Design philosophy revision in metropolises distribution system by comparative chromosome genetic algorithm

The significant values of power losses of low voltage distribution system in Iranian large cities, have been actuated the planning managers to use the small transformers and set them as near as possible to the load points to reduce the LV network. In this paper the correctness of this approach has been evaluated and the optimum ranges for distribution substation capacities with respect to the load density and dispersion in various areas of large cities have been determined. A three stage distribution network planning algorithm based on comparative Genetic Algorithm and prim method has been applied. Various random generated cases for location and values of the load points in different categories of load densities and dispersions have been studied to conclude the reliable relationship between these characteristics of the load and the optimum feeding transformer capacity.