Cluster Interconnection Studies Under FERC Order No. 2023

Part A — Cluster Interconnection Studies at a Glance

For two decades, most US generator interconnection queues ran on a first-come, first-served serial basis: each request was studied in order of its queue position, and each study assumed that every request ahead of it would be built. That model broke under the weight of solar, wind and battery storage requests. Queues swelled to more than two terawatts nationally, studies ran years late, and speculative requests withdrew after triggering restudies for everyone behind them.

FERC Order No. 2023, issued in July 2023 and clarified in Order No. 2023-A, replaced serial studies with a first-ready, first-served cluster process. Requests that meet financial and site-control readiness tests are grouped into a cluster and studied together. Network upgrades are identified for the cluster as a whole and allocated among its members. Withdrawals carry real financial penalties. Transmission providers face penalties for late studies.



To move from the old queue to the new process, each transmission provider ran transition procedures: a transitional serial study for the most advanced requests and a transitional cluster study for eligible requests that were not yet far enough along. New England's first transitional cluster, completed in 2026, offers one of the clearest public views of what a cluster study produces — and what it costs.

The transitional cluster in numbers

3.4 Short-circuit analysis

Short-circuit studies identify breakers whose interrupting capability would be exceeded by the cluster's added fault current. Inverter-based resources contribute much less fault current than synchronous generators — typically 1.0 to 1.2 times rated current — but many large inverter plants clustered on one bus still add up.

3.5 Stability and EMT analysis

In the New England transitional cluster, steady-state, short-circuit, stability, EMT and torsional interaction screening were all performed. Detailed violation results are typically redacted as CEII in public reports, but the upgrade list reveals what was found: thermal overloads, voltage and stability problems requiring STATCOMs, and breaker or substation work.



3.6 Capacity deliverability testing

Capacity deliverability tests whether a resource's capacity can reach load during system peak, together with all other capacity resources, under stressed but credible conditions. The details are region-specific, but the core idea is common:

1. Dispatch existing and queued capacity resources to serve peak load plus reserves.
2. Increase output of the resources near the studied project (the "study area") while reducing output elsewhere, representing the worst credible local dispatch.
3. Apply contingencies and check thermal limits.
4. Identify upgrades needed for the project's capacity to be deliverable without bottling up other capacity resources.

Deliverability tests are often more restrictive than energy service tests, because they assume high simultaneous output in the study area. That is why the New England transitional cluster offered an automatic downgrade from capacity to energy service when capacity testing alone produced thermal violations.

In New England, capacity service enables participation in the Forward Capacity Market. The value of that participation — weighed against any additional deliverability upgrades — is a commercial decision that engineering results must inform.

3.7 Inverter-based resource performance in cluster studies

Storage-heavy and renewable-heavy clusters place inverter-based resource performance at the center of the study:

When many inverter-based resources cluster in a weak part of the network, EMT studies may reveal control interactions that positive-sequence studies cannot. Mitigations range from control tuning to grid-forming capability, synchronous condensers or STATCOMs — the latter appearing among the New England cluster's network upgrades.

4. Worked examples

4.1 Distribution factors and thermal overloads



The impact of a project on a monitored line is quantified by its distribution factor (DFAX) — the fraction of the project's injection that flows on the line for a given contingency:

Consider a 115 kV line with a long-term emergency rating of 240 MVA and a pre-cluster post-contingency flow of 200 MVA. Three cluster projects affect it:

Post-contingency flow rises to 285 MVA, or 118.75% of the long-term emergency rating. No single project would overload the line: A alone reaches 250 MVA (104%), B alone 230 MVA (96%) and C alone 205 MVA (85%). The overload is a cluster effect — exactly the condition the cluster process is designed to capture and allocate fairly.

4.2 Allocating the upgrade

Suppose reconductoring the line costs $40 million. Three allocation methods give very different answers:

The proportional impact method assigns cost according to each project's contribution to the need, which is why Order No. 2023 adopted it for system network upgrades. Many regions also apply a minimum distribution factor threshold (often around 3% to 5%); a project below the threshold may be excluded from both the upgrade's cost and the violation list.

The general expression for proportional impact allocation is:

4.3 Study cost allocation

Study costs are allocated differently from upgrade costs. In the New England transitional cluster, study and restudy costs are allocated 50% per capita and 50% per MW. Suppose the cluster's total study cost were $6 million across 23 projects and 4,826.5 MW. A 200 MW project's share would be:

Order No. 2023 lets transmission providers choose a per capita share between 10% and 50%, with the rest allocated per MW. A higher per capita share burdens small projects relatively more; a higher per MW share burdens large projects.

4.4 Withdrawal penalties



Using the same $255,000 study cost share:

The $178 million figure is the average network upgrade cost for the 18 projects studied for energy service for the first time in the New England transitional cluster (average size about 213 MW, or about $836/kW). Under the pro forma rules, a withdrawal penalty applies only when the withdrawal materially affects other customers' costs or timing, and penalties are waived when network upgrade costs rise sharply after a study report (more than 25% after a cluster restudy, or more than 100% after the facilities study). Funds are first used to cover restudy costs and then to offset upgrade cost increases for remaining customers.

4.5 Interconnection agreement deposits

At agreement execution, the New England transitional cluster required a deposit of 20% of identified network upgrade costs. For the highest-cost project in the cluster, about $541 million, that is about $108 million — due within days of agreement execution. For the average project with first-time energy service studies, it is about $35.6 million. Developers must plan financing for these deposits well before the study report is issued.

4.6 Short-circuit impact of inverter clusters

A 300 MW battery project contributes roughly:

at 115 kV, assuming 1.2 per unit fault current on a 300 MVA base. If a 115 kV bus already has 38 kA of fault duty and 40 kA breakers, one such project raises duty to about 39.8 kA (99.5% of rating), and a second nearby project would exceed it. Breaker replacements, current-limiting reactors or bus splitting then become network upgrades.



4.7 STATCOM sizing for voltage support

The voltage change at a bus produced by a reactive power injection is approximately:

A 400 Mvar STATCOM at a 345 kV bus with 10,000 MVA of short-circuit capacity raises voltage by about 4% — a large and fast correction for post-contingency voltage collapse or weak-grid stability problems. At a 115 kV distribution-adjacent bus with 1,500 MVA, a 66 Mvar device produces a similar 4.4% change. Reactive devices of 400 Mvar, 250 Mvar and 66 Mvar appear among the New England transitional cluster upgrades, along with containerized 8 Mvar distribution-voltage STATCOMs where transmission-voltage devices would have been oversized and physically constrained.



4.8 Reconductoring with high-temperature, low-sag conductor

Aluminum conductor steel-supported (ACSS) conductor can operate continuously at much higher temperatures than conventional ACSR, increasing ampacity without heavier structures. A 345 kV line rebuilt with double-bundled 1590 kcmil ACSS and a summer long-term emergency rating of 2,557 MVA carries:

or about 2,140 A per subconductor — within the capability of 1590 kcmil ACSS at elevated temperature. Because ACSS weighs about the same as ACSR of similar size, it can often be installed on existing structures. When structures are aged wood H-frames, however, a full rebuild may be unavoidable regardless of conductor choice. NESC clearance requirements at maximum sag must be checked for any higher-temperature operation.

4.9 Worked example: commercial readiness deposits through the process

For a 200 MW regular-cluster request with an eventual network upgrade allocation of $178 million, the pro forma Order No. 2023 deposit path is approximately:

The capital at risk rises sharply as the study progresses, which is the intent: a developer that is not ready to build is pushed to withdraw early, when the impact on others is smallest. Individual transmission providers may adopt variations accepted by FERC, so the applicable tariff governs.

4.10 Worked example: what an upgrade allocation means for project economics

A network upgrade allocation is a capital cost. Annualized with a capital recovery factor:

at an 8% cost of capital over 20 years, CRF ≈ 0.102. The annual cost per kW-year of different allocations is:

For a battery project, these annualized costs must be recovered from capacity revenues, energy arbitrage, ancillary services and any contracted payments. An allocation of $1,500/kW or more can exceed the entire annual revenue a storage project can expect from capacity markets in many years. Some network upgrade costs may be eligible for transmission credits or other recovery mechanisms depending on the region's tariff, which changes the calculation; developers should model both cases.

4.11 Worked example: study delay penalties

Order No. 2023 imposes penalties on transmission providers for late studies — for example, $1,000 per business day for a late cluster study, $2,000 for a late cluster restudy or affected system study, and $2,500 for a late facilities study, each subject to caps tied to study deposits. A cluster study 40 business days late would accrue $40,000 in penalties, distributed among customers that remained in the cluster. The amounts are modest relative to project costs; their purpose is accountability and transparency, with transmission providers able to seek relief for delays outside their control.

5. Keentel's analysis of the cluster results



5.1 Cost concentration

Project-level costs are dominated by shared system network upgrades. Five of the seven most expensive projects per kW connected at 115 kV, and four of those sat in one western area where cumulative battery injections overloaded the same corridors.



5.2 Cost by point-of-interconnection voltage

The 115 kV projects carried nearly twice the per-kW cost of the 345 kV projects. Lower-voltage networks have lower thermal capacity per circuit and are more easily overloaded by clustered injections; their upgrades are also less likely to be shared with large projects at higher voltages.



5.3 Upgrade types

5.4 Schedule

Construction time estimates ranged from 6 to 93 months from notice to proceed. Upgrades longer than five years usually involve multiple line rebuilds, siting and permitting, and outage coordination on heavily used corridors. For battery storage projects that can be built in 12 to 24 months, network upgrade schedules — not project construction — set the commercial operation date. Limited operation studies, interim deliverability and phased interconnection can sometimes bridge the gap.

6. Alternative transmission technologies

Order No. 2023 requires transmission providers to evaluate alternative transmission technologies when identifying network upgrades. A typical set of guidelines maps each technology to the violation types it may address:

Keentel's observations from the New England results:

• ACSS is now standard. Transmission owners treated high-temperature, low-sag conductor as their default reconductoring choice, with composite-core conductors reserved for long spans, river crossings or structures with spare strength and clearance.
• Power flow controllers were considered but often more expensive than reconductoring for single thermal violations.
• Transmission switching was rejected where opening circuits would weaken reliability, especially on critical interfaces.
• Modular STATCOMs created value: containerized distribution-voltage units kept two projects viable at lower cost than transmission-voltage devices.
• Dynamic line ratings (not in the list above, but required for consideration under FERC's related transmission rules) may offer further near-term capacity on wind-cooled lines.

7. Contingent facilities, affected systems and restudies

7.1 Contingent facilities

Contingent facilities are unbuilt network upgrades or higher-queued projects on which a project's interconnection depends. If a contingent facility is delayed or cancelled, the project's schedule and costs may change. Interconnection agreements must identify them, and developers should track their status as closely as their own construction.

7.2 Affected systems

When a cluster affects a neighboring transmission system — another utility's lower-voltage system within the region, or a neighboring region — an affected system study determines whether additional upgrades are required. Order No. 2023 standardized affected system study timelines and cost allocation. In New England, projects requiring affected system studies could request an extended agreement execution deadline of 30 days after receipt of the affected system study report.

7.3 Restudies

The New England transitional cluster had no formal restudy window, but restudies may occur if withdrawals of transitional cluster projects or higher-queued requests change results. The extent depends on where and how much capacity withdraws. Restudy costs are allocated 50% per capita and 50% per MW among affected projects, and are offset first by forfeited deposits from withdrawn requests. Restudies can run in parallel with the first regular cluster study.

7.4 Withdrawal cascades

Because shared upgrades are allocated across the cluster, a withdrawal can raise costs for those who remain:

In the example of Section 4.2, if Project A withdraws and the $40 million upgrade is still needed (flow without A is 235 MVA, just under the 240 MVA rating, so it would not be), the upgrade could disappear entirely. If instead the base-case flow were 220 MVA, B and C alone would produce 255 MVA, the upgrade would remain, and B's share would rise from $14.1 million to $34.3 million. Withdrawal penalties exist precisely to fund cases like this.

8. Large loads and the interconnection queue

Data centers and other large loads are not subject to generator interconnection procedures, but they compete for the same transmission capacity. A cluster study's base case includes forecast load; a large load that appears after the base case is set can change flows, create new violations and require restudies. Increasingly, regions and utilities are developing large-load interconnection processes, co-location rules and surplus interconnection options for load-plus-generation sites. Developers of data center campuses with on-site generation or storage must understand both the load and generator interconnection processes, because the storage or generation portion may enter a generator cluster.

8A. How cluster processes compare across US regions

Every FERC-jurisdictional transmission provider now runs some form of cluster study, but the details differ. Engineers working in several regions must keep the differences straight.

The common lesson: interconnection service definitions, deposit schedules, withdrawal penalties and allocation methods are tariff-specific. A strategy that works in one region can fail in another, and published summaries often compress these differences. Engineering teams should always work from the current tariff and business practice manuals, and confirm interpretations with the transmission provider during scoping.

9. A practical developer strategy for the cluster era

9.1 Before the cluster window

1. Obtain CEII access and planning models. Use the transmission provider's base cases, not public approximations.
2. Screen points of interconnection. Run distribution factor analysis on candidate POIs against known constraints; avoid corridors already loaded by queued projects.
3. Consider higher-voltage POIs. The New England results show 345 kV POIs averaging about half the per-kW cost of 115 kV POIs.
4. Right-size the request. Model multiple sizes; a smaller request may avoid an expensive upgrade threshold.
5. Choose the service type deliberately. Energy-only service, capacity service, or capacity with an automatic downgrade option.
6. Secure site control and financing for deposits, including the 20% network upgrade deposit at agreement execution.
7. Prepare accurate models. Validated PSS®E and PSCAD models of the actual inverters, collector system and plant controller.

9.2 During the cluster study

1. Attend the scoping meeting and confirm modeling assumptions for the project.
2. Track other cluster members and their likely withdrawals.
3. Review interim results critically: check model data, ratings, contingency definitions and upgrade scope.
4. Comment within the comment window with specific engineering evidence.
5. Evaluate alternatives: POI changes allowed under material modification rules, size reductions, service downgrades, surplus interconnection.

9.3 After the study

1. Negotiate the interconnection agreement within the 60-day window.
2. Post deposits on time to avoid deemed withdrawal.
3. Track contingent facilities and network upgrade schedules.
4. Submit as-purchased and as-built data, keeping the project's equipment rated for the duties assumed in the study.

9.4 Worked example: screening two POIs

A developer considers a 200 MW battery at either of two POIs:

The 345 kV POI has higher direct costs but much lower network upgrade risk. Screening with actual base cases, rather than proximity alone, changes the decision.

9.5 Surplus interconnection and material modifications

Two provisions give developers flexibility outside the full cluster process:

A storage addition at an existing solar or wind site via surplus interconnection can avoid the cluster process entirely, provided total injection stays within the existing interconnection service. For many developers, this is the fastest way to add storage in congested regions.

10. Engineering quality in cluster studies

10.1 Model accuracy

Cluster results are only as good as the models submitted. Common problems include:

• generic inverter models that do not match the selected equipment;
• incorrect reactive capability curves;
• plant controller settings inconsistent with the interconnection requirements;
• collector system impedances that differ from the design;
• PSCAD models that fail quality checks or do not initialize.

NERC reliability guidelines on inverter-based resource modeling and IEEE 2800-2022 performance requirements set expectations for model fidelity and plant performance. Transmission providers increasingly reject or delay requests with poor models.

10.2 Ratings

Thermal violations depend on facility ratings. Transmission owner rating methodologies, governed by NERC FAC-008, determine normal, long-term emergency and short-term emergency ratings. A rating error — such as an outdated limiting element — can create or hide a violation. Developers and their engineers should check that limiting elements and ratings in the study match current transmission owner data.

10.3 Contingency definitions

N-1 and N-1-1 contingencies, breaker-failure and stuck-breaker events, and common-tower contingencies must match the transmission owner's protection and substation configuration. An incorrect contingency definition can drive an expensive upgrade.

10.4 Battery charging

Storage-dominated clusters must be tested in charging mode at light load. A battery charging at full power is a large load on the system; clustered charging can overload facilities in the opposite direction from discharge. Operating restrictions on charging can sometimes avoid upgrades.

10.5 Reviewing a cluster study report: Keentel's checklist

Public cluster reports redact violation details, short-circuit results, stability results, cost allocation details and contingent facilities as CEII. Interconnection customers receive the full report. Keentel's review of a cluster report for a client follows a structured checklist:

Questions should be raised during the comment period with specific engineering evidence: corrected ratings, revised models, alternative upgrade proposals or sensitivity results. Transmission providers are far more responsive to precise technical comments than to general objections about cost.

10.6 Interim versus final reports

The interim report is the main opportunity to influence results. Between interim and final reports in the New England transitional cluster, the process allowed comments, refinement of upgrade scopes (for example, adjusting distribution-voltage STATCOM costs from 10 Mvar to the required 8 Mvar rating) and correction of data. After the final report, changes generally come only through restudies triggered by withdrawals, or through detailed design during the interconnection agreement phase.

11. Common misconceptions corrected

12. Key takeaways

1. First-ready, first-served replaced first-come, first-served. Readiness, not queue date, now drives priority.
2. Clusters reveal cumulative impacts. Upgrades no single project would trigger are identified and shared.
3. Location and POI voltage dominate cost. Screen POIs with actual base cases.
4. Know the allocation rules. Study costs, substation upgrades and system upgrades are allocated differently.
5. Price withdrawal risk. Transitional penalties of nine times study costs, and regular penalties up to 20% of upgrade costs, change the economics of speculative requests.
6. Plan for long upgrade schedules and large deposits.
7. Model quality is a commercial issue. Accurate models reduce delays and avoid unnecessary upgrades.

13. How Keentel Engineering helps

Keentel Engineering supports developers, independent power producers, utilities and large-load customers through every stage of the interconnection process, from 4 kV to 765 kV.

Why clients choose Keentel Engineering

• Interconnection expertise across regions: ISO-NE, PJM, MISO, SPP, ERCOT, CAISO, NYISO and non-RTO utilities.
• Study depth: steady-state, short-circuit, stability and EMT under one team.
• Commercial understanding: engineering results translated into cost, schedule and risk decisions.
• Practical design experience: substations, transmission lines and inverter-based plants.
• Licensed professional engineering: a Florida-registered firm with offices in Tampa, Austin, Sacramento and Baltimore, supporting projects nationwide.

Preparing for a cluster request window or reviewing cluster study results? Talk to Keentel Engineering at (813) 389-7871, contact@keentelengineering.com, or book a 15-minute scoping call at calendly.com/keentel-engineering/15min.

Part C — Technical FAQ: Cluster Interconnection Studies

References and Further Reading

Links were current at publication (October 2026).



• FERC Order No. 2023, Improvements to Generator Interconnection Procedures and Agreements, 184 FERC ¶ 61,054 (2023), and Order No. 2023-A, 186 FERC ¶ 61,199 (2024). Federal Energy Regulatory Commission.
• FERC pro forma Large Generator Interconnection Procedures (as revised by Order No. 2023). Federal Energy Regulatory Commission.
• ISO New England, Interconnection Service Study Planning Models (iso-ne.com, System Planning section). ISO New England Inc.
• ISO New England Open Access Transmission Tariff, Schedules 22, 23 and 25, and Planning Procedure PP5-6. ISO New England Inc.
• ISO New England public materials on the transitional cluster study and next steps (2025–2026). ISO New England Inc.
• NERC Reliability Standards FAC-008, TPL-001, MOD-032 and MOD-026/027; NERC reliability guidelines on inverter-based resource modeling. North American Electric Reliability Corporation.
• IEEE Std 2800-2022, Interconnection and Interoperability of Inverter-Based Resources Interconnecting with Associated Transmission Electric Power Systems. IEEE.
• IEEE Std 738, Calculating the Current-Temperature Relationship of Bare Overhead Conductors. IEEE.
• National Electrical Safety Code (NESC), IEEE C2. IEEE.
• Lawrence Berkeley National Laboratory, Queued Up: Characteristics of Power Plants Seeking Transmission Interconnection (annual editions).
  

Disclaimer

This article is general technical information for educational purposes. It is not engineering, legal or financial advice for any specific project and does not create a professional relationship. Figures summarized here are drawn from public, planning-grade, non-binding estimates and may change through restudy, detailed design and construction; Keentel's aggregations and per-kW calculations are its own analysis, and projects are anonymized. Tariff provisions vary by transmission provider and change over time; readers must consult the current FERC-accepted tariff, transmission provider procedures and qualified legal counsel. Worked examples use illustrative values. Engineering work must follow applicable IEEE standards, the NESC, NERC requirements and transmission provider criteria, sealed by a licensed professional engineer where required.

FERC, NERC, IEEE, ISO New England and other names are the property of their respective owners. Keentel Engineering is not affiliated with or endorsed by these organizations or any software vendor.