Inter-area Oscillations in Power Systems by Arturo Roman Messina

Author:Arturo Roman Messina
Language: eng
Format: epub
Publisher: Springer US, Boston, MA

Fig. 5.2Real-time small-signal stability centralized control strategy

On the other hand, Prony analysis [20], has been successfully applied for off-line signal and model identification in power systems [21, 22, 23], and a wide variety of fields [24, 25, 26]. It has been shown that comparing with Fourier transform-based techniques, Prony’s method is more powerful when it works [27]. Issues such as system’s nonlinearities [28], power system’s high-order model [29, 30], noise, low signal-to-noise ratios (SNRs), and defining the time window or data length [31, 32] have limited its use in real-time applications. However, the problem suggested by [29] was already solved by one upgrade in [33], and the high-frequency modes [30] are not an issue in small-signal stability studies, as the low-frequency nature of inter-area and local modes. Reference [32] analyzes the issues of sample size selection, data length selection, and noise in Prony method. However, as seen later in this chapter with simple rules and a smoothing filter, these issues can be strongly reduced in power system’s small-signal stability studies.

In [28], the authors apply the Hilbert spectral analysis obtaining the instantaneous frequency and damping. In [34], the authors use matrix pencil method and Hankel total least-squares method to determine also the frequency and damping of power system modes. They results are comparable to Prony method. Moreover, in [34] the authors use the same principles presented in this work to increase the reliability of the oscillation monitoring system (OMS).

The controller framework is summarized in Fig. 5.2. Rules for applying the right power swing phase compensation by strategically located SVCs are developed. Also, they are demonstrated in the Kundur’s two-area power system [35] and briefly in the Western Systems Coordinating Council (WSCC) large-scale power system. Practical rules for applying multi-Prony analysis successfully on real-time basis for detection of low damped oscillations are developed. Off-line rules for selection and use of the appropriate signals for improving the accuracy, the observability, and the reliability of multi-Prony analysis are obtained through simulations in the two-area power system. Later, these rules are shown with real noisy data and final online rules are proposed and proved with the WSCC power system for a validated case of the August 10, 1996 blackout [36].

The proposed approach is based on previous works [37, 33, 38] and tested with real data and a validated simulation case [36]. Reference [37] introduces the applicability and usefulness of Prony in power system, while [38] improves substantially the accuracy of Prony’s estimations. Also, the idea of grouping signals by dominant modes to increase the observability of certain modes was first suggested in [33]. We have taken these previous results and developed some more rules to improve the accuracy, the reliability, and the observability of the multi-Prony analysis for the purpose of a real-time controller. The framework has been extended into an OMS at Washington State University [34] and a prototype version of OMS has been recently implemented at Tennessee Valley Authority [39].

Finally, the main objective of this work is on designing a practical real-time wide-area

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.