Theoretical and Experimental Limitations of RoCoF Estimation

Abstract

A precise estimation of the Rate of Change of Frequency (RoCoF) is crucial for secure power system operation. In fact, RoCoF is strictly related to the amount of the available physical and/or virtual inertia of the system and the severity of the active power unbalance following a disturbance. For this reason, it is widely exploited in different protection systems, e.g., Anti-Islanding, Under Frequency Load Shedding (UFLS) and wide-area protection systems. The new paradigm of modern power systems, with a low-inertia and converter-based generation assets, is increasing the transient severity, making the frequency and the RoCoF estimation more complex and less precise for the actual devices. This work addresses this issue by proposing a numerically robust approach based on concepts inherited from differential geometry and fluid mechanics. The proposed approach is then tested with high-sampling real experimental measurements and used to develop a faster control logic for a RoCoF-based UFLS control scheme. The proposed approach provides information to protections regarding the nature of the contingency which can be used to improve its response.

Figures (13)

• Figure 1: Measurements of the normalized voltage measurements during the transformer energisation event in the $\alpha\beta\gamma$ frame.
• Figure 2: Instantaneous frequency, $\boldsymbol{\omega_\upsilon}$, and qss frequency, $\boldsymbol{\omega}_{\rm QSS}$, results of the transformer energisation case study.
• Figure 3: Time derivative of instantaneous and qss frequency, $\boldsymbol{\omega'_\upsilon}$ and $\boldsymbol{\omega}_{\rm QSS}'$ respectively, of the transformer energisation case study.
• Figure 4: Overview of time derivative circulation, $\Gamma'$, results of the transformer energization case study (a), and detailed view of the results under the event (b). Periods where condition \ref{['metric']} is not satisfied are gray shadowed.
• Figure 5: rocof estimation as an average of time derivative of instantaneous frequency (Conventional) and by definition in \ref{['eq:rocof']} (QSS-based) for $\Delta t_w=500$ms. Red-dotted line represents a realistic threshold for system stability control.