Algorithm selection and combining multiple learners for residential energy prediction

Balancing supply and demand management in energy grids requires knowing energy consumption in advance. Therefore, forecasting residential energy consumption accurately plays a key role for future energy systems. For this purpose, in the literature a number of prediction algorithms have been used. Th...

Full description

Saved in:
Bibliographic Details
Published in:Future generation computer systems Vol. 99; pp. 391 - 400
Main Authors: Güngör, Onat, Akşanlı, Barış, Aydoğan, Reyhan
Format: Journal Article
Language:English
Published: Elsevier B.V 01.10.2019
Subjects:
Algorithm selection
Combining multiple learners
Electricity consumption prediction
Time series prediction
Electricity consumption prediction
Algorithm selection
Time series prediction
Combining multiple learners
ISSN:0167-739X, 1872-7115
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Balancing supply and demand management in energy grids requires knowing energy consumption in advance. Therefore, forecasting residential energy consumption accurately plays a key role for future energy systems. For this purpose, in the literature a number of prediction algorithms have been used. This work aims to increase the accuracy of those predictions as much as possible. Accordingly, we first introduce an algorithm selection approach, which identifies the best prediction algorithm for the given residence with respect to its characteristics such as number of people living, appliances and so on. In addition to this, we also study combining multiple learners to increase the accuracy of the predictions. In our experimental setup, we evaluate the aforementioned approaches. Empirical results show that adopting an algorithm selection approach performs better than any single prediction algorithm. Furthermore, combining multiple learners increases the accuracy of the energy consumption prediction significantly. •Energy prediction is important for utilities to balance supply and demand.•A single prediction algorithm might not perform well for a variety of households.•In our data set, we see that TESLA outperforms other time series prediction methods.•Algorithm selection performs better than a single prediction algorithm.•Random Forest as our algorithm selection method gives the minimum prediction error.•Combining multiple learners further increases energy prediction accuracy.
ISSN:0167-739X
1872-7115
DOI:10.1016/j.future.2019.04.018