Frequency Quality Metrics based on Second-Order Derivative and Autocorrelation
Pith reviewed 2026-05-10 16:34 UTC · model grok-4.3
The pith
Power systems can show good frequency quality by standard metrics yet poor quality by metrics based on second-order derivatives and autocorrelation.
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
Core claim
The paper claims that frequency quality metrics constructed from the second-order derivative of frequency and from its stochastic autocorrelation function reveal aspects of grid performance that standard transmission-system-operator metrics miss, so that a power system may register as having good frequency quality by existing criteria while registering as having poor frequency quality by the new criteria.
What carries the argument
The proposed frequency quality metrics that quantify second-order frequency dynamics through the time derivative and that quantify stochastic properties through the autocorrelation function of frequency time series.
Load-bearing premise
Second-order derivative and autocorrelation metrics provide a more accurate assessment of true dynamic frequency behavior than the metrics currently used by transmission system operators.
What would settle it
A recorded grid disturbance in which the new metrics show no stronger correlation with the severity or duration of frequency excursions than the standard metrics already in use.
Figures
read the original abstract
This industry-oriented paper originates from the observation that current frequency quality metrics utilized by transmission system operators (TSOs) fail to fully capture the dynamic behavior of the grid frequency. Motivated by this gap, the paper proposes novel frequency quality metrics based on second-order dynamics and stochastic autocorrelation. Using real-world data with 0.1 s and 1 s resolution from the Irish, Great Britain and Nordic systems and running dynamic stochastic simulations, the paper shows that the proposed metrics bring new and counterintuitive insights in terms of how good or poor the frequency quality of power grids is beyond current well-known metrics. In particular, the paper shows that a power system may show good frequency quality using standard metrics and poor frequency quality using the proposed metrics. Overall, the paper contributes to improve the understanding of frequency quality.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper proposes new frequency quality metrics based on the second-order derivative of frequency and autocorrelation to better capture dynamic grid behavior beyond standard TSO metrics. Using real data from the Irish, GB, and Nordic systems plus dynamic stochastic simulations, it shows that systems can rate as good quality under conventional metrics but poor under the proposed ones, contributing new insights into frequency quality assessment.
Significance. If the new metrics demonstrate superior relevance to stability, they could refine TSO practices and grid monitoring. The application to multiple real-world datasets and simulations is a positive aspect, offering concrete counterexamples to standard assessments. However, without links to operational outcomes, the practical significance for improving stability remains provisional.
major comments (2)
- [Results section] The results from real data (Irish/GB/Nordic systems) and simulations claim new insights into frequency quality, but no statistical significance tests, error bars, or data exclusion rules are reported, leaving the robustness of the 'good vs. poor' discrepancies unverified.
- [Simulation and Case Study sections] The central claim that the proposed metrics provide a more accurate assessment requires evidence that elevated second-derivative or autocorrelation values correlate with or predict stability events (e.g., nadir breaches or RoCoF thresholds); this validation is absent from the simulation and data analysis.
minor comments (2)
- [Methods] Clarify the exact definitions and units of the second-order derivative and autocorrelation metrics with explicit equations in the main text.
- [Abstract] The abstract could briefly quantify one example discrepancy (e.g., a specific metric value difference) to strengthen the summary of contributions.
Simulated Author's Rebuttal
We thank the referee for the constructive comments on our manuscript. We address each major comment point by point below, indicating where revisions will be made.
read point-by-point responses
-
Referee: [Results section] The results from real data (Irish/GB/Nordic systems) and simulations claim new insights into frequency quality, but no statistical significance tests, error bars, or data exclusion rules are reported, leaving the robustness of the 'good vs. poor' discrepancies unverified.
Authors: We acknowledge that formal statistical significance tests, error bars, and explicit data exclusion rules are not included in the current manuscript. The discrepancies between standard and proposed metrics are demonstrated through direct computation and visualization on the full datasets from the Irish, GB, and Nordic systems as well as the stochastic simulations. In the revised version, we will add appropriate statistical tests (such as paired t-tests or non-parametric equivalents) to quantify the significance of the observed differences, include error bars on summary plots where variability across time windows or simulation runs is relevant, and clearly document the data preprocessing and exclusion criteria applied to the frequency time series. revision: yes
-
Referee: [Simulation and Case Study sections] The central claim that the proposed metrics provide a more accurate assessment requires evidence that elevated second-derivative or autocorrelation values correlate with or predict stability events (e.g., nadir breaches or RoCoF thresholds); this validation is absent from the simulation and data analysis.
Authors: The manuscript's core contribution is to show that the proposed second-order derivative and autocorrelation metrics can identify frequency quality issues not captured by conventional TSO metrics, resulting in different assessments of the same systems. It does not assert that these metrics are more accurate specifically because they correlate with or predict discrete stability events such as nadir breaches or RoCoF threshold violations. The simulations and real-data analysis focus on revealing dynamic behavior and counterintuitive quality ratings. We agree that establishing direct predictive links to operational events would strengthen practical relevance, but this is outside the stated scope of proposing and applying the new metrics. We will add a brief discussion of this limitation and its implications for future work. revision: no
Circularity Check
No significant circularity; metrics defined independently and applied to external data
full rationale
The paper defines frequency quality metrics directly from the second-order derivative of frequency and from autocorrelation functions, which are standard mathematical constructs in dynamics and stochastic processes. These definitions are applied to independent real-world datasets from the Irish, Great Britain, and Nordic power systems plus separate stochastic simulations. No parameter fitting to target outcomes, no self-referential definitions, and no load-bearing self-citations are present in the derivation chain. The comparison between standard TSO metrics and the proposed ones is therefore an external evaluation rather than a tautology.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.