Locating voltage collapse points using evolutionary computation techniques

In recent years, evolutionary computation (EC) techniques have proven to be an useful alternative approach for solving many highly nonlinear power system planning and operation problems. The objective of this paper is to investigate mathematically-complex voltage collapse problems using EC techniques, in particular the particle swarm optimization (PSO) and differential evolution (DE) algorithms. It demonstrates the exceptional searching capabilities of both the PSO and DE algorithms to locate voltage collapse point solutions (also widely known as nose points or critical points), which are at least comparable to those obtained using the well-known continuation power flow (CPF) technique. The feasibility and practicality of this approach has been tested on a 3-machine 9-bus, the IEEE 118-bus and the IEEE 300-bus power systems.