Analysis of Angular Threshold Criteria for Transient Instability Identification in Uncertain Power Systems

Abstract

This paper addresses the problem of finding a threshold (critical value of rotor angle) for the identification of transient instability by the sole use of the information on rotor angles. Several threshold values for rotor angles have been used in the past without a clear agreement on a standard value. In this paper a probabilistic approach to the problem is proposed, so that for any network a threshold can be defined based on set accuracy, while at the same time minimizing the instability identification time. Results demonstrate that highly accurate robust rotor angle thresholds can be found, even for highly uncertain power networks with a high penetration of Renewable Energy Sources (RES).

Methodology for Optimal Deployment of Corrective Control Measures to Ensure Transient Stability of Uncertain Power Systems

“…There is currently no general agreement on a definitive angle threshold for transient instability detection in multi-machine stability studies. However, in [19], a thorough analysis on the subject was done, concluding that Center Of Inertia (COI) frame based thresholds offer some conceptual and practical advantages compared to approaches measuring the difference between pairs of rotor angles. Therefore, the angle data in the COI frame are used in this paper to determine the samples at which the clustering should be performed, as it gives more insights into how groups of machines separate from each other while approaching their stability limits.…”

“…where δT is the transient instability threshold and will also denote from now on the threshold for clustering. According to literature, δT is a value usually between 100-180° [19]. This range is hence chosen for the clustering threshold assessment in this study.…”

Non-heuristic Identification of the Optimal Number of Oscillation Patterns in Uncertain Power Systems

“…The required number of MC simulations defined in [1] to achieve satisfactory accuracy was 6,000. In this paper, as in [2,9], the number of simulations is increased to 10,000 to obtain more unstable cases for credible statistical analysis. The case study used for testing has a nominal RES penetration level of 20%.…”

“…Hence, it is better to define this instant in terms of an angle threshold for transient instability identification rather than an actual point in time after a disturbance. In [9], a thorough analysis on the subject was done using thresholds based on the Maximum Angle Difference Between Two Generators (MADBTG) and on the angular distance from the Center Of Inertia (COI) reference frames. Both will be used in this study to measure system separation for evaluating the clustering results.…”

“…where Hi is the inertia constant of generator i, and NG is the total number of generators. For the MADBTG and COI frames based thresholds, the value of δT in (1) and 2for the clustering assessment exploration is defined for the ranges between 100°-360° and 100°-180°, respectively [9]. With respect to the cut-off thresholds (δC), values up to 100° are explored, so an overlap with the δT values (which start at 100°) is avoided.…”