Efficient calculation of critical eigenvalue clusters in the small signal stability analysis of large power systems

The paper presents a methodology for the calculation of a selected set of eigenvalues, considered critical in the small signal stability analysis of power systems. It analyzes several alternatives for refining a preliminary rough solution obtained by subspace iterations. These alternatives range from constant-matrix iterative refinement to Newton's method. Due to an adaptive solution strategy, the overall algorithm is very robust. Newton's method is much faster than existing approaches. The performance of these methods is demonstrated on several test systems.<
>