Numerical issues in the location of singularity-induced bifurcation points

Abstract

This paper addresses some numerical issues in the identification of singularity-induced (SI) bifurcation points in power systems using the Voltage Stability Toolbox (VST). The VST benefits from the visualization and symbolic computation capabilities of MATLAB and its extended Symbolic Toolbox to implement the local static and dynamic bifurcation analysis methods. Among different bifurcations located, singularity-induced bifurcation is of interest in this paper. The authors' simulation results indicate that identification and location of SI bifurcation points are method and tool dependent. For example, the step-size in Newton-Raphson-Seydel (NRS) algorithm is one of the key issues in the identification of SI bifurcation points. It is observed that the dynamics of SI bifurcation are so fast that one ought to choose an appropriate step-size to identify and locate it. In general, depending on the system under investigation, a small step-size provides a better chance to capture SI bifurcation, at the cost of increasing simulation time. Tradeoff studies of this effect are presented.