Optimal Substation Location and Network Routing using an Improved GA Based Solution Approach

The paper deals with a problem of optimizing substation locations and network routing to serve a given load demand. An objective function of fixed costs of substations and lines and the present worth cost of energy loss of the line segments is optimized, subject to a set of constraints. The problem is combinatorial and thus they have a discrete objective function. Considering the limitations of the conventional methods available to solve such problems, researchers have used various intelligent computational methods, derived from nature. Genetic Algorithm is one such technique now used popularly for solving power system problems. This paper presents yet another GA based solution procedure that uses a simple codification scheme, genetic operators and control parameters to model the Distribution System Problem within the framework of GA. The model has been intensively tested in real distribution networks, which proves their practical application to large power distribution systems with relatively smaller solution times. The paper presents an improved GA based solution approach and the results obtained when applied on to an existing large sub-transmission network.