The increasing integration of converter-based assets in power systems creates the need for practical metrics to assess whether new devices behave as expected once connected to the grid. This is particularly relevant for wind power plants, grid-forming converters, and other power-electronics-based resources, where the dynamic response depends on the control implementation, parameter tuning, operating point, grid strength, and interaction with the surrounding network. From a system operator perspective, the key question is not only whether the asset is stable, but whether its behaviour is sufficiently close to the expected one defined during studies, model validation, or the connection process.

This paper proposes the use of the Dynamic Similarity Index (DSI) as a frequency-domain metric for converter-based asset validation. The DSI was identified as a potentially useful approach to quantify how far the response of a new asset is from a reference or expected behaviour. The DSI compares the dynamic response of a system under test with a selected reference model and provides a frequency-dependent indication of their mismatch. Lower DSI values indicate that the asset is dynamically closer to the reference, while higher values indicate larger deviations. The original formulation of the DSI was proposed to assess voltage-source behaviour in converter-dominated power systems, but the same concept can be extended to assess whether a new converter-based asset follows an expected frequency-domain response.

The proposed methodology is organised in three steps. First, a reference response is defined, for example from a validated model, a manufacturer-provided model, a grid-code-compliant response, a connection-study model, or commissioning measurements. Second, the response of the asset under test is obtained from simulation, frequency-response identification, impedance scans, hardware-in-the-loop tests, or field measurements. Third, both responses are compared in the frequency domain and the DSI is computed over the frequency range of interest. This allows not only quantifying the deviation from the expected behaviour, but also identifying the frequency ranges where the asset departs the most from the reference.

A case study will be considered to illustrate the approach. At component level, the DSI will be used to compare a new converter against its expected reference behaviour under different grid strengths, operating points, and control settings.

The expected contribution is to present the DSI as an operator-oriented metric for converter-based asset validation. The approach provides a compact and interpretable answer to a practical question: is the new asset behaving as expected? By quantifying deviations from a reference dynamic response and locating the relevant frequency ranges, the proposed metric can support model validation, commissioning, control tuning, compliance assessment, and post-event analysis in converter-dominated power systems.