A Novel Method for Active Distribution Network Loss Minimization with Brain-Storm-Optimization-Based Robust Optimization

In this paper, a new method is proposed to minimize network losses in active distribution networks (ADNs). In recent years, studies on ADNs have been prevalent because of the emergence of distributed energy resources (DERs) such as Photovoltaic (PV) systems, wind power generation, etc. One of the challenges is how to handle the uncertainties of such renewables affected by weather conditions. This paper proposes an Evolutionary-Computation (EC)-based Robust Optimization (RO) method for minimizing the network losses in ADN s. This paper aims at reducing the risks of the obtained solutions in ADNs with parameter uncertainties in nonlinear combinatorial optimization problems. RO evaluates feasible solutions by considering the worst scenario through data sampling while EC works to evaluate better solutions that escape from local minima. As a method of EC, this paper makes use of Brain Storm Optimization (BSO) to obtain better solutions. It plays a key role to separate solution candidates into several clusters to evaluate better solutions by selecting one of four rules randomly. The effectiveness of the proposed method is demonstrated in the IEEE 69-node distribution network.