Eigen-Analysis of Large Delayed Cyber-Physical Power System by Time Integration-Based Solution Operator Discretization Methods

In this paper, solution operator discretization methods with linear multistep and implicit Runge-Kutta (SOD-LMS/IRK) are presented for efficient eigen-analysis of large delayed cyber-physical power system (DCPPS). First, the time integration-based discretization generates highly structured approximate matrices to the solution operator. Exploitation of the structure ensures low computational burden and high efficiency in solving the matrix-inversion-vector product involved in eigenvalue computation. Second, the implicitly restarted Arnoldi algorithm is employed to compute critical eigenvalue from the solution operator's discretized matrices. SOD-LMS/IRK are endowed with scalability in analyzing very large DCPPS by fully utilizing the inherent sparsity in the augmented system state matrices. The main contribution of the presented SOD-LMS/IRK is the improved efficiency and scalability of existing SOD-PS (pseudo-spectral collocation) and SOD-LMS/IRK. Numerical results on the 16-generator 68-bus test system, a 516-bus, and a 33028-bus real-life large transmission systems validate the effectiveness of the proposed SOD-LMS/IRK.