Closingthe Planning Gaps: How Transmission Improves Resource Adequacy in US Power Systems

November 06, 2026 | Blog

Part A — Resource Adequacy and Transmission Planning at a Glance

Every US utility, regional transmission organization (RTO) and state commission asks the same two questions about the future grid. Will there be enough generation, storage and demand response to serve load through the hottest summer afternoon and the coldest winter morning? And will the transmission network be able to deliver that supply where it is needed, reliably and at reasonable cost?

In most of the country those questions are answered by three separate processes, run by different organizations, on different schedules, with different models and different data:

• Resource adequacy (RA) assessments, run by balancing authorities, RTOs or utilities, estimate the probability of running short of supply and set the planning reserve margin (PRM) and resource accreditation rules.
• Resource planning — integrated resource plans (IRPs) at vertically integrated utilities and state agencies, and capacity expansion studies at RTOs — chooses the future generation and storage portfolio.
• Transmission planning, run by transmission owners (TOs) and regional planning entities, tests the future network against NERC reliability standards and, increasingly, economic and multi-value criteria.

Each process uses the outputs of the others, but usually from the previous cycle. Transmission topology and import limits enter the RA study as fixed assumptions. The future generation portfolio enters the transmission study as a fixed input. Nobody asks, inside one integrated model, whether a new 765 kV line or HVDC tie could replace a gigawatt of local peaking capacity at lower total cost.

Keentel Engineering calls these handoffs the planning gaps. This guide explains where they come from, why they raise costs for customers, how transmission actually delivers resource adequacy value, how to quantify that value with standard probabilistic methods, and what has changed with FERC Order No. 1920.

Keentel's companion article on cluster interconnection studies covers how individual generators connect to the grid under FERC Order No. 2023. This article looks one level up: how the system decides how much generation it needs and how much transmission should carry it.

The planning chain in one table

What Keentel's analysis shows

To make the value of transmission concrete, Keentel built a two-area probabilistic resource adequacy model. Each area has a 10,000 MW peak, a fleet of 100–600 MW thermal units with forced outage rates of 4–6%, and a year of daily peak loads with seasonal shape and weather variation. Each area alone needs about 11,955 MW of installed capacity — a 19.6% planning reserve margin — to meet a loss-of-load expectation (LOLE) of 0.1 day per year. Keentel then connected the two areas with a tie of 0 to 3,000 MW and asked how much capacity each area really needs.

Figure 1. Keentel's two-area probabilistic model. Left: LOLE in Area A when each area keeps its isolated 0.1 day/yr fleet and a tie of increasing capacity is added. Right: installed capacity each 10 GW-peak area must carry to meet 0.1 day/yr, with the equivalent planning reserve margin on the right axis. Results are illustrative.

Three results stand out:

• A 2,000 MW tie cut each area's capacity need by about 1,000 MW when the two areas' peak loads were only moderately correlated (correlation coefficient about 0.5, with peaks shifted by about two months). Across both areas that is about 2,000 MW of capacity that never has to be built.
• Even with identical weather, the tie was worth about 670 MW per area. Generator forced outages are largely independent between regions, so sharing reserves helps even when loads peak together. Load diversity adds value on top of outage diversity.
• The benefit saturates. Beyond about 1,500–2,000 MW, more tie capacity adds almost no RA value in this model, because the binding events become ones where both areas are short at the same time. The first gigawatt of interregional transfer capability is worth far more than the fifth.



At an illustrative capacity cost of $100/kW-year, the 2,000 MW of avoided capacity in the low-correlation case is worth about $200 million per year — more than the roughly $135 million per year annualized cost of a $1.8 billion transmission project at 7% over 40 years. That comparison counts only resource adequacy, before any production cost, congestion or resilience benefits. It is exactly the benefit that today's separated processes tend to leave out.

Engineering review of common statements about planning integration

Five lessons

• Transmission is a resource adequacy resource. It shares reserves, captures weather and load diversity, and lets regions hold less capacity for the same reliability. It should be valued that way.
• The value is strongly nonlinear. The first few hundred megawatts of transfer capability between weakly correlated regions are worth the most. Studies that test only one large project, or only one tie size, miss the shape of the curve.
• Correlated extreme weather is the binding case. Winter storms that hit several regions at once — and correlated failures of gas supply and generating units in the cold — limit how much help neighbors can give. Multi-area models need correlated weather years and correlated outages, not just independent unit outages.
• Data and horizon alignment is the cheapest fix. Using the same load forecasts, resource assumptions and weather years in RA, IRP and transmission studies removes much of the inconsistency before any new model is built.
• Order No. 1920 creates the opening. Its 20-year horizon and its required "reduced loss of load probability or reduced planning reserve margin" benefit give planners a formal place to count RA value — if the RA and transmission teams actually exchange models and results.
  

Part B — Technical Guide: Closing the Planning Gaps

1. The vocabulary of resource adequacy

Resource adequacy is the ability of the electric system to supply aggregate demand and energy requirements at all times, accounting for scheduled and reasonably expected unscheduled outages of system components. It is a planning concept, measured probabilistically, and distinct from operating reliability (surviving the next contingency) and from transmission security (meeting NERC TPL performance requirements for specified contingencies).



1.1 Probabilistic metrics

The daily LOLE is computed as the sum, over the days of the year, of the probability that available capacity falls below the daily peak load:

Expected unserved energy sums the expected shortfall over every hour:

Here C is available capacity (installed capacity minus random forced outages, plus energy-limited and weather-dependent output) and L is load. The 0.1 day/yr target is a long-standing convention, not a physical law; it corresponds to one loss-of-load day in ten years. Modern systems with high shares of solar, wind and storage can meet 0.1 day/yr while still carrying large EUE in a single long winter event, which is why many planners now examine several metrics together.

1.2 Planning reserve margin



The planning reserve margin translates the probabilistic target into a simple capacity requirement:

Two conventions are in use. An installed-capacity (ICAP) PRM counts nameplate megawatts and must therefore be large enough to cover forced outages. An unforced-capacity (UCAP) or accredited-capacity PRM first derates each resource for its expected unavailability, so the required margin is smaller. Comparing PRMs across regions without checking the convention is a common mistake.

where EFORd is the equivalent demand forced outage rate. For variable and energy-limited resources, accreditation uses ELCC instead.

1.3 Effective load carrying capability (ELCC)

ELCC is the amount of perfectly reliable capacity that would provide the same reliability improvement as the resource being evaluated. It is found by adding the resource to the system, measuring the new LOLE or EUE, and then finding the amount of perfect capacity that achieves the same result.

• Average ELCC spreads the combined value of a resource class across all of its megawatts.
• Marginal ELCC measures the value of the next megawatt. As solar penetration grows, the net peak moves into the evening and the marginal ELCC of solar falls sharply. Storage shows a similar but slower decline as the net peak widens beyond its duration.

Several RTOs now use marginal or class-average ELCC for accreditation, and several are extending ELCC-style accreditation to thermal units using performance during the riskiest hours. Transmission does not receive an ELCC in these programs; its value appears through import limits, zonal requirements and the PRM itself.

Worked example 1 — LOLE from a capacity outage probability table

A small system has six identical 100 MW units, each with a forced outage rate of 5%. The probability that exactly k units are out is binomial:

The year's daily peaks are grouped into four bands: 30 days at 500 MW, 60 days at 450 MW, 120 days at 400 MW and 155 days at 350 MW. A shortfall occurs on a 500 MW day if two or more units are out (probability 0.03277), on a 450 MW day under the same condition, and on 400 MW and 350 MW days if three or more units are out (probability 0.00223).

Small systems with large units need very large reserve margins to reach 0.1 day/yr, because losing one unit is a large fraction of the fleet. This is the first, and oldest, reason interconnection pays: pooling makes each unit a smaller share of the combined fleet. The same convolution, extended to hundreds of units in 50 MW steps, is the engine of Keentel's two-area model.

2. How resource adequacy is assessed in the United States

2.1 The assessment process

A modern RA study has five building blocks:

• Load. Hourly load shapes for many historical weather years (often 20–40 or more), scaled to the forecast year, with economic uncertainty and new large loads such as data centers layered on top.
• Thermal resources. Unit capacities, forced outage rates and, increasingly, temperature-dependent derates and correlated outages in extreme cold or heat.
• Variable resources. Wind and solar profiles synchronized to the same weather years as load, so the correlation between weather, load and renewable output is preserved.
• Energy-limited resources. Storage, hydro and demand response with their duration and use limits, dispatched in the model to minimize shortfall.
• Transmission. Usually a single zone, or a few zones linked by fixed transfer limits, plus an assumption about how much help neighboring regions can provide.

The model is solved by convolution (for simple systems) or by sequential Monte Carlo simulation, which samples thousands of years of random outages against the weather years and records every shortfall.

2.2 Correlated events change the answer

Two winter events changed how US planners think about adequacy. In February 2021 a prolonged cold-weather event in the central United States caused massive simultaneous outages of gas-fired, coal, nuclear and wind generation, much of it linked to freezing equipment and natural gas supply interruptions, and forced large-scale rotating outages in Texas. In December 2022 a fast-moving winter storm across the Eastern Interconnection produced very high forced outages, again concentrated in gas-fired generation, and rotating outages in parts of the Southeast.



Both events showed that outages are not independent random draws. They are correlated with temperature, with fuel supply and with each other — and they can hit several regions at once. For transmission planning this cuts both ways. It limits how much help a neighbor can give during a wide-area event, which is why RA studies cap external assistance. But it also shows the value of strong ties between regions whose weather extremes do not coincide, and of the ability to move power across long distances during regional emergencies.

2.3 RA programs in the RTOs (summarized)

Program details change frequently; confirm current rules with each RTO before relying on them.

3. Resource planning and capacity expansion

3.1 What an IRP or capacity expansion study does

A capacity expansion model chooses which resources to build, retire and operate over a 15–25 year horizon to minimize total cost subject to load, policy and reliability constraints. Reliability enters as a constraint: accredited capacity must meet peak load times one plus the PRM, often with seasonal and local requirements. Transmission is usually represented as a "pipe-and-bubble" or transport model — zones connected by fixed flow limits with no Kirchhoff's voltage law, no contingencies and no losses beyond a simple factor.

3.2 Why the representation matters

A transport model is not wrong; it is coarse. Its weaknesses for this problem are specific:

• Transfer limits are inputs, fixed from the previous transmission cycle, so the model cannot trade transmission against generation.
• Interface limits in a real network depend on the dispatch pattern and on contingencies; a single number hides that.
• Capacity expansion uses accredited capacity values computed once, outside the model. If transmission changes the PRM or the ELCC of resources in each zone, the expansion model never sees it.

Worked example 2 — PRM and accreditation for a mixed portfolio

A 10,000 MW-peak area has the following resources. Thermal units are accredited at ICAP × (1 − EFORd); variable and storage resources at an illustrative ELCC.

On an installed basis the PRM looks generous:

On an accredited basis the margin is only 9.4%. If the region's accredited requirement is, say, 12%, this area is about 260 MW short even though its installed reserve margin is 42%. The same area with access to more import capability could meet its target through the tie — but only if the RA study credits that import and the capacity expansion model knows it is available.

Worked example 3 — declining marginal capacity value of solar

Keentel modeled a simplified 10,000 MW system whose daily load peaks at 5 p.m. and a solar fleet whose output follows a midday arc ending around 7 p.m. The capacity value is measured as the reduction in net peak load per megawatt of solar added.

Once the net peak shifts past sunset, more solar adds no capacity value in this simple model. Real ELCC studies, which consider many weather years and storage, show a similar shape. Transmission interacts with this effect: a tie to a region whose net peak occurs at a different hour can restore part of the lost value, because the combined system's net peak is broader and lower. A single-zone ELCC study cannot see that benefit.

4. Transmission planning in the United States

4.1 Reliability planning under NERC TPL-001-5.1

NERC Reliability Standard TPL-001-5.1 requires Planning Coordinators and Transmission Planners to assess their systems annually for a near-term horizon (years one through five) and a long-term horizon (years six through ten). The assessment uses steady-state, short-circuit and stability studies across a defined set of planning events, from single contingencies (P1) through extreme events, with stated performance requirements. Where violations are found, the planner must develop a Corrective Action Plan.

TPL-001 is deterministic. It asks whether the system meets performance requirements for specified contingencies at specified load and dispatch conditions. It does not ask how likely a capacity shortfall is, and it assumes that enough resources exist to serve the modeled load. That is why transmission reliability studies "ignore resource adequacy": it is not their job under the standard.

4.2 Regional and interregional planning under FERC Order No. 1000

FERC Order No. 1000 (2011) required each FERC-jurisdictional transmission provider to participate in a regional planning process that produces a regional plan, considers transmission needs driven by public policy requirements, and allocates the costs of selected regional facilities roughly commensurate with benefits. It also required interregional coordination procedures between neighboring regions. In practice, regional processes produced reliability, economic and policy-driven portfolios, while very few interregional projects were selected, because each region's own process had to find the project beneficial on its own terms.

4.3 Long-term regional planning under FERC Order No. 1920

FERC Order No. 1920, issued in May 2024 and modified on rehearing by Orders No. 1920-A and 1920-B, requires FERC-jurisdictional transmission providers to conduct long-term regional transmission planning. Its key elements are:

• A planning horizon of at least 20 years, with long-term scenarios developed and updated at least every five years.
• At least three long-term scenarios, including a scenario that reflects high-impact, low-frequency extreme weather and other events.
• At least seven categories of factors in developing the scenarios, including federal, state and local laws and policies, utility resource plans and load forecasts, trends in technology and fuel costs, retirements, generator interconnection requests and withdrawals, and corporate commitments.
• At least seven benefits to be measured when evaluating long-term projects: avoided or deferred reliability transmission facilities and aging infrastructure replacement; a benefit that can be characterized as either reduced loss of load probability or reduced planning reserve margin; production cost savings; reduced transmission energy losses; reduced congestion due to transmission outages; mitigation of extreme weather events and unexpected system conditions; and capacity cost benefits from reduced peak energy losses.
• A formal role for states in cost allocation for long-term regional facilities, with an engagement period.
• Requirements to consider right-sizing of replacement facilities and certain grid-enhancing technologies.

The second required benefit is the formal bridge this article is about. For the first time, FERC-jurisdictional regions must measure whether new long-term transmission reduces loss of load probability or reduces the planning reserve margin. Compliance filings and implementation schedules vary by region, and several regions received extensions; planners should confirm the status in each region.

4.4 Other national studies

• NERC's Interregional Transfer Capability Study, delivered in late 2024 at Congress's direction, assessed transfer capability between neighboring regions and identified roughly 35 GW of prudent additions to interregional transfer capability that would strengthen reliability under extreme conditions. It is an assessment, not a plan or a mandate, but it gives planners a consistent national baseline.
• The US Department of Energy's National Transmission Needs Study assessed historical and anticipated transmission needs and identified large benefits from additional interregional capacity, particularly between regions with different weather and resource patterns.
• RTO long-range portfolios — for example MISO's long-range transmission planning tranches, SPP's integrated transmission planning with new 765 kV facilities, PJM's regional transmission expansion planning, CAISO's transmission planning process and ERCOT's 765 kV plans — show that 765 kV and 500 kV AC backbones, and in some cases HVDC, are back on the planning agenda.

5. How transmission delivers resource adequacy value

Transmission does not generate a megawatt, but it changes how many megawatts each region must hold. Five mechanisms are at work.

5.1 The two-area model

In a two-area model, Area A experiences a shortfall only if its own margin is negative and Area B cannot make up the difference over the tie. With M as the available margin (capacity minus load) and T the tie capacity:

This simple model makes three implicit assumptions worth stating: Area B helps only from its own surplus (it does not shed its own load to help A), the tie is fully available when needed, and no internal constraints limit delivery inside either area. Real multi-area models add tie outage rates, internal zones, sharing rules and priority between areas.

Worked example 4 — Keentel's two-area result in detail

Keentel's model uses two areas with 10,000 MW peaks. Each fleet contains 100–600 MW units with forced outage rates of 4–6%, totaling 11,500 MW before tuning; firm capacity was then added or removed in both areas until each met 0.1 day/yr in isolation (11,955 MW, PRM 19.6%). Daily peak loads follow a seasonal curve with a weather-driven random component; the low-correlation case shifts Area B's seasonal peak by about 60 days and gives a daily-peak correlation of about 0.49, while the high-correlation case uses identical load shapes. The capacity outage probability tables are convolved exactly in 50 MW steps and the margin distribution of Area B is combined with that of Area A on each day.

Results are in Figure 1 and the Part A table. In the low-correlation case, the capacity each area must carry falls from 11,955 MW to 10,946 MW with a 2,000 MW tie:

In the high-correlation case the saving is about 673 MW per area, all from outage diversity. Most of the benefit arrives by 1,000–1,500 MW of tie capacity; beyond about 2,000 MW the curve is flat.

The practical meaning: if a region's RA study assumes a fixed 500 MW of external assistance, it has already captured only the steep first part of this curve and cannot see the value of the next 1,500 MW. If it assumes zero assistance, it overstates its capacity need by up to a gigawatt.

5.2 Quantifying the RA value of transmission: four methods

Worked example 5 — does the RA value pay for the line?

Suppose the 2,000 MW tie in Worked example 4 is a 765 kV line costing $1.8 billion. At a 7% cost of capital over 40 years, the capital recovery factor is:

The annualized cost is $1.8 billion × 0.0750 ≈ $135 million per year, before operations and maintenance. At an illustrative avoided capacity cost of $100/kW-year:

On RA value alone, the line roughly pays for itself in the high-correlation case and returns about 1.5 times its cost in the low-correlation case. Production cost savings, loss reductions, congestion relief and extreme-weather benefits come on top. These numbers are illustrative; real studies must use the regions' actual fleets, weather years, line costs and cost of new entry.

5.3 Why external assistance is capped

RA studies commonly cap the help a region can receive from its neighbors, sometimes at a fixed number of megawatts or at zero. The reasons are real:

• Simultaneous scarcity. During wide-area extreme weather, neighbors may be short at the same time.
• Firmness. Without firm contracts or firm transmission service, there is no guarantee that a neighbor's surplus will be offered.
• Deliverability. Internal constraints in either region may prevent surplus capacity from reaching the tie.
• Double counting. If every region counts its neighbors' surplus, the same megawatts may be counted twice.

The problem is not the cap itself but how it is set. A fixed cap applied in every scenario, regardless of the transmission in the model, removes the transmission's RA value by assumption. Keentel recommends deriving assistance limits from multi-area probabilistic modeling with correlated weather, and recomputing them when transmission changes.

6. Transfer capability: the link between the two worlds

The number that connects RA studies and transmission studies is transfer capability: how many megawatts can be moved from one area to another without violating thermal, voltage or stability limits under normal and contingency conditions. NERC defines first-contingency incremental transfer capability (FCITC) as the amount of power, above base-case flows, that can be transferred while respecting limits after the most limiting single contingency.

Linear sensitivity factors make the calculation transparent. The power transfer distribution factor (PTDF) gives the change in flow on a monitored element per megawatt of transfer; the line outage distribution factor (LODF) gives the share of an outaged line's flow that shifts onto the monitored element.

Worked example 6 — FCITC on a 345 kV interface

A 345 kV tie line between two areas has an emergency rating of 1,250 MVA (treated as MW here for simplicity). Its base-case flow is 400 MW and its PTDF for an A-to-B transfer is 0.45. A parallel 345 kV line carries 300 MW, has a PTDF of 0.35, and has an LODF of 0.55 onto the monitored line.

Without contingencies, the transfer limit set by this line would be:

After loss of the parallel line, the monitored line's base flow becomes 400 + 0.55 × 300 = 565 MW, and its effective sensitivity becomes 0.45 + 0.55 × 0.35 = 0.6425:

The first-contingency limit is about 1,070 MW — much lower than the 1,889 MW that a "pipe" model using the line's rating might assume. Real FCITC studies scan thousands of contingencies, include voltage and stability limits and model dispatch-dependent base flows. They also show why an RA study that uses a single transfer number from the previous planning cycle can be badly out of date when new lines, retirements or large loads change the flow pattern.

7. The seven gaps and how to close them



Keentel groups the integration problems into seven gaps. The first three concern regional planning; the last four concern interregional planning and markets.

7.1 An integrated planning loop

Keentel recommends replacing the one-way chain in the typical process diagram with a loop:

• Step 1 — Common scenarios. Agree on 20-year scenarios, including load growth from data centers and electrification, retirements and policy, consistent with Order No. 1920.
• Step 2 — Candidate transmission. Define a set of transmission options with transfer capabilities computed by nodal FCITC analysis, including several sizes for each interregional interface.
• Step 3 — Multi-area RA. Compute LOLE, EUE and the required PRM for each transmission option using correlated weather years and derived assistance limits.
• Step 4 — Capacity expansion. Re-optimize the resource portfolio with the PRM and import limits from Step 3.
• Step 5 — Production cost and reliability testing. Test the combined portfolio for production cost benefits and TPL-001 reliability performance.
• Step 6 — Iterate. Repeat until the transmission plan and the resource plan stop changing materially.

The loop sounds heavy, but most of its value comes from the first iteration: discovering that the RA model's import assumptions and the transmission model's resource assumptions disagree.

7.2 Data and modeling checklist

• Use the same weather years for load, wind and solar across RA, IRP and transmission studies.
• Model temperature-dependent and correlated thermal outages, particularly in winter.
• Represent new large loads (data centers, industrial electrification) with realistic flexibility and ride-through assumptions.
• Derive zonal transfer limits from current nodal models, and record the dispatch and contingency that set each limit.
• Test several tie sizes to capture the nonlinear value curve.
• Report LOLE, LOLH and EUE together, and report the PRM convention (ICAP or accredited).
• Document every assumption about external assistance and its basis.

8. Cost allocation and incentives

Engineering can show that a line reduces capacity needs; cost allocation decides who pays and who benefits.

A controllable HVDC link, for example a ±525 kV or ±600 kV voltage-source converter line between two regions, is attractive for capacity sharing because its flow can be scheduled and its deliverability is clear. But without a market mechanism that credits its capacity value to the party paying for it, much of that value flows to customers who did not fund it. Closing Gap 7 requires market design, not only engineering.

9. What engineers and planners should assess



• LOLE, LOLH and EUE for each transmission option, using a multi-area probabilistic model.
• The shape of the RA value curve versus tie or interface size, not a single point.
• Load, weather and resource correlation between the regions being connected, including extreme-weather years.
• First-contingency transfer capability of each option from nodal studies, including voltage and stability limits.
• External assistance assumptions, derived rather than assumed.
• PRM and accreditation impacts, stated in the region's convention.
• Avoided capacity cost versus annualized transmission cost, alongside production cost and resilience benefits.
• NERC TPL-001-5.1 performance of the final combined portfolio.
• Cost allocation and incentive design that matches who pays with who benefits.
  

Part C — Three Anonymous Case Studies

These composite case studies are based on the kinds of planning problems Keentel Engineering sees on US projects. Names, locations and identifying details have been changed or combined.

Case study 1 — A utility IRP that assumed zero help from its neighbors

Situation. A vertically integrated utility in the Southeast with a 9 GW summer peak and a growing winter peak was preparing its integrated resource plan. Its RA study modeled the utility as an island with zero external assistance, producing a required installed reserve margin of about 22% and a need for roughly 1,200 MW of new combustion turbines within six years. The utility had two 500 kV ties and several 230 kV ties to neighboring systems.

Analysis. Keentel reviewed the RA model and the transmission data. The neighbors' peaks were moderately correlated with the utility's in summer but much less so in winter, when the utility's morning peak came before a neighboring system's. Keentel built a three-area probabilistic model using the same 35 weather years for all three areas, with thermal outage rates that rose in extreme cold and with assistance limited by first-contingency transfer capability computed from the current nodal planning case. Assistance was capped during the coldest historical days to reflect simultaneous scarcity across all three areas.

Engineered solution. With derived assistance limits, the required reserve margin fell to about 17%, and the new-capacity need fell by about 400 MW. Keentel also evaluated an upgrade of one 230 kV tie to 345 kV, which raised winter transfer capability by about 600 MW and reduced the need by a further 150 MW at a fraction of the cost of new generation.

Outcome. The utility filed its IRP with a reduced combustion turbine build, a transmission upgrade in its regional plan and a documented method for external assistance that its commission staff accepted. The capacity savings were roughly four times the annualized cost of the transmission upgrade.

Case study 2 — An RTO zone where the import limit was out of date

Situation. A load zone inside a Midwest RTO was facing a local capacity shortfall after the announced retirement of a 700 MW coal unit. The zone's capacity import limit used in the RA program came from a transmission study two years earlier, before several 345 kV upgrades were completed and before a large data center campus requested 600 MW of new load at the zone's edge.

Analysis. A group of load-serving entities asked Keentel to examine whether the import limit was still valid. Keentel rebuilt the transfer study in the current nodal case, scanning all single and selected multiple contingencies, and found two offsetting effects. The completed 345 kV upgrades raised the thermal import limit by about 450 MW. But the new data center load, located near the limiting interface, shifted flows and created a new post-contingency overload on a 138 kV path that the previous study had never seen. The net change was an increase of about 150 MW — far less than the 450 MW a simple update would have assumed.

Engineered solution. Keentel identified a low-cost 138 kV reconductoring and a series reactor on a parallel path that together relieved the new limit, raising the net import capability by about 500 MW. The load-serving entities submitted the projects through the RTO's planning process with a documented RA benefit, and coordinated the data center's interconnection study so that its load was modeled consistently in both the RA and transmission studies.

Outcome. The RTO's next RA cycle used the updated import limit. The zone's local capacity requirement was met without emergency retention of the retiring unit, and the reconductoring was in service before the retirement.

Case study 3 — A merchant HVDC developer that needed a capacity product

Situation. A developer proposed a ±525 kV HVDC link of about 2,000 MW between a wind-rich region and a load-heavy region in another RTO. Production cost studies showed attractive energy arbitrage, but the developer's financing depended on selling capacity value to load-serving entities in the importing region. The importing RTO's capacity program treated external resources conservatively and capped external assistance in its RA study.

Analysis. Keentel built a two-region probabilistic model with 30 aligned weather years, including the major winter storms of recent years. The analysis showed that the regions' wind patterns and peak loads were weakly correlated on most of the risky days, and that the link could deliver about 1,300 MW of equivalent firm capacity to the importing region at 0.1 day/yr, falling to about 900 MW in the extreme-weather scenario when the exporting region was also stressed. Keentel also documented the converters' availability, the firmness of the exporting region's surplus, and the deliverability of the imported power from the converter station into the importing region's load centers through nodal transfer analysis.

Engineered solution. Keentel prepared an RA value report that the developer used to negotiate firm capacity contracts with two load-serving entities in the importing region, backed by firm transmission service and resource commitments in the exporting region. The report also supported the developer's request that the RTO recognize the contracted imports as accredited capacity.



Outcome. The developer secured contracts for about 900 MW of firm capacity — the extreme-weather value rather than the average — which was enough to reach financial close. The RTO later opened a stakeholder process on accreditation of firm external capacity backed by dedicated transmission.

Part D — 20-Question Technical FAQ

Keentel Services, References and Contact

How Keentel Engineering Helps

Keentel Engineering supports utilities, RTO stakeholders, developers, large-load customers and state agencies at the point where resource adequacy, resource planning and transmission planning meet.

• Transfer capability studies. Nodal FCITC and import-limit analyses at 69 kV through 765 kV, with thermal, voltage and stability limits, to give RA and capacity expansion models defensible transfer numbers.
• Multi-area resource adequacy modeling. Probabilistic LOLE, LOLH and EUE studies with aligned weather years, correlated outages and derived external assistance.
• Transmission RA value and benefit-cost analysis. PRM reduction, equivalent firm capacity and avoided capacity cost analyses, structured to support Order No. 1920 long-term benefit measurement.
• IRP and planning support. Review of utility IRP and RTO planning assumptions for consistency between load, resource and transmission data.
• Interconnection and large-load studies. Generator interconnection studies under Order No. 2023 and large-load (data center) interconnection studies that feed consistent assumptions into planning models.
• HVDC and backbone transmission studies. Steady-state, stability and EMT studies for 500 kV and 765 kV AC and HVDC interregional options.
• NERC compliance. TPL-001-5.1 assessments, MOD-032 data and FAC-related support for Planning Coordinators and Transmission Planners.

Why Clients Choose Keentel

• We work across the silos. Our engineers understand nodal transmission studies, probabilistic RA models and capacity expansion, and we connect them.
• US standards fluency. NERC TPL and MOD standards; FERC Orders No. 1000, 1920 and 2023; RTO capacity constructs.
• Power system studies from 4 kV to 765 kV, including EMT modeling for HVDC and inverter-based resources.
• Transparent methods. Every assumption on assistance, accreditation and transfer capability is documented so regulators and stakeholders can follow it.
• Responsive senior engineers who stay with the project from scoping through regulatory filings.

Talk to Keentel Engineering

Is your resource plan built on transmission assumptions from the last planning cycle — or your transmission plan blind to the capacity it could save? Keentel Engineering can help you quantify the resource adequacy value of transmission and close the planning gaps.

• Phone: (813) 389-7871
• Email: contact@keentelengineering.com
• Book a 15-minute call: calendly.com/keentel-engineering/15min
• Web: keentelengineering.com

References

• FERC Order No. 1920, Building for the Future Through Electric Regional Transmission Planning and Cost Allocation, 2024; Orders No. 1920-A and 1920-B on rehearing.
• FERC Order No. 1000, Transmission Planning and Cost Allocation by Transmission Owning and Operating Public Utilities, 2011.
• FERC Order No. 2023, Improvements to Generator Interconnection Procedures and Agreements, 2023.
• NERC Reliability Standard TPL-001-5.1, Transmission System Planning Performance Requirements.
• NERC Reliability Standard MOD-032, Data for Power System Modeling and Analysis.
• NERC, Interregional Transfer Capability Study, 2024.
• NERC, Long-Term Reliability Assessment and Probabilistic Assessment reports (annual).
• NERC, Glossary of Terms Used in NERC Reliability Standards (Transfer Capability, First Contingency Incremental Transfer Capability).
• FERC, NERC and Regional Entity staff reports on the February 2021 cold weather event and the December 2022 winter storm.
• US Department of Energy, National Transmission Needs Study, 2023.
• IEEE Power & Energy Society, Resource Adequacy Working Group publications on probabilistic adequacy metrics.
• R. Billinton and R. N. Allan, Reliability Evaluation of Power Systems, 2nd ed., Plenum Press, 1996.
• A. J. Wood, B. F. Wollenberg and G. B. Sheblé, Power Generation, Operation, and Control, 3rd ed., Wiley, 2014.
• RTO resource adequacy and transmission planning manuals (MISO, PJM, SPP, ERCOT, CAISO) — current editions.

Disclaimer

This article is provided by Keentel Engineering for general technical education. It is not engineering, legal or regulatory advice for any specific project or filing. The models, fleets, costs and case studies are illustrative or composite, and results for real systems depend on actual data, models and regulatory rules. FERC orders, NERC standards and RTO rules change, and their implementation varies by region; always confirm current requirements with the applicable regulator, RTO and transmission provider. Keentel Engineering accepts no liability for decisions made on the basis of this article without a project-specific engineering review.

© 2026 Keentel Engineering. All rights reserved. Written by Keentel Engineering.