Prediction of Power Outages in Distribution Network with Grey Theory

Abstract

Annual power outages in distribution network are highly related to the reliability of the power grid and directly affect the customers' satisfaction. The severe weather conditions, increasing loads as well as aging equipment are all potential threatens to the electrical grid infrastructure. A good prediction of the number of outages is essential for the maintenance planning and cost benefit analysis of investment. In order to predict the out-of-service cases in the power grid, the GM (1,1) (first-order Grey Modelling) forecasting method is introduced in this paper. To improve the accuracy of the prediction, the PSO (particle swarm optimization) algorithm is applied for the parameter optimization in the modeling. The number of outages in the next two years of a medium-voltage urban distribution network are predicted based on the records in the past 7 years. The good performance of the simulation results verifies the proposed forecasting method.