Multiobjective optimization method for distribution system configuration using Pareto optimal solution

A distribution network has a huge number of configuration candidates because the network configuration is determined by the states of many sectionalizing switches (opened or closed) installed in terms of keeping power quality, reliability, and so on. Since the feeder current and voltage depend on the network configuration, the distribution loss, voltage imbalance, and bank efficiency can be controlled by changing the states of these switches. In addition, feeder currents and voltages change in response to the outputs of distributed generators (DGs) such as photovoltaic generation systems, wind turbine generation systems, and so on, that are connected to the feeder. Recently, the total number of DGs connected to distribution networks has increased drastically. Therefore, many configuration candidates of distribution networks must be evaluated multiply from various viewpoints such as distribution loss, voltage imbalance, bank efficiency, and so on, considering the power supply from the connected DGs. In this paper, the authors propose a multiobjective optimization method from three evaluation viewpoints, (1) distribution loss, (2) voltage imbalance, and (3) bank efficiency, using Pareto optimal solutions. In the proposed method, after several high-ranking candidates with small distribution loss are extracted by combinatorial optimization, each candidate is evaluated from the viewpoints of voltage imbalance and bank efficiency using the Pareto optimal solution, after which the minimum-loss configuration is chosen as the best configuration among these solutions. Numerical simulations were performed for a real-scale system model consisting of 72 distribution feeders and 234 sectionalizing switches in order to examine the validity of the proposed method.