05/06/2026
A power-system model can run without errors and still be wrong. This is one of the most important lessons in modern power-system engineering. When engineers study a new solar plant, wind farm or battery energy storage system, the simulation may show that everything is stable. The plant may appear to remain connected during faults, support voltage and respond correctly to changes in frequency. But a successful simulation does not automatically mean that the real plant will behave in the same way. The model must be checked against the actual equipment installed at the site. This process is called model validation. For Uzbekistan, this subject is becoming increasingly important. The country is rapidly connecting utility-scale solar, wind and battery-storage projects. Recent projects also combine renewable generation with BESS, making plant behaviour more dependent on inverter controls and software settings. Model validation should therefore become a normal part of grid connection, commissioning and long-term operation.
What does model validation actually mean?
In simple terms, model validation answers one question:
Does the simulation model behave like the real power plant?
Imagine that a fault causes the grid voltage to fall. The simulation may show that a solar plant remains connected, provides reactive current and restores active power smoothly after the fault. However, the real plant may behave differently. Its inverters may reach their current limits. Protection functions may operate. Active power may recover more slowly. The plant power controller may react with a delay. Some inverter blocks may temporarily reduce their output. If the simulated response and the measured plant response are different, the model needs to be corrected. This is why there is an important difference between verification and validation. Verification checks whether the model is complete, correctly built and able to run in the required software. Validation checks whether the model represents the real equipment. Both are necessary. A model that runs successfully but does not match field performance can give engineers false confidence. International grid operators already use commissioning tests and real disturbance records to compare simulated and measured responses. AEMO, for example, requires finalised simulation models and validation reports based on commissioning or network-event data. NERC guidance also recommends using actual grid disturbances and measured plant data to verify inverter-based resource models.
Why are solar, wind and BESS models different?
Traditional power systems were mainly based on synchronous generators. Their behaviour is influenced by physical equipment such as rotating mass, excitation systems and turbine governors. Engineers have many years of experience modelling these machines. Solar PV, modern wind turbines and BESS are different. They are connected to the grid through power-electronic converters. Their behaviour is mainly determined by control logic and software settings. During a disturbance, their response can depend on:
This means that two plants with the same MW rating may respond very differently during the same grid event. Even two plants using the same inverter manufacturer may behave differently because their controller settings, transformer arrangement, protection logic or software versions are not the same. NERC’s model-verification guidance specifically recognises that inverter-based resources need suitable testing, measured data and disturbance-based verification to confirm that their dynamic models represent actual plant behaviour.
What is the risk of using an incorrect model?
The most dangerous model is not always the model that fails. An obvious error is usually found and corrected. The greater risk is a model that looks professional, runs without errors and produces believable graphs, but does not correctly represent the real plant. Such a model can lead to incorrect conclusions about:
For example, a study may show that a plant provides strong reactive support during a voltage dip. But the real inverter may prioritise active current or reach its total current limit earlier than the model. The simulation would then overestimate the plant’s ability to support the grid. This is not simply a modelling issue. It can affect connection decisions, operational limits, protection settings and system security. NERC’s MOD-026 standard was developed specifically to verify that generator excitation or plant volt/var control models accurately represent real equipment response for system studies.
How should validation be carried out?
The process can be practical and straightforward. Before commissioning, the developer should provide the required steady-state, short-circuit and dynamic models. An EMT model should also be provided when the connection is weak or when fast inverter-control interactions need to be studied. During commissioning, actual plant signals should be recorded. These normally include voltage, current, active power, reactive power, frequency, controller references, equipment status and protection events. The plant should then be tested in its real operating modes. Typical tests may include voltage control, reactive-power response, active-power ramping, frequency response and BESS charging or discharging. The same test should be reproduced in the simulation software. Measured and simulated results should then be compared. Engineers should look at response time, overshoot, settling time, current limits and post-fault recovery. When the results do not match, the model should be tuned and tested again. This is the key point:
The final approved model should represent the commissioned plant—not only the original design.
AEMO’s modelling and commissioning framework follows this general approach by requiring models, commissioning data and validation evidence to be submitted and maintained.
When are RMS and EMT models needed?
RMS and EMT models have different purposes. RMS models are suitable for many system-wide studies. They are commonly used for transient stability, voltage recovery, frequency response and major transmission contingencies. They are relatively fast and can be used to study a large network with many generators and operating conditions. EMT models represent faster electrical and control behaviour in greater detail. They become important when a project is connected to a weak grid, when several inverter-based plants are located close to each other, when grid-forming control is used or when fast interactions cannot be represented accurately in an RMS model. EMT models should not automatically replace RMS models. Both types are useful. The important requirement is that both models represent the same plant configuration, controls, limits and protection philosophy. NERC’s EMT modelling guideline recommends clear EMT model requirements for newly connecting inverter-based resources and highlights the growing need for EMT studies as power systems become more converter-based.
What should Uzbekistan introduce?
Uzbekistan does not need to copy another country’s process word for word. The requirements should be suitable for the Uzbek power system, available software, local engineering capability and project-development process. A practical first step would be to require model verification and commissioning-based validation for major new solar, wind and BESS projects. More detailed RMS and EMT requirements could then be applied to larger projects, weak-grid connections, grid-forming BESS and projects with a significant impact on system stability. The responsibility should also be clear. The developer should provide accurate project data. The OEM should provide correct equipment and control information. The consultant should perform the technical assessment. The grid operator should define the requirements and approve the final models. Uzbekistan should also consider creating a controlled national database of approved models. This would help ensure that grid studies use the correct and latest version of each plant model. The model should be updated when there is a major firmware change, controller modification, protection-setting change, plant expansion or equipment replacement. This national database is our recommendation for Uzbekistan. It is based on the practical need for consistent models, version control and reliable system-wide studies.
This is not additional paperwork
Model validation should not become another formal document prepared only to receive approval. Its purpose is much more practical. When a real disturbance occurs, the grid does not need a beautiful report. It needs the plant to behave as expected. The plant should remain connected when required, respect its current limits, support voltage, respond to frequency and recover correctly after the disturbance. A validated model gives the grid operator confidence that the simulation represents what will happen in the real system. It also protects the developer. A correct model can help avoid unnecessary operating restrictions, repeated studies and excessive network reinforcement caused by incorrect assumptions.
Conclusion
Uzbekistan is building a more modern, renewable and inverter-based power system. That transition requires not only new power plants and transmission infrastructure, but also better engineering models.
A solar plant, wind farm or BESS should not only be physically connected to the grid. It should also be represented correctly in the national power-system model. Model validation is the bridge between what engineers see in simulation and what the equipment actually does in the field. Introducing this requirement now will be easier and safer than trying to correct large numbers of inaccurate models later.
References
1. Australian Energy Market Operator — Power System Model Guidelines, Version 3.0.
Defines power-system model and data requirements for generators and network participants.
2. Australian Energy Market Operator — Commissioning Guideline.
Describes commissioning evidence, final simulation models and model-validation reporting.
3. North American Electric Reliability Corporation — Power Plant Model Verification for Inverter-Based Resources.
Provides guidance on testing, measured data and disturbance-based verification of solar, wind and other inverter-based resources.
4. NERC Reliability Standard MOD-026-1.
Covers verification of generator excitation and plant volt/var control models used in power-system studies.
5. NERC — Electromagnetic Transient Modeling and Simulations Reliability Guideline.
Provides recommendations for EMT modelling and model requirements for inverter-based resources.
6. World Bank — Uzbekistan renewable-energy and storage project documentation.
Provides background on the country’s expansion of utility-scale solar, wind and BESS projects.