Optimal distribution substation locating in large distribution systems

The problem of distribution system planning is a large-scale, very complex and difficult to define problem. This paper presents the application of an efficient genetic algorithm (GA) for the optimal design of large distribution system, solving the optimal sizing and locating of substation using their corresponding fixed and variable cost as well as any operating and optimization constraints. As a new concept loss characteristic matrix has introduced in this optimal substation placement problem. In order to reduce the computation time and being confident from the feasibility of huge search space the basic operator of GA has been designed such a way that avoids and suppresses from infeasible solution. The value of loads density in every load black determined by mid-term and long-term load forecasting module. The developed GA-based software has been tested in a large distribution system with real size.