Precision enhancement of the traditional sub‐synchronous analysis by using frequency components of instantaneous magnetic flux

Abstract

Sub-synchronous resonance (SSR) is a great danger to the stable and safe operation of the synchronous generator and power network. Detecting this condition in a fast and precise method is very important for power engineers. During the last decade, many methods have been proposed for SSR detection. The most famous of these methods use shaft speed analysis. On the other hand, it is noteworthy that SSR is a phenomenon that needs the use of both mechanical and electrical quantities for its analysis. Therefore, if both kinds of the signal are used, the protection system reliability can be enhanced significantly and its trip is more efficient and consistent. Subsequently, to enhance the precision and sensitivity of the SSR detection, this paper proposes the use of frequency analysis of the magnetic flux linkage of the generator as an electrical signal along with the shaft speed as a mechanical signal. The magnetic flux is a local index that is available at the relay location and does not require a communication channel to transmit. This index contains a lot of useful information about the performance of the machine. However, the main disadvantage of this index is its accessibility, which requires invasive sensors.

Fast and accurate method for sub‐synchronous oscillation detection

“…Due to the limited number of datasets from field-recorded SSO cases, the commonly used IEEE first benchmark model for SSO study [3,5,6,[25][26][27] is utilized in order to analyze the performance of the proposed method. A similar model (IEEE second benchmark model) has been used in [10,12,18,28,29] to prove the effectiveness of the presented methods. The power system model is shown in Figure 5, where a 575 V is generated with a 50 Hz synchronous frequency ( f s ).…”

“…Both the current and voltage signals are required. In [18], the authors proposed the use of frequency analysis of the magnetic flux linkage of the generator as an electrical signal along with the shaft speed as a mechanical signal to detect the SSO. This method consumes 0.6 s to discover the SSO.…”

“…The results showed that the threshold value for the current signal is higher than the threshold value for voltage signal because the sub-synchronous component within the current signal is more obvious than the sub-synchronous component within the voltage signal. The SSO was detected after 0.1 s. On the basis of the previous works, the researchers have used different input signals such as current signal [9,21,22], voltage signal [6,10,14,21], active power signal [11], and both the electrical and mechanical signals [7,12,18]. In [6], the SSO detection methods that have been published before 2020 have been summarized in terms of the used approach, input signal, and SSO detection time.…”