLLYW T 115 1 124.93 88.51 46.4 14 1 BF 4283 KEELER-PEARL-OSTRANDER PacifiCorp - West 45143 HLLYWOOD 115 45328 HOLLYW T 115 1 124.92 88.51 46.39 14 1 BF 4283 KEELER-PEARL-OSTRANDER Tacoma Power 46631 N-EAST 115 46743 PRAX TP 115 1 112.78 72.78 39.99 2 0 N-2: C3_13 TAC-COW #1 230 KV & TAC-SW 230 KV Portland General Elec 43761 GLENCOE# 115 43581 TABOR 115 1 121.24 81.25 39.99 9 0 BF 4283 KEELER-PEARL-OSTRANDER Portland General Elec 43007 ALDERCRT 115 43761 GLENCOE# 115 1 109.3 70.36 38.94 2 1 BF 4283 KEELER-PEARL-OSTRANDER PacifiCorp - West 45299 TROUTDAL 115 47214 RUNYAN 115 1 116.22 78.41 36.24 6 5 BF NORTH BONNEVILLE - ROSS 230 #1 LINE Portland General Elec 43039 BETHEL 230 90131 BETSAN11 230 1 108.15 73.15 35 4 0 BF 4475 PEARL-MARION & PEARL 500/230 BK2 Bonneville Power Admin 41403 7MILE 115 40541 HOOD RVR 115 1 104.53 74.27 30.72 9 3 N-2: BIG EDDY-OSTRANDER 500 & BIG EDDY-TROUTDALE 230 Bonneville Power Admin 41403 7MILE 115 41071 THE DALS 115 1 104.53 74.27 30.72 9 3 N-2: BIG EDDY-OSTRANDER 500 & BIG EDDY-TROUTDALE 230 Portland General Elec 40824 PEARL E 230 43773 PEARL # 230 1 118.32 104.71 30.64 3 0 L_PEARLW230-SHERWOOD230C1-MS Portland General Elec 43313 MCLOUGLN 230 90126 BIGMCL13 230 1 113.53 85.54 29.65 15 0 N-2: BIG EDDY-OSTRANDER 500 & BIG EDDY-TROUTDALE 230 Bonneville Power Admin 45021 BOYER 115 43223 GRNDROND 115 1 104.04 76.82 27.5 7 4 L_CARLTON230-TILLAMOK230C1 Portland General Elec 43801 CARVER1 115 43595 TOWNCNTR 115 1 101.09 74.06 27.03 1 0 BF 4283 KEELER-PEARL-OSTRANDER Northern Wasco PUD 47003 3MILE TP 115 47019 CHEIGHTS 115 1 108.69 84.87 23.82 2 0 N-1: The Dalles-Chenoweth Northern Wasco PUD 47001 2ND ST 115 47003 3MILE TP 115 1 105.82 82.02 23.8 2 0 N-1: The Dalles-Chenoweth Bonneville Power Admin 40213 CHEMAWA 230 40211 CHEMAWA 115 1 113.18 99.04 23.2 8 0 T_SALEM230-SALEM115C1 Bonneville Power Admin 47028 DISCV NW 115 41071 THE DALS 115 1 101.78 80.5 21.28 1 0 L_CHEIGHTS115-SEUFERT115C1 Portland General Elec 43773 PEARL # 230 43527 SHERWOOD 230 1 107.29 86.94 20.35 2 0 L_PEARLW230-SHERWOOD230C1-MS PacifiCorp - West 43291 LINNEMAN 230 45301 TROUTPP1 230 1 142.71 125.08 17.62 7 0 L_TROUTDW230-TROUTPP2230C1 Seattle City Light 46403 BOTHELL 230 90150 BOTSNO21 230 2 98.21 82.16 16.05 1 0 L_BOTHELL230-SNOKS1230C1-MS Bonneville Power Admin 90150 BOTSNO21 230 41008 SNOK S3 230 2 98.19 82.16 16.04 1 0 L_BOTHELL230-SNOKS1230C1-MS Bonneville Power Admin 40306 COVINGT5 500 40304 COVINGTE 230 5 109.18 95.74 13.44 1 0 N-2: RAVER-COVINGTON & RAVER-TACOMA 500 PacifiCorp - West 44900 DIXNV230 230 45093 DIXONVLE 230 1 112.9 102.46 11.27 3 0 BF 5081,84,87 Clear Alvey 500kV Bus PacifiCorp - West 40049 ALVEY 230 46291 MCKEN TP 230 1 120.47 109.54 10.94 1 0 BF SANTIAM 230 (BUS SECTIONALZING BREAKER) PacifiCorp - West 43215 GRESHAM 230 45303 TROUTPP2 230 1 114.91 104.87 10.03 5 0 BF BSBF MCLOUGHLIN TIE V668 230KV Portland General Elec 43684 BLUELAKE 115 43685 BLUELAKE 230 1 107.35 97.46 9.89 1 0 N-2: GRESHAM-TROUTDALE-TROUTDALE,LINNEMAN-TROUTDALE 54

C5: Chief Joe-Monroe #2 500kV & Remove one 345kV,Monroe-Snoh 230kV added Another robust line addition alternative is the Chief Jo-Monroe 500kV #2 and removal of one Chief Jo- Snohomish 345kV line and associated 345/230kV transformers. The PV summary is in the following table. PV Scenarios Overview PV Scenario Max Shift N-2: PAUL-STSOP, PAUL-OLY 1065.63 BF 4272 INGLEDOW-CUSTER-MON 1200 N-2: Pearl-Ostrander 500kV/Pearl-Marion 500kV 3498.88 BF 4519 Cust-Mon #1 500kV & Mon Caps 3509.38 N-2: CUSTER-MONROE 1&2 3568.75 BF 4530 RAVER-PAUL-SATSOP 3589.5 L_FAIRMONT230-SHELTON230C3 3762.5 BF A1610 FAIRMONT 230 3762.5 L_FAIRMONT230-SHELTON230C4 3764.5 N-2: PAUL-ALLSTON 2 & ALLSTON-NAPAVINE 500 4000 N-2: MONROE-CUSTER #1 500 & CHIEF JO-SNOHOMISH #4 4050 BF 4554 OLYMPIA-PAUL-TONO 4062.5 BF 4526 ECHO LAKE-MONROE-CUSTER 4193.75 BF OLYMPIA WEST BUS 4381.25 N-2: PAUL-SATSOP 500 & OLYMPIA-SATSOP 230 4423.88 L_MAPLEV3230-MV-SVC230C1 4503.13 BF 4548 ALLSTON-PAUL-SATSOP 4714.5 BF 4322,24,94 Clear Keeler 500kV 4762.5 BF OLYMPIA 230 E BUS - FLT ANY LN 4762.5 BF 4540 NAPAVINE-PAUL-SATSOP 4762.5 BF 4394 ALLSTON-KEELER-PEARL 4762.5 BF 4502 NAPAVINE-ALLSTON-KEELER 4904 N-1: Paul-Satsop & Satsop 500/230kV tx 4993.75 BF KEELER 230 - FAULT ANY LINE 5021.88 N-2: OLY-SHLT #3&4 D. CKT 5253.13 BF 5179 VANTAGE-SCHULTZ-RAVER #4 5300 N-2: Schultz-EchoLake & Schultz-Raver 500kV 5303.13 N-2: Schultz-Echo Lake 500kV / Schultz-Raver #4 500kV 5303.13 BF 5121 MAPLE VALLEY-ECHO LAKE-SCHULTZ-ECHO LAKE CAPS 5431.25 BF 5111 MON-ECHO L-ECHO L CAPS 5431.25 BF 5075 SCHULTZ-ECHO LAKE-ECHO LAKE CAPS 5431.25 A check of monitored bus self dv/dq and voltage at Raver indicates the maximum generation shift is overstated. 55

PU Volt dv/dq 0.00015 0.00014 0.00013 0.00012 0.00011 0.0001 0.00009 0.00008 0.00007 0.00006 0.00005 0.00004 0.00003 0.00002 0.00001 0-0.00001-0.00002-0.00003-0.00004-0.00005-0.00006 Low Voltage Solution with Schultz-Raver&Schultz-EchoLake overstates maximum shift 0 500 1,000 1,500 2,000 2,500 3,000 Nominal Shift 3,500 4,000 4,500 5,000 N-2: Schultz-EchoLake & Schultz-Raver 500kV: RAVER_500.0 (40869) Build Date: February 15, 2012 1.084 1.082 1.08 1.078 1.076 1.074 1.072 1.07 1.068 1.066 1.064 1.062 1.06 1.058 1.056 1.054 1.052 1.05 1.048 1.046 1.044 1.042 1.04 1.038 1.036 1.034 1.032 1.03 1.028 1.026 1.024 1.022 1.02 1.018 1.016 0 By trial and error, the maximum shift without a low voltage solution was found to be about 4200 MW 500 1,000 1,500 2,000 2,500 Nominal Shift 3,000 3,500 4,000 4,500 5,000 N-2: Schultz-EchoLake & Schultz-Raver 500kV: RAVER_500.0 (40869) Build Date: February 15, 2012 56

The contingency analysis comparing branch loading before and after the maximum generation shift shows overloads Line Records From NumFrom Name From NomTo NumbeTo Name To Nom kvcircuit Max % Loading CMax % Ld Cont CompWorst IncrViolationsNew ViolaMax % Ld Cont Name 43348 MURRAY H 230 43541 ST MARYS 230 1 158.07 79.56 81.34 6 3 N-1: KEELER500-PEARL500C1 45299 TROUTDAL 115 45303 TROUTPP2 230 1 127.85 93.48 44.21 39 3 BF 4283 KEELER-PEARL-OSTRANDER 40013 ACTON 115 40187 CASCD LK 115 1 119.03 84.84 36.05 80 16 N-2: BIG EDDY-OSTRANDER 500 & BIG EDDY-TROUTDALE 230 40187 CASCD LK 115 40541 HOOD RVR 115 1 121.02 86.88 36.04 117 10 N-2: BIG EDDY-OSTRANDER 500 & BIG EDDY-TROUTDALE 230 45073 CULLY 115 45328 HOLLYW T 115 1 114.02 88.64 35.2 14 0 BF 4283 KEELER-PEARL-OSTRANDER 45143 HLLYWOOD 115 45328 HOLLYW T 115 1 114.02 88.64 35.19 14 0 BF 4283 KEELER-PEARL-OSTRANDER 43761 GLENCOE# 115 43581 TABOR 115 1 110.51 81.5 29.01 8 0 BF 4283 KEELER-PEARL-OSTRANDER 43007 ALDERCRT 115 43761 GLENCOE# 115 1 99.32 70.6 28.71 1 0 BF 4283 KEELER-PEARL-OSTRANDER 47117 ARIEL 115 40175 CARDWELL 115 1 106.07 105.45 28.26 7 4 L_VIEWTAP115-MERWIN115C1 45299 TROUTDAL 115 47214 RUNYAN 115 1 106.12 78.54 27.57 4 3 BF ROSS 230 EAST BUS, FAULT NORTH BONNEVILLE #2 40824 PEARL E 230 43773 PEARL # 230 1 116.16 104.76 26.37 3 0 L_PEARLW230-SHERWOOD230C1-MS 45021 BOYER 115 43223 GRNDROND 115 1 98.64 77 21.63 2 0 L_CARLTON230-TILLAMOK230C1 43313 MCLOUGLN 230 90126 BIGMCL13 230 1 106.33 85.7 20.63 4 0 N-2: BIG EDDY-OSTRANDER 500 & BIG EDDY-TROUTDALE 230 47003 3MILE TP 115 47019 CHEIGHTS 115 1 103.91 84.99 18.92 2 0 N-1: The Dalles-Chenoweth 47001 2ND ST 115 47003 3MILE TP 115 1 101.04 82.14 18.89 2 0 N-1: The Dalles-Chenoweth 40217 CHENOWTH 115 47028 DISCV NW 115 1 118.97 102.01 16.95 5 0 L_CHEIGHTS115-SEUFERT115C1 43773 PEARL # 230 43527 SHERWOOD 230 1 103.61 87.05 16.55 1 0 L_PEARLW230-SHERWOOD230C1-MS 43291 LINNEMAN 230 45301 TROUTPP1 230 1 139.62 125.22 14.4 7 0 L_TROUTDW230-TROUTPP2230C1 40049 ALVEY 230 46291 MCKEN TP 230 1 121.11 109.6 11.52 1 0 BF SANTIAM 230 (BUS SECTIONALZING BREAKER) 43763 HEMLOCK# 115 43477 ROCKWD 2 115 1 101.66 90.77 10.9 1 0 N-2: GRESHAM-TROUTDALE-TROUTDALE,LINNEMAN-TROUTDALE 40213 CHEMAWA 230 40211 CHEMAWA 115 1 109.77 99.13 10.72 4 0 T_SALEM230-SALEM115C1 44900 DIXNV230 230 45093 DIXONVLE 230 1 110.45 102.52 8.65 3 0 BF 5081,84,87 Clear Alvey 500kV Bus 40306 COVINGT5 500 40304 COVINGTE 230 5 105.83 97.49 8.34 1 0 N-2: RAVER-COVINGTON & RAVER-TACOMA 500 43215 GRESHAM 230 45303 TROUTPP2 230 1 112.57 104.95 7.63 4 0 BF BSBF MCLOUGHLIN TIE V668 230KV 43684 BLUELAKE 115 43685 BLUELAKE 230 1 104.37 97.53 6.83 1 0 N-2: GRESHAM-TROUTDALE-TROUTDALE,LINNEMAN-TROUTDALE 45195 MERIDINP 230 45197 MERIDINP 500 2 126.64 121.24 5.4 1 0 BF 11R6 DIXONVILLE-MERIDIAN-MERIDIAN TX 1 + REAC 57

Q_sync (Mvar) New Failed Solutions (Comp Solved= before gen displacement, Solved=after gen displacement) Contingency Records Label Solved Comp Solv BF 4272 INGLEDOW-CUSTER-MON NO YES BF 4530 RAVER-PAUL-SATSOP NO YES BF 4554 OLYMPIA-PAUL-TONO NO YES BF A1613 FAIRMONT 230 NO YES BF A1622 FAIRMONT 230 NO YES BF A79 SHELTON 230 NO YES BF A83 SHELTON 230 NO YES BF OLYMPIA WEST BUS NO YES BUS: Andrew York Bus Section Brkr Fail 115 NO YES N-2: PAUL-ALLSTON 2 & ALLSTON-NAPAVINE 500 NO YES N-2: PAUL-SATSOP 500 & OLYMPIA-SATSOP 230 NO YES N-2: PAUL-STSOP, PAUL-OLY NO YES N-2: Pearl-Ostrander 500kV/Pearl-Marion 500kV NO YES BF 4519 Cust-Mon #1 500kV & Mon Caps NO YES At the 4193 MW generation shift limit for WOCN, several contingencies outside of the WOCN are also at or above a voltage stability limit. Four examples are shown below. These include eastside contingencies (Schultz-Vantage&Schultz-Wautoma), I-5 outages (Raver-Paul, Paul-Allston&Napavine-Allston), and cross mountain lines to the Willamette Valley (John Day-Marion&Buckley-Marion). 240 230 QV Curves 220 210 200 190 180 170 160 150 Schultz-Wautoma&Schultz-Vantage, critical bus at Allston 140 130 120 110 100 90 80 70 60 50 40 30 20 10 0-10 -20-30 -40-50 -60 1.02 1.03 1.04 1.05 1.06 Voltage (pu) 1.07 1.08 1.09 1.1 gfedcb gfedcb **BUS** 40045 (ALLSTON_500.0),**CASE** BASECASE gfedcb **BUS** 40869 (RAVER_500.0),**CASE** BASECASE **BUS** 40749 (MONROE_500.0),**CASE** BASECASE gfedcb **BUS** 40827 (PEARL_500.0),**CASE** BASECASE Build Date: February 15, 2012 58

Q_sync (Mvar) Q_sync (Mvar) **BUS** 40821 (PAUL_500.0),**CASE** BASECASE 240 220 200 180 160 140 120 Raver-Paul outage, critical bus at Paul 100 80 60 40 20 0-20 -40 0.99 1 1.01 1.02 1.03 1.04 Voltage (pu) 1.05 1.06 1.07 1.08 1.09 1.1 Build Date: February 15, 2012 QV Curves 180 170 160 150 140 130 120 110 JohnDay-Marion&Buckley-Marion, critical bus in Pearl/Keeler/Allston area 100 90 80 70 60 50 40 30 20 10 0-10 -20-30 -40 1.02 1.03 1.04 1.05 1.06 Voltage (pu) 1.07 1.08 1.09 1.1 gfedcb **BUS** 40045 (ALLSTON_500.0),**CASE** BASECASE gfedcb **BUS** 40699 (MARION_500.0),**CASE** BASECASE gfedcb **BUS** 40827 (PEARL_500.0),**CASE** BASECASE Build Date: February 15, 2012 59

Q_sync (Mvar) **BUS** 40821 (PAUL_500.0),**CASE** BASECASE 800 750 700 650 600 550 500 450 Paul-Allston&Napavine-Allston, critical bus at Paul, 4193 shift was over voltage stability limit 400 350 300 250 200 150 100 50 0 0.98 1 1.02 1.04 1.06 1.08 1.1 Voltage (pu) 1.12 1.14 1.16 1.18 1.2 Build Date: February 15, 2012 60

C6: Chief Joe-Monroe #2&3 500kV & Remove both 345kV,Monroe-Snoh 230kV added The TCRM studies indicated benefit of converting the unused 345kV line between Monroe and Snohomish to a 230kV circuit between Monroe and Snohomish. The PV simulation summary and the low voltage solution check is shown below. PV Scenarios Overview PV Scenario Max Shift N-2: PAUL-STSOP, PAUL-OLY 1065.63 BF 4272 INGLEDOW-CUSTER-MON 3345.75 N-2: Pearl-Ostrander 500kV/Pearl-Marion 500kV 3398.88 N-2: CUSTER-MONROE 1&2 3630.13 BF 4519 Cust-Mon #1 500kV & Mon Caps 3630.13 BF 4530 RAVER-PAUL-SATSOP 3803.13 N-2: PAUL-SATSOP 500 & OLYMPIA-SATSOP 230 4159.38 BF 4502 NAPAVINE-ALLSTON-KEELER 4225 BF 4554 OLYMPIA-PAUL-TONO 4320.75 BF 4548 ALLSTON-PAUL-SATSOP 4320.75 BF OLYMPIA WEST BUS 4320.75 N-2: PAUL-ALLSTON 2 & ALLSTON-NAPAVINE 500 4320.75 BF OLYMPIA 230 E BUS - FLT ANY LN 4359.38 BF 4540 NAPAVINE-PAUL-SATSOP 4362.5 N-2: MONROE-CUSTER #1 500 & CHIEF JO-SNOHOMISH #4 4418.75 BF 4322,24,94 Clear Keeler 500kV 4512.5 BF 4394 ALLSTON-KEELER-PEARL 4512.5 N-1: Paul-Satsop & Satsop 500/230kV tx 4706.25 BF KEELER 230 - FAULT ANY LINE 5000 N-2: Schultz-Echo Lake 500kV / Schultz-Raver #4 500kV 5003.13 L_KEEL-SVC230-KEELER230C1 5042.63 N-2: COULEE-SCHULTZ 1&2 5055.75 N-2: SCHULTZ-WAUTOMA 500 & VANTAGE-SCHULTZ 500 5168.75 N-2: Schultz-EchoLake & Schultz-Raver 500kV 5381.25 N-2: Chief Jo-Monroe 2&3 5428.13 BF 5111 MON-ECHO L-ECHO L CAPS 5518.75 BF 5075 SCHULTZ-ECHO LAKE-ECHO LAKE CAPS 5525 BF 5121 MAPLE VALLEY-ECHO LAKE-SCHULTZ-ECHO LAKE CAPS 5533.25 61

dv/dq 0.00006 0.00005 0.00004 0.00003 0.00002 0.00001 0 Low Voltage Solution with Schultz-Raver&Schultz-EchoLake overstates maximum shift -0.00001-0.00002-0.00003-0.00004-0.00005 0 500 1,000 1,500 2,000 2,500 3,000 Nominal Shift 3,500 4,000 4,500 5,000 N-2: Schultz-EchoLake & Schultz-Raver 500kV: RAVER_500.0 (40869) Build Date: February 15, 2012 62

PU Volt 1.085 1.08 1.075 1.07 1.065 1.06 1.055 1.05 1.045 By trial and error, the maximum shift without a low voltage solution was found to be about 5000 MW 1.04 1.035 1.03 1.025 1.02 0 1,000 2,000 3,000 Nominal Shift 4,000 5,000 N-2: Schultz-EchoLake & Schultz-Raver 500kV: RAVER_500.0 (40869) Build Date: February 15, 2012 63

APPENDIX D TCRM Analysis TCRM (Transmission Curtailment Risk Measure) analysis was performed as part of the evaluation of the proposed Cross Cascades North reinforcement projects to identify potential impacts on transfers across the Northern Intertie and through the Puget Sound Area. TCRM analysis was developed by the Puget Sound Area Study Team (PSAST) to predict the probability of needing to reduce transfers to remain within reliability criteria. Whenever an individual nomogram point indicated that the Operational Transfer Capability (OTC) on the Northern Intertie would be less than the level of firm commitments, the amount of shortfall was quantified. Each of the shortfall amounts were multiplied by weighting factors and then combined into a single value by summing the weighted amounts together to obtain a TCRM value for each transfer direction. The following assumptions were used in the TCRM analysis: 1) PSAST projects consist of: a. SCL 6 ohm inductors at Broad St b. Bothell-SnoKing 230kV #1&#2 High Temp conductors c. Lakeside 230kV line and transformer addition d. Delridge-Duwamish 230kV High Temp conductors e. Third Covington 500/230kV transformer addition 2) Monroe-Novelty 230kV line upgraded to 100C 3) Removed N-2: MON-ECH-SK and MON-NOV/ N-2: MON-ECH-SK and MON-SAM secondary contingencies 4) For North to South studies: a. Series caps on lines into Monroe were bypassed b. SCL inductors at Broad St were switched out 5) For South to North studies: a. Series caps on lines into Raver or Echo Lake were bypassed b. SCL inductors at Broad St were switched in 6) Initial transfer level = 1500MW 7) Assumed firm transmission: a. 1500 MW South to North b. 2100 MW North to South 64

The result of TCRM analysis is summarized in the table below: PROJECT CASE North to South South to North TCRM Change from Base Case TCRM Change from Base Case PSAST Projects In - BASE CASE 45,525 20,863 Schultz series capacitor size increase & Schultz-Raver #4 reconductor 43,880-1,645 15,291-5,572 All Projects Below have Schultz Series Caps & Schultz- Raver #4 upgrade Chief Joe-Monroe #2 500kV 59,326 13,801* 12,014-8,849 Chief Joe-Monroe #2 500kV & Remove one 345kV,Monroe-Snoh 230kV added Chief Joe-Monroe #2&3 500kV & Remove both 345kV,Monroe-Snoh 230kV added 62,608 17,083 16,769-4,094 370,694 325,169* 23,936 3,073 Sickler-Monroe 500kV (with series compensation) 53,823 8,298* 5,139-15,724 Sickler-Monroe 500kV, convert to 500kV Coulee-Schultz section of Coulee-Oly 287kV 51,487 5,962 5,189-15,674 Upgrade Rocky Reach-Maple Valley 345kV to 500kV 47,579 2,054 69,630 48,767 Upgrade Rocky Reach-Maple Valley 345kV to 500kV, convert Coulee-Schultz section of CouOly 46,590 1,065 136,934 116,071* Upgrade Coulee-Olympia 287kV to 500kV 40,504-5,021 19,651-1,212 Sickler-Raver 500kV 39,691-5,834 41909 21,046* Sickler-Raver 500kV, convert to 500kV Coulee-Schultz section of Coulee-Oly 287kV 37,419-8,106 48,974 28,111 * TCRM increase can be reduced with additional prerequisite projects as illustrated below. 65

The TCRM values determined for the base case with the PSAST projects can be compared to the TCRM for each proposed Cross Cascades North project to determine its impact on the Northern Intertie. In cases where there is a significant decrease in TCRM, the proposed project would decrease exposure to curtailments of transfers between the Northwest and British Columbia. Conversely in cases where the TCRM significantly increases, the proposed project would increase the likelihood of curtailments of transfers and prerequisite projects will be needed to mitigate. The results show that new transmission line options that terminate at Monroe (Chief Joe-Monroe #2 or Sickler-Monroe) reduce the South to North TCRM and increase the North to South TCRM without the implementation of additional prerequisite projects. The TCRM analysis also shows which facilities and secondary contingencies contribute to the final TCRM value. From this information, additional prerequisite projects were identified to mitigate the impact of WoCN reinforcements on the Northern Intertie. For illustrative purposes, a few WOCN projects and potential additional prerequisite projects are highlighted below: a) Chief Joe-Monroe #2 500kV: The North to South TCRM analysis shows that the top limiting facility is the Custer W-Portal Way transformer. Around 70% of the TCRM value can be mitigated with a second Portal Way 230/115kV transformer. A sensitivity study that modeled a second Portal Way transformer showed that the TCRM for the base case was reduced to 15,060 and the TCRM with Chief Joe- Monroe #2 was reduced to 19,137. b) Chief Joe-Monroe #2&3 500kV option that removes both 345kV lines and adds 230kV between Monroe and Snohomish: For this proposed line project, the top limiting facilities were the Monroe 500/230kV transformer and the Custer-Portal Way transformer. This analysis shows that a second Monroe 500/230kV transformer and a second Custer-Portal Way transformer would be prerequisites to mitigate the impact of the project. 66

c) Sickler-Monroe 500 kv with series compensation: Similar to Chief Jo-Monroe #2 line, this project also requires a second Portal Way 230/115kV transformer. d) Upgrade of Rocky Reach-Maple Valley to 500kV: The large increase in South to North TCRM is associated with the limiting facility of Maple Valley 500/230kV transformer. To mitigate this issue, a second Maple Valley 500/230kV transformer will be required. Alternatively, the removal of the secondary contingency Breaker Failure 5114: Raver-Echo Lake- Monroe will also decrease the TCRM for this project. e) Sickler-Raver 500 kv addition: For this project, the limiting facility Sammamish-Sammamish E 230kV switch contributes most significantly to the increase in TCRM. A prerequisite project could be an upgrade of the switch. 67

8338 NE Alderwood Rd Suite 140 Portland, OR 97220 (503) 943-4940 www.columbiagrid.org