Abstract

Distributed slack bus (DSB) models for the resolution of power flow problems are a more realistic approach than the classical single slack bus concept. DSB models have been widely implemented in iterative approaches such as in Newton-Raphson power flow methods. However, iterative algorithms may lead to incorrect solutions or may fail to reach convergence under inappropriate initial estimations. The holomorphic embedding load-flow method (HELM) deals with the solution of the nonlinear power flow equations by means of complex analysis techniques, resulting in a deterministic and noniterative method, which is independent of initial estimations. As such, this paper proposes, from two different methodologies, a novel formulation and implementation of the DSB model for the HELM, which overcomes in that way the disadvantages of iterative methods and also incorporates the inherent advantages of DSB models. Moreover, this paper performs convergence and speed capability tests of the two proposed methodologies using two real-size benchmark systems.