Robust and efficient approach based on Richardson extrapolation for solving badly initialised/ill-conditioned power-flow problems

In this study, the authors focus on solving the power-flow (PF) problems of badly initialised/ill-conditioned power systems, where the solution of these systems is a challenge for most of state-of-the-art PF techniques. This challenge is increased in case of large and very large-scale power systems which have a huge number of variables. Consequently, the PF techniques used should have a high degree of efficiency in order to handle the large vectors and matrix computations. In this study, a novel PF approach based on the Richardson extrapolation is proposed to solve these systems. The proposed approach is validated using various badly initialised/ill-conditioned systems, comparing its performance with several well known PF techniques. The obtained results prove that the proposed approach is robust and efficient enough to properly manage the badly initialised/ill-conditioned systems even if they consist of a huge number of buses, whereas the other PF techniques face different convergence difficulties.