New Approach to Peak Load Forecasting based on EMD and ANFIS

This paper presents a new approach to Peak Load Forecasting based on EMD and ANFIS. EMD is used to decompose the load data into several Intrinsic Mode Functions (IMFs); then appropriate ANFIS models are developed for these IMFs. The target model proposed in this paper, EMD–ANFIS, is achieved by combining the predictions from these IMF–ANFIS models together and this is used for forecasting purposes. A real life data obtained from Power Holding Company of Nigeria (PHCN), Bida, was used to evaluate the forecast accuracy of the proposed model. The results revealed that the proposed EMD–ANFIS model yields better results when compared to ANN and EMD–ANN models. The proposed EMD–ANFIS model recorded 2.76% and 50.05% improvements over EMD–ANN and traditional ANN models, respectively as judged by the overall MAPE of the models.