**7. Conclusions**

In the paper a novel algorithm for the analysis and parameters estimation of inter-area oscillations has been proposed. The method consists of three steps: (i) the separation, according to an optimal bisection frequency, of the input signal in individual mono-component oscillations; (ii) reshaping of each mono-component in order to minimize the distortion due to the separation; (iii) correction of Gibbs effect and parameters estimation through a NLS algorithm.

Several tests, both with numerical and experimental signals have been carried out in order to assess the method. With respect to solutions already available in literature, the proposed method is characterized by excellent performance in terms of accuracy of the parameter estimates and robustness to noise.

As regards the computational burden and the response time, the proposed method allows to acquire and process a reduced number of samples, in favor of the computational resources. In particular, the described results have been obtained by processing 500 samples acquired with a sampling rate of 50 S/s. Concerning the response time, the time required by the algorithm execution is negligible with respect to the time spent for the acquisition of the signal samples. The method performance, in fact, is sensitive to the spectral resolution and, then, to the size of the analyzed time window. By considering the typical values of frequency and damping coefficient characterizing the inter-area oscillations, the authors have selected, as optimal trade off between the spectral resolution and the response time, a time window of 10 s.

The ongoing activity is focused on the development of a dedicated measurement instrument. At this aim, the realized algorithm has to be converted from MATLAB® and transferred on an embedded system.

**Author Contributions:** conceptualization, D.L. and A.L.; data curation, F.B., L.P.D.N. and S.T.; methodology, F.B., L.P.D.N., A.L. and S.T.; software, S.T.; supervision, D.L.; validation, F.B. and L.P.D.N.; writing—original draft, F.B., L.P.D.N. and S.T.; writing—review and editing, D.L. and A.L.

**Funding:** This research received no external funding.

**Acknowledgments:** The authors would like to thank Salvatore Lodato for his technical and scientific contribution to the research activity and to the drafting of the paper.

**Conflicts of Interest:** The authors declare no conflict of interest.
