Electricity Peak Load Forecasting using CGP based Neuro Evolutionary Techniques

DOI

10.1080/18756891.2016.1161365How to use a DOI?

Keywords

load forecasting; evolutionary algorithm; Cartesian Genetic Programming; CGPANN

Abstract

Proficient Economic and Financial planning is critical to the successful and efficient operation of power generating and distributing units. This planning becomes quite facile if the accurate and precise knowledge regarding the required power load is ascertained. This research is an innovative effort to bring forward different electric peak load forecasting models based on the Neuroevolutionary technique known as the Cartesian Genetic Programming Evolved Artificial Neural Network(CGPANN). Although CGPANN in itself is not novel but its application to power load forecasting is quite unique because of its innate ability to identify the best computationally efficient predictive model along with the recognition of the best appropriate features for load forecasting. Both the Feedforward and Recurrent CGPANN setups have been evolved here for peak load forecasting on a daily basis. The different setups developed have been trained and tested on the peak load data of United Kingdom National Grid. The models developed in this research have been tested and compared against previously proposed machine learning models. In comparison to contemporary models in the field, a more efficient and accurate peak load forecasting model has been produced using CGPANN based prediction methods.

Copyright

© 2016. the authors. Co-published by Atlantis Press and Taylor & Francis

Open Access

This is an open access article under the CC BY-NC license (http://creativecommons.org/licences/by-nc/4.0/).

Download article (PDF)
View full text (HTML)

TY  - JOUR
AU  - Gul Muhammad Khan
AU  - Rabia Arshad
PY  - 2016
DA  - 2016/04/01
TI  - Electricity Peak Load Forecasting using CGP based Neuro Evolutionary Techniques
JO  - International Journal of Computational Intelligence Systems
SP  - 376
EP  - 395
VL  - 9
IS  - 2
SN  - 1875-6883
UR  - https://doi.org/10.1080/18756891.2016.1161365
DO  - 10.1080/18756891.2016.1161365
ID  - Khan2016
ER  -