R0, R1 and R2 are the three maturity stages of the models and data AEMO and the network service provider (NSP) use to assess a generating system. R0 is design-stage data at connection application, R1 models represent the equipment actually procured and support the performance standards assessment before connection, and R2 models are validated against tests on site after commissioning.
Each stage is a gate. Late or inconsistent R1 models stall the performance standards assessment, and failed R2 validation can hold the plant below full output until the models are fixed.
AEMO's current Power System Model Guidelines and the NSP's requirements define the detail, including accuracy criteria. Check the version that applies to your project.
R0, R1 and R2 in One Table
R1 and R2 come from the registered data categories in NER Schedule 5.5; R0 is the industry's shorthand for preliminary data before them. In practice, the industry uses all three labels as shorthand for the three model and data packages a project produces as it moves from application to operation.
| Stage | When it is used | Where the data comes from | Typical deliverables |
|---|---|---|---|
| R0 | Connection enquiry and application | Preliminary design, typical OEM data, sometimes generic models | Single line diagram, indicative plant parameters, preliminary PSS/E and PSCAD models, early study results |
| R1 | Performance standards assessment, before connection and registration | Contracted equipment: OEM inverter models, final plant controller design, design calculations, factory test results | Site-specific PSS/E and PSCAD models, releasable user guide, benchmarking report, data sheets, GPS compliance studies |
| R2 | After commissioning, once site tests are complete | Measurements from commissioning and hold point tests on the installed plant | Validation report comparing simulation with site recordings, updated models and parameters, confirmed settings |
Do not treat R0 as disposable. It drives the early studies and the access standards you negotiate, and a materially different R1, for example after an inverter change, means redoing some of that work.
What an R1 Submission Contains
The R1 package is where most of the modelling effort sits. It must describe the plant you will build well enough that AEMO and the NSP can run their own studies and reach your answers.
R1 Submission Contents
- PSS/E model: site-specific positive-sequence dynamic model of the whole generating system, in the software version the NSP and AEMO specify
- PSCAD model: EMT model with the OEM's detailed inverter representation and the plant controller, set up to run at the specified time step
- Model source code access: an arrangement for AEMO and the NSP to see or receive the source code, or an agreed alternative such as escrow
- Benchmarking report: PSS/E and PSCAD responses compared for the same set of disturbances, with differences explained
- Releasable user guide: a model description that can be shared with other parties, covering structure, parameters, initialisation and limits
- Plant controller detail: the actual control structure, gains, deadbands, ramp rates and coordination with inverter-level controls
- Network and collector data: transformers, cables, reactive plant and auxiliary loads, matching the issued-for-construction design
- Compliance studies: the GPS assessment showing each access standard is met, using the R1 models
Models of the real equipment
R1 models must represent the specific inverter, firmware and plant controller on order. A generic model, or one from the same OEM's earlier product line, is the most common reason an R1 package comes back. The firmware matters because inverter control behaviour changes between releases, and the model has to match what is installed.
Source code access and the releasable user guide
AEMO and the NSP need to understand what the model does, not just run it. That is why the guidelines expect access to source code under agreed confidentiality terms, and a user guide that can be released to other parties who study the same part of the network. Agree the access arrangement with the OEM at procurement. It is much harder to negotiate after the contract is signed.
Benchmarking between PSS/E and PSCAD
The two models describe the same plant in different tools. The benchmarking report shows they respond consistently to a defined set of tests such as voltage dips, setpoint steps and frequency disturbances. Where they differ, the report has to explain why, for example a sub-cycle effect that only the EMT model captures. Our overview of PSS/E and PSCAD modelling explains why AEMO asks for both.
R2: Validating the Models Against Site Tests
R2 closes the loop. Once the plant is built, the tests run during commissioning are recorded, and the same events are replayed through the models. If the simulated response matches the measured one within the accuracy expected by the guidelines, the models are accepted as representing the plant.
Tests used for validation
The validation tests are normally built into the commissioning program, with tests at each staged hold point. Our guide to AEMO hold point testing explains how that staged release works. Typical tests include:
- Voltage and reactive power setpoint steps, to check the plant controller's response time, settling and overshoot
- Active power setpoint steps and ramp rate tests
- Power factor and reactive power mode changes
- Frequency response tests, often by injecting a simulated frequency signal into the plant controller
- Tests at different output levels, since the plant can behave differently at low and high output
Fault ride-through is not normally tested by applying a real fault to the network. It is usually covered by type testing and by checks that the installed protection and control settings match those in the model.
Accuracy expectations
AEMO's model guidelines define how closely simulated and measured responses must match, using quantities such as rise time, settling time and the error between the two traces over defined windows. The exact criteria depend on the guideline version that applies to your project. Read them from that document rather than relying on figures quoted from an earlier project.
High-speed recorders must be installed, time-synchronised and capturing the right signals before testing starts. A test that was not recorded properly usually has to be repeated.
Re-tuning
Where the model and the plant disagree, one of them has to change. Sometimes the model is corrected to reflect the as-built plant, for example a measured transformer impedance or a controller delay. Sometimes the plant settings are adjusted and the test is repeated. Either way, any change to the model means re-checking the benchmarking, and a change to the plant may also mean re-checking GPS compliance.
The fastest R2 outcomes usually come from preparing simulated responses for every planned test before the testing window opens. The comparison can then be made on site, the same day, rather than weeks later.
Common Reasons Submissions Get Rejected
Most rejections come from inconsistencies a careful reviewer finds quickly, not deep modelling errors.
| Problem | What it looks like | How to avoid it |
|---|---|---|
| Version mismatch | Model built in a different PSS/E or PSCAD version, or for different inverter firmware, from what was specified or installed | Fix software versions and firmware at the start, and record them in the user guide |
| PSS/E and PSCAD inconsistency | The two models give materially different responses to the same disturbance, with no explanation | Benchmark early and explain every material difference in the report |
| Missing plant controller detail | Plant controller shown as a simplified block, or gains and limits that do not match the delivered controller | Get the controller model and settings from the supplier with the inverter models |
| Design drift | Collector, transformer or reactive plant data in the model no longer matches the latest design | Freeze model inputs against a controlled design revision |
Most trace back to a model updated in one place and not the others. One configuration record covering firmware, software versions, settings and design revision prevents most of them. Our article on common GPS application pitfalls covers the wider application process.
Who Does What
Model submissions involve four parties, and delays often happen in the handovers between them.
| Party | Role in R0, R1 and R2 |
|---|---|
| Inverter and plant controller OEM | Supplies the inverter and controller models, source code access, type test results and firmware details. Supports re-tuning if R2 validation fails |
| Consultant (connection applicant's engineer) | Builds the site-specific PSS/E and PSCAD models, runs benchmarking and GPS studies, writes the user guide, prepares test procedures and the R2 validation report |
| Network service provider | Reviews models and studies for its network, runs its own system studies, and witnesses or reviews commissioning tests |
| AEMO | Reviews models against its guidelines, assesses performance standards, approves staged release through commissioning, and accepts the validated models |
The developer sits above all four. The OEM's modelling obligations need to be in the supply contract, with deliverable dates tied to the R1 program. If they are not, the consultant cannot finish the R1 package on time, whatever the effort.
If you are earlier in the process, our guides to generator performance standards and steady-state versus dynamic studies explain what the models are ultimately used to prove.
Get Your Models Submission-Ready
ElectraGlobe builds and benchmarks PSS/E and PSCAD models for solar, wind and BESS projects, prepares R1 submissions, and supports R2 validation through commissioning. If you have an R1 package in preparation, a set of AEMO comments to close out, or test results that do not match the model, our power system studies team can tell you quickly where the gaps are. Get in touch to discuss your project.
Frequently Asked Questions
What is R2 model validation?
R2 validation compares the plant's measured response during commissioning tests with the response the models predict for the same events. If the two match within the accuracy expected by AEMO's model guidelines, the models are accepted as representing the installed plant. The validated models are then used for future system studies involving the plant.
When are R1 models submitted?
R1 models are submitted once the main equipment is contracted and the design is firm enough to model the plant that will actually be built. They support the performance standards assessment before connection and registration. In practice, the timing is driven by when the OEM can deliver site-specific inverter and plant controller models.
Who provides the inverter models?
The inverter manufacturer provides the inverter models, and usually the plant controller models if it supplies the controller. The applicant's consultant then integrates them into site-specific PSS/E and PSCAD models of the whole generating system. Model delivery and source code access should be written into the supply contract.
What happens if R2 validation fails?
The cause of the mismatch is investigated, and either the model is corrected to reflect the as-built plant or the plant settings are adjusted and the test repeated. Until the models are accepted, progress through the commissioning hold points can be delayed. Any change may also require benchmarking and parts of the GPS assessment to be re-checked.