Abstract: Technology enhancement of Distributed Generations, as well as
deregulation and privatization in power system industry, shows a new
perspective for power systems and subsystems. When a substantial portion of
generation is in the form of dispersed and small units, a new connection
pattern emerges whereby the dispersed units are embedded in reticulation
infrastructure. Now, the flow of power is no longer the same as in the
conventional systems, since the dispersed generating plants contribute with
generation also at the distribution grids level. Connection of generation to
distribution grids cannot effectively be made, unless the some especial control
and monitoring tools are available and utilizable. State estimation in these
kinds of networks, often called mixtribution, is the preliminary and essential
tool to fulfill this requirement and also is the subject of this article.
Actually, state estimation is an optimization problem including discrete and
continuous variables, whose objective function is to minimize the difference
between calculated and measured values of variables, i.e. voltage of nodes, and
active/reactive powers in the branches. In this paper, a new approach based on
Ant Colony Optimization (ACO) is proposed to solve this optimization problem.
The feasibility of the proposed approach is demonstrated and compared with
methods based on neural networks and genetic algorithms for two test
systems.
Keywords: Distributed generation, state estimation, Ant Colony Optimization (ACO)
