Optimal location, sizing and allocation of subtransmission substations using K-means algorithm

The present article presents a novel and applied approach based on clustering to determine optimal locations, sizing of sub-transmission substations with their associated service area without determining location of candidate substations. The goal of this optimization is to minimize all of fixed costs and operation costs while all of constraints are met. Cost of equipment, construction, MV feeder cables, and power losses as well as existing substations are considered in cost function. Also we've considered different constraints such as voltage drop, substation power capacity limit, thermal limit, and radial network. Proposed model includes a method for applying planning constraints in K-means based algorithm that it leads to convert K-means based algorithm into an applied tool with most accuracy. We employed impact factor to adjust effect magnitude of load power to determine new sub-transmission locations. This algorithm is tested by real urban network to verify effectiveness and feasibility of proposed model.