NERC MOD-026-2 consolidates the old MOD-026-1 and MOD-027-1 model verification standards. Owners provide dynamic models, including electromagnetic transient models for inverter-based resources, and show that the models match measured plant behavior. MOD-027-1 retires when MOD-026-2 takes effect, so new verification work is planned to MOD-026-2.
Short answer
NERC MOD-026 and MOD-027 used to be two separate standards of the North American Electric Reliability Corporation (NERC). MOD-026-1 covered excitation and voltage control models, and MOD-027-1 covered turbine-governor and frequency response models. The Federal Energy Regulatory Commission (FERC) approved MOD-026-2 on 19 February 2026. Both older standards retire immediately before MOD-026-2 takes effect, and until then they stay in force. MOD-026-2 then carries the whole verification job.
New verification work for MOD-027 topics is therefore planned to the combined standard. The wider framework is in the US electrical codes and grid interconnection guide.
What changed
The old pair focused on synchronous generators. The new standard is broader and was part of NERC’s response to FERC Order 901 on inverter-based resources:
- It includes registered inverter-based resources (IBR), flexible AC transmission system (FACTS) devices and high-voltage direct current (HVDC) systems, along with generating facilities.
- Generator Owners (GO) and Transmission Owners (TO) provide positive-sequence dynamic models. Under R3 they also provide electromagnetic transient (EMT) models for IBR, FACTS and HVDC.
- Legacy facilities, with a commercial operation date before MOD-026-2 takes effect, are excluded from R3 where the manufacturer no longer supports an EMT model. The owner documents that lack of support.
- It covers verification and validation, with new glossary definitions for each.
- It works together with MOD-032 and MOD-033, covered in our MOD-032 and MOD-033 guide.
The implementation plan phases the requirements. Read NERC’s standard page and the plan for the dates that apply to your facility.
What the owner provides
The Transmission Planner (TP) needs models it can run in its study software. For each facility the owner typically supplies:
- The positive-sequence dynamic model with parameters and the software version it was prepared for.
- The EMT model, where the standard asks for it.
- Documentation of how the parameters were obtained.
- Evidence that the simulated response matches the measured one.
The evidence comes from tests. For a generator, that means step tests on the voltage regulator and on the governor. For an inverter plant, it means measured responses at the plant controller. The guide on automatic voltage regulator tests shows how the regulator data is gathered, and the inertia constant and generator models guide explains the model parameters behind it.
How to compare model and measurement
A good comparison uses the same disturbance in both worlds. The test injects a step in voltage reference or power reference, the recorder captures the plant response, and the simulation is run with the same step. The two curves are compared on peak, rise time, settling and damping. Where the model differs, the parameters are adjusted, and the adjusted model is documented.
Acceptance criteria are set by the standard and by the planner. Do not assume a generic percentage.
Worked example (illustrative)
The data and the band below are illustrative assumptions, not acceptance criteria from the standard.
Assumptions. A plant controller receives a 10 MW power step. Six samples of the measured and simulated active power rise are compared. The tolerance assumed for the exercise is 5 percent root-mean-square error and 10 percent maximum error, both relative to the 10 MW step.
| Sample | Measured MW | Simulated MW | Error MW |
|---|---|---|---|
| 1 | 0.0 | 0.0 | 0.0 |
| 2 | 3.2 | 3.9 | 0.7 |
| 3 | 6.1 | 6.8 | 0.7 |
| 4 | 8.4 | 8.9 | 0.5 |
| 5 | 9.6 | 9.8 | 0.2 |
| 6 | 10.0 | 10.0 | 0.0 |
Step 1. Square the errors. 0 + 0.49 + 0.49 + 0.25 + 0.04 + 0 = 1.27.
Step 2. Mean and root. 1.27 ÷ 6 = 0.2117, and the square root is 0.46 MW.
Step 3. Relative RMS error. 0.46 ÷ 10 = 4.6 percent, which passes the assumed 5 percent band.
Step 4. Maximum error. 0.7 ÷ 10 = 7 percent, which passes the 10 percent band.
Step 5. Look at the shape. The simulation is faster than the plant in samples 2 to 4. That may point to a delay in the real controller that the model leaves out. A pass on numbers still deserves a note to check that delay.
Practical points for owners
Ask for the models early. Manufacturers need time to prepare an EMT model, and the model has to match the firmware that will be installed. Ask for the model and its software version at the equipment purchase stage, not at commissioning.
Plan the tests with the models in mind. A step test that was designed only to check the controller may not give the planner the data needed to validate the model. Decide with the planner which disturbances will be used for comparison, and design the recorders to capture them.
Keep model and plant in step. A model is a snapshot. After a firmware change, a controller replacement or a setting change that alters the response, the model needs an update. A model that no longer matches the plant fails validation, and an out-of-date model is worse than a simple but correct one.
Know what the models are for. The positive-sequence model is used for wide-area studies, and the EMT model is used for fast interactions at specific locations. Each has its own limits. A model that works in one tool may not run in another, so confirm the software and version with the planner.
Legacy facilities. The standard handles older facilities differently, especially where the manufacturer no longer supports an EMT model. If your plant is old, ask the planner whether it is identified as legacy before spending effort on a model that is not required.
For the machine parameters behind these models, see the guide on synchronous generator inertia and models.
Checklist for the engineer
- Confirm which facilities fall under MOD-026-2 and which are legacy.
- Read the current standard text and implementation plan on NERC’s site.
- Ask the manufacturer for the models, the software versions and their limits.
- Plan the field tests that produce validation data before commissioning ends.
- Set up recorders with enough resolution for the response you will compare.
- Compare model and measurement on the same disturbance.
- Document every parameter you tuned and why.
- Update the models after control or firmware changes.
Who signs and seals
In the United States, study and model packages are normally reviewed and sealed by a professional engineer licensed in the state of the project. Our work is delivered ready for your engineer of record to review and stamp. Tiza does not sign, seal or certify work in the United States. Compliance with a NERC standard stays with the registered owner; our studies support its analysis and evidence.
How Tiza can help
If your plant needs dynamic studies and model documentation for the Transmission Planner, Tiza provides power-system studies in English, ready for your engineer of record to review and stamp. Contact us with the plant data and the planner’s model requirements.
Figure: flow from manufacturer model to plant model, field test, comparison and updated model.
Frequently asked questions
Is MOD-027 still a current standard?
Only until MOD-026-2 takes effect. FERC approved MOD-026-2 in February 2026, and MOD-026-1 and MOD-027-1 retire immediately before that date. MOD-026-2 consolidates both, so new verification work is planned to it.
What is the difference between verification and validation?
Verification checks that the model is a correct representation of the plant and its controls. Validation compares the model's simulated response with measured behavior. FERC's approval added glossary definitions for both terms.
Who gives the models to the Transmission Planner?
The Generator Owner or Transmission Owner. For registered inverter-based resources, FACTS devices and HVDC systems, this includes electromagnetic transient models, with limited exceptions for legacy facilities.