Optimal distribution network expansion planning under uncertainty by evolutionary decision convergence

We address the problem of distribution network expansion planning under uncertainty and present a new methodology capable of dealing with uncertainty in a realistic complex problem. Uncertainty is represented by a set of scenarios. A strategic objective problem is then formulated to fit the availability of reliable first stage information. An evolutionary algorithmic procedure is proposed to find a robust decision policy; by successive adjustments to the scenarios structure, convergence is achieved for the first stage decisions. A comprehensive example is presented to illustrate the convergence process and the robustness of the achieved decision. (C) 1997 Elsevier Science Ltd.