Short-term Load Forecasting Model Based on RBF Neural Network Optimized by Artificial Bee Colony Algorithm

The short-term load forecasting model based on RBF neural network optimized by the artificial bee colony algorithm is constructed in this paper. RBF-NN is a multi-layer feedforward neural network based on the principle of multivariable interpolation. It can map any complex nonlinear relationship and has the ability of global approximation. The convergence speed of RBF-NN is fast. The artificial bee colony algorithm is used to train the RBF neural network in the paper. The artificial bee colony algorithm has the advantages of simple implementation, good global search ability and strong robustness. It can quick jump out of the local optimum. Through the forecasting test for the load system of actual distribution network, the results verified that the model based on RBF neural network optimized by the artificial bee colony algorithm can obtain the satisfactory prediction result.