A Hybrid Intelligent Method for Estimating Distribution Network Reconfigurations

This paper proposes a hybrid intelligent method for estimating distribution network reconfigurations. As the idea of smart grid is well-spread in the world, a good method is required to deal with distribution network operation. In this paper, distribution network reconfigurations are discussed to evaluate the distribution network losses. The emergence of renewable energy such as PV system and wind power generation makes nodal voltage magnitudes fluctuate due to weather conditions. In this paper, a hybrid intelligent method of RBFN (Radial Basis Function Network) of ANN (Artificial Neural Network) and Regression Tree of Data Mining is proposed to estimate distribution network reconfigurations to reduce distribution network losses efficiently. RBFN is used to estimate network reconfigurations from the network conditions. As a prefilitering technique, Regression Tree plays a key role to classify input variables into some clusters where RBFN is constructed at each cluster. The use of the technique makes the learning process of RBFN much easier. The proposed method is successfully applied to a sample system. A comparison is made of the proposed and the conventional methods in terms of errors and computational time.