System Oscillation Assessment for IBRs
As the power system continues to transition towards inverter-based resources, assessment of dynamic stability has become increasingly important. One area that requires particular attention is the assessment of system oscillations, commonly known as SSO studies. While SSO describes oscillations below the 50 Hz fundamental, NESO’s assessment scope also considers oscillatory behaviour above 50 Hz, extending up to 500 Hz for certain studies.
System Oscillations refers to electrical or control interactions that can occur at frequencies above or below the synchronous frequency of the power system. These interactions may arise between the plant controls, converter systems, transformers, cables, grid impedance and nearby network elements. If not adequately damped, such oscillations can lead to sustained or growing responses, which may affect plant performance and wider system stability.
NESO requires system oscillation studies to be undertaken for inverter-based resources to demonstrate that the plant does not introduce poorly damped or unstable oscillatory behaviour. The required assessment depends on the type of plant and the nature of the connection, as set out in NESO’s guidance. Depending on the project, this may include Small Signal Injection (SSI) studies, Active Frequency Scan (AFS) studies and Step Change studies. These assessments are used to confirm stable plant behaviour across the relevant operating conditions, fault levels and network configurations.
The move towards MIMO-based system oscillation assessment reflects the increasing complexity of modern grid connections. As more inverter-based resources connect to the transmission and distribution networks, stability assessments must capture the full interaction between plant controls and the grid. A matrix-based approach provides a more complete view of plant-grid interactions. It identifies coupling effects, impedance behaviour and stability through eigenvalue-based Nyquist analysis. This is particularly important for BESS and other converter-based projects, where control interactions may not be visible using simplified methods. For project developers and OEMs, this means that technical feasibility studies for system oscillation assessment are becoming more detailed and data driven.
NESO’s Updated Approach to Active Frequency Scan Studies
The National Electricity System Operator has updated the expectations for Active Frequency Scan studies in V2 of “Guidance for Oscillation Assessment for Inverter Based Resources” in September 2025. The previous approach was largely based on single-input single-output (SISO) methods, where the system response was assessed using a simpler representation of the plant and network interaction.
The updated methodology places greater emphasis on a multiple-input multiple-output, or MIMO, impedance-based approach. Instead of considering only a single scalar impedance relationship, the plant and grid are represented using full impedance matrices. These matrices capture the coupling between different axes, such as d-axis, q-axis and zero-sequence components in the DQ0 reference frame.
This is important because converter-connected plant does not always behave as a simple decoupled system. Control loops, phase-locked loops, current controllers, voltage controllers and grid-forming or grid-following control strategies can create interactions between axes. A MIMO approach is therefore more representative of the actual behaviour of inverter-based resources.
Table-1 shows a summary of the difference in approach in the Active frequency Scan.
| Assessment Area | Previous Approach | Updated MIMO Approach |
|---|---|---|
| Injection Method | Positive-sequence disturbance injection | Multi-axis disturbance injection to capture full plant-grid behaviour |
| Impedance representation | Positive-sequence or equivalent impedance | Full 3×3 DQ0 impedance matrices |
| Plant-grid model | Simplified SISO representation | MIMO representation of plant and grid |
| Coupling terms | Not explicitly assessed | Off-diagonal DQ0 coupling terms included |
| Stability method | Based mainly on impedance plots | Based on return ratio eigenvalues |
| Nyquist plots | Not included in assessment | Eigenvalue-based Nyquist assessment against (−1,0) |
| Main outputs | Magnitude, phase, resistance and reactance plots | Plant/grid DQ0 matrices, eigenvalues and Nyquist plots |
MIMO Methodology
The study should be performed in a suitable EMT simulation environment, such as PSCAD, using detailed a plant models, including OEM inverter and Plant Control models.
The first step is to derive the plant impedance matrix via the perturbation method. Three scans are performed with a small signal voltage perturbation injected in the direct (d), quadrature (q) and zero (0) axis respectively.
The disturbance magnitude is set to 0.5% of the nominal voltage, and the scan is performed from 1 Hz to 100 Hz in 1 Hz increments and from 100 Hz to 500 Hz in 10 Hz increments. For each of the tests, the voltage and current perturbations are measured across each axis to formulate the I_dq0 and V_dq0 matrices and the plant impedance matrix, Z_plant, is derived via
A separate 3×3 complex impedance matrix is generated for the Plant and the grid at each scanned frequency. In the dq0 reference frame, each matrix has the following form:
The diagonal terms represent the direct-axis responses, while the off-diagonal terms represent coupling between the d, q and zero-sequence components. For example, Z_dq represents the d-axis voltage response resulting from a q-axis current disturbance. The cross-coupling terms are retained because converter controls may create significant interaction between the different axes, which cannot be represented by a SISO scan.
Nyquist Stability Criteria
The Nyquist stability criteria are then assessed using the eigenvalues of the Loop Gain Matrix (L). Since, for the MIMO assessment, are represented by full DQ0 impedance matrices, the loop gain matrix is therefore calculated from the grid and the plant impedance matrices as:
At each scanned frequency, this loop gain matrix contains the combined effect of the plant, the grid and the coupling between the d, q and zero-sequence axes. To assess this behaviour in a simpler way, the eigenvalues of the loop gain matrix are calculated. These eigenvalues represent the main interaction modes of the plant-grid system at each frequency. For a 3×3 DQ0 matrix, three eigenvalues are obtained at each frequency. As the frequency is swept across the study range (1 – 500 Hz), each eigenvalue forms a trajectory on the Nyquist plane. These trajectories are referred to as eigenloci.
The Nyquist plot is then used to check whether any of these eigenloci encircle the critical point [−1,0]. If the eigenloci do not encircle the critical point, the plant-grid system is considered stable for the assessed operating condition. If an encirclement occurs, it indicates that the interaction between the plant controls and the grid impedance results in an unstable system. An Example of Nyquist Results are shown in Figure 1.
System Oscillation Assessment – Our Approach
Blake Clough Consulting applies advanced MIMO Active Frequency Scan techniques to System Oscillation Assessment studies, using both standard PSCAD frequency scanning tools and our custom internal solutions to support complex grid requirements.
The MHI/PSCAD DFScan tool provides a structured workflow for MIMO impedance scanning and stability assessment. However, as the latest DFScan functionality is currently available in beta versions of PSCAD, compatibility can be challenging for some OEM black-box models that are developed and validated in earlier PSCAD versions. To address this, Blake Clough Consulting has developed an internal PSCAD-based scanning block and post-processing toolbox that implements the same MIMO frequency scanning principles while providing greater flexibility for project-specific models and workflows. Our internal workflow automates disturbance injection, DQ0 plant and grid impedance calculations, eigenvalue evaluation and Nyquist stability assessment. The post-processing engine automatically generates DQ0 impedance matrices, impedance plots, eigenvalue results, Nyquist plots and critical crossing-point gain margins.
As NESO’s requirements for system oscillation assessment continue to evolve, Blake Clough Consulting is actively preparing its study workflows, tools and internal processes to align with the latest guidance. By developing automated MIMO frequency scanning and post-processing capabilities, we are well placed to support developers and OEMs in completing system oscillation assessments, supporting developer on their Projects’ connection process.
