Successive power flows with adaptive step-length increments for fast approximation of the maximum loading point

摘要

This study is concerned with the problem of computing the maximum loading point (MLP) in large-scale power systems. A modified asymptotic numerical method (ANM) with λ-parameterisation is used to fast approximate the MLP, which needs to solve a successive of power flows with adaptive load and generation increments. The ANM with Newton corrections is used to deal with the problem of reactive limits violations. The saddle-node bifurcation is identified by the accumulation of small step-lengths. The proposed algorithm has been tested on the IEEE-300, 9241-bus and a real 35,000-bus power system in China. Some comparisons with other algorithms have been performed. Numerical results reveal that a lower number of the Jacobian matrix factorisations are needed with the proposed method, which significantly reduces the computational costs.