Abstract
The study is based on time series modeling with special application to electric load consumptions modeling in a distributed environment. Existing theories and applications about artificial neural networks, backpropagation learning method, Nguyen-Widrow weight initialization technique and autocorrelation analysis were explored and applied. An adaptive stopping criterion algorithm was also integrated to BP to enable the ANN to converge on the global minimum and stop the training process without human intervention. These algorithms were combined together in designing the parallel adaptive multi-layer perceptron (PAMLP). In the simulation, the electric load consumptions of the seven (7) power utilities from Alaska in 1990–2013 were obtained from the official website of U.S. Energy Information Administration. The data set was divided into three overlapping parts: training, testing and validation sets, based on the principles of sliding-window training and walk-forward testing methods. The PAMLPs were trained and tested using the sliding-window method with 15-year window size and walk-forward testing method, respectively. The accuracy of each forecasting model produced by PAMLP was then measured using the respective out-of-sample validation sets using RMSD, CV (RMSD), and SMAPE (0% ≤ SMAPE ≤ 100%). In the monthly basis time series forecasting, the average CV (RMSD) at 7.79% and SMAPE at 3.12% for all utilities show the effectiveness of the PAMLP system across different time horizons and origin.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Ramarkrishna, R., Boiroju, N.K., Reddy, M.K.: Neural networks forecasting model for monthly electricity load in Andhra Pradesh. Int. J. Eng. Res. Appl. 2(1), 1108–1115 (2012)
Kandananond, K.: Forecasting electricity demand in Thailand with an artificial neural network approach. Energies 4, 1246–1257 (2011)
Edwards, R.E., et al.: Predicting future hourly residential electrical consumption: a machine learning case study. Energy Build. (2012). https://doi.org/10.1016/j.enbuild.2012.03.010
Haykin, S.: Neural Networks: A Comprehensive Foundation. Pearson Education Inc., Upper Saddle River, NJ, USA (2009)
Meng, M., Shang, W., Niu, D.: Monthly electric energy consumption forecasting using multiwindow moving average and hybrid growth models. J. Appl. Math. (2014)
Meng, B.M., Niu, D., Sun, W.: Forecasting monthly electric energy consumption using feature extraction. Energies 4(10), 1495–1507 (2011)
US Energy Information Administration. Monthly Energy Review. http://www.eia.gov/electricity/data.cfm (2014). Accessed 26 May 2014
Kown, D., Kim, M., Hong, C., Cho, S.: Short term load forecasting based on BPL neural network with weather factors. Int. J. Multimedia Ubiquitous Eng. 9(1), 415–424 (2014)
Yang, J., Rivard, H., Zmeureanu, R.: Building energy prediction with adaptive artificial neural networks. In: Proceedings of the 9th International IBPSA Conference, Montreal, Canada, pp. 1401–1408 (2005)
Arroyo, D., Skov, M., Huynh, Q.: Accurate electricity load forecasting with artificial neural networks. In: Proceedings of the 2005 International Conference on Computational Intelligence for Modelling, Control and Automation, and International Conference of Intelligent Agents, Web Technologies and Internet Commerce (2005)
Lin, F., Yu, X.H., Gregor, S., Irons, R.: Time series forecasting with neural networks. Complex. Int. 2 (1995)
Nguyen, D., Widrow, B.: Improving the learning speed of 2-layer neural networks by choosing initial values of the adaptive weight. In: Proceedings of the International Joint Conference on Neural Networks, San Diego, CA, USA, vol. 3, pp. 21–26 (1990)
Lalis, J.T., Gerardo, B.D., Byun, Y., Ha, Y.: Ubiquitous stopping criterion for backpropagation learning in multilayer perceptron neural networks. In: Proceedings of the 7th International Conference on Information Security and Assurance, Cebu City, Philippines, vol. 21, pp. 294–298 (2013)
Lalis, J.T., Maravillas, E.: Dynamic forecasting of electric load consumption using adaptive multilayer perceptron (AMLP). In: Proceedings of the International Conference on Humanoid, Nanotechnology, Information Technology, Communication and Control, Environment and Management, Palawan, Philippines (2014)
Acknowledgements
This work (Grants No. C0515862) was supported by Business for Cooperative R&D between Industry, Academy, and Research Institute funded Korea Small and Medium Business Administration in 2017.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2019 Springer Nature Switzerland AG
About this chapter
Cite this chapter
Lalis, J.T., Gerardo, B.D., Byun, Y. (2019). A Mechanism for Online and Dynamic Forecasting of Monthly Electric Load Consumption Using Parallel Adaptive Multilayer Perceptron (PAMLP). In: Lee, R. (eds) Computer and Information Science. ICIS 2018. Studies in Computational Intelligence, vol 791. Springer, Cham. https://doi.org/10.1007/978-3-319-98693-7_12
Download citation
DOI: https://doi.org/10.1007/978-3-319-98693-7_12
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-98692-0
Online ISBN: 978-3-319-98693-7
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)