Computational & Technology Resources
an online resource for computational,
engineering & technology publications
PROCEEDINGS OF THE THIRD INTERNATIONAL CONFERENCE ON SOFT COMPUTING TECHNOLOGY IN CIVIL, STRUCTURAL AND ENVIRONMENTAL ENGINEERING
Edited by: Y. Tsompanakis
Load and Energy Billing Forecasting of an Electricity Utility
K. Figueiredo1, M.M.B.R. Vellasco2 and J.C.A. Mattoso3
1Department of Applied Mathematics and Computational Science,
UEZO, Rio de Janeiro, Brazil
K. Figueiredo, M.M.B.R. Vellasco, J.C.A. Mattoso, "Load and Energy Billing Forecasting of an Electricity Utility", in Y. Tsompanakis, (Editor), "Proceedings of the Third International Conference on Soft Computing Technology in Civil, Structural and Environmental Engineering", Civil-Comp Press, Stirlingshire, UK, Paper 32, 2013. doi:10.4203/ccp.103.32
Keywords: prediction of billing and load, neural networks, climate and economic.
This study presents a methodology for predicting electrical energy load and billing that considers climate data and economic-financial factors. Using the clustering method, the concession area of a distributing company was divided into 17 microclimates according to the climate characteristics of districts (of the city of Rio de Janeiro) and municipalities (of the state of Rio de Janeiro). From the microclimates, the billing and load series were separated by consumption class (residential, commercial, public power, public service and rural). Models based on multilayer perceptron (MLP) multistep neural networks and linear regression were developed for the monthly prediction with a maximum of 15 steps forward (multistep). Adequate performance was achieved for the majority of the consumption and load classes.
purchase the full-text of this paper (price £20)