An ensemble model for day-ahead electricity demand time series forecasting

Authors:
Wen Shen;Vahan Babushkin;Zeyar Aung;Wei Lee Woon
Affiliations:
Masdar Institute of Science and Technology, Abu Dhabi, Uae;Masdar Institute of Science and Technology, Abu Dhabi, Uae;Masdar Institute of Science and Technology, Abu Dhabi, Uae;Masdar Institute of Science and Technology, Abu Dhabi, Uae
Venue:
Proceedings of the fourth international conference on Future energy systems
Year:
2013

Citing 5
Cited 0

Efficient Implementation of the Fuzzy c-Means Clustering Algorithms

IEEE Transactions on Pattern Analysis and Machine Intelligence
LBF: A Labeled-Based Forecasting Algorithm and Its Application to Electricity Price Time Series

ICDM '08 Proceedings of the 2008 Eighth IEEE International Conference on Data Mining
Energy Time Series Forecasting Based on Pattern Sequence Similarity

IEEE Transactions on Knowledge and Data Engineering
A Cluster Separation Measure

IEEE Transactions on Pattern Analysis and Machine Intelligence
Clustering of the self-organizing map

IEEE Transactions on Neural Networks

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this work, we try to solve the problem of day-ahead prediction of electricity demand using an ensemble forecasting model. Based on the Pattern Sequence Similarity (PSF) algorithm, we implemented five forecasting models using different clustering techniques: K-means model (as in original PSF), Self-Organizing Map model, Hierarchical Clustering model, K-medoids model, and Fuzzy C-means model. By incorporating these five models, we then proposed an ensemble model, named Pattern Forecasting Ensemble Model (PFEM), with iterative prediction procedure. We evaluated its performance on three real-world electricity demand datasets and compared it with those of the five forecasting models individually. Experimental results show that PFEM outperforms all those five individual models in terms of Mean Error Relative and Mean Absolute Error.