Evolving RBF predictive models to forecast the Portuguese electricity consumption

Pedro M. Ferreira, António E. Ruano, Rui Pestana, L. Kóczy

Research output: Chapter in Book/Report/Conference proceedingConference contribution

5 Citations (Scopus)

Abstract

The Portuguese power grid company wants to improve the accuracy of the electricity load demand (ELD) forecast within an horizon of 24 to 48 hours, in order to identify the need of reserves to be allocated in the Iberian Market. In this work we present some preliminary results about the identification of radial basis function (RBF) neural network (NN) ELD predictive models and about the performance of a model selection algorithm. The methodology follows the principles already employed by the authors in different applications: the NN models are trained by the Levenberg-Marquardt algorithm using a modified training criterion, and the model structure (number of neurons and input terms) is evolved using a Multi-Objective Genetic Algorithm (MOGA). The set of goals and objectives used in the MOGA model optimisation reflect different requirements in the design: obtaining good generalisation ability, good balance between one-step-ahead prediction accuracy and model complexity, and good multi-step prediction accuracy. A number of experiments were carried out, whose results are presented, producing already a number of models whose predictive performance is satisfactory.

Original languageEnglish
Title of host publicationIFAC Proceedings Volumes (IFAC-PapersOnline)
Volume2
EditionPART 1
DOIs
Publication statusPublished - 2009
Event2nd IFAC International Conference on Intelligent Control Systems and Signal Processing - Istanbul, Turkey
Duration: Sep 21 2009Sep 23 2009

Other

Other2nd IFAC International Conference on Intelligent Control Systems and Signal Processing
CountryTurkey
CityIstanbul
Period9/21/099/23/09

Fingerprint

Electricity
Genetic algorithms
Neural networks
Model structures
Neurons
Industry
Experiments

Keywords

  • Electricity load demand
  • Modeling
  • Neural networks
  • Prediction
  • Radial basis functions

ASJC Scopus subject areas

  • Control and Systems Engineering

Cite this

Ferreira, P. M., Ruano, A. E., Pestana, R., & Kóczy, L. (2009). Evolving RBF predictive models to forecast the Portuguese electricity consumption. In IFAC Proceedings Volumes (IFAC-PapersOnline) (PART 1 ed., Vol. 2) https://doi.org/10.3182/20090921-3-TR-3005.00073

Evolving RBF predictive models to forecast the Portuguese electricity consumption. / Ferreira, Pedro M.; Ruano, António E.; Pestana, Rui; Kóczy, L.

IFAC Proceedings Volumes (IFAC-PapersOnline). Vol. 2 PART 1. ed. 2009.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Ferreira, PM, Ruano, AE, Pestana, R & Kóczy, L 2009, Evolving RBF predictive models to forecast the Portuguese electricity consumption. in IFAC Proceedings Volumes (IFAC-PapersOnline). PART 1 edn, vol. 2, 2nd IFAC International Conference on Intelligent Control Systems and Signal Processing, Istanbul, Turkey, 9/21/09. https://doi.org/10.3182/20090921-3-TR-3005.00073
Ferreira PM, Ruano AE, Pestana R, Kóczy L. Evolving RBF predictive models to forecast the Portuguese electricity consumption. In IFAC Proceedings Volumes (IFAC-PapersOnline). PART 1 ed. Vol. 2. 2009 https://doi.org/10.3182/20090921-3-TR-3005.00073
Ferreira, Pedro M. ; Ruano, António E. ; Pestana, Rui ; Kóczy, L. / Evolving RBF predictive models to forecast the Portuguese electricity consumption. IFAC Proceedings Volumes (IFAC-PapersOnline). Vol. 2 PART 1. ed. 2009.
@inproceedings{77ff1af1e794493f9e49ab9aacdb7b12,
title = "Evolving RBF predictive models to forecast the Portuguese electricity consumption",
abstract = "The Portuguese power grid company wants to improve the accuracy of the electricity load demand (ELD) forecast within an horizon of 24 to 48 hours, in order to identify the need of reserves to be allocated in the Iberian Market. In this work we present some preliminary results about the identification of radial basis function (RBF) neural network (NN) ELD predictive models and about the performance of a model selection algorithm. The methodology follows the principles already employed by the authors in different applications: the NN models are trained by the Levenberg-Marquardt algorithm using a modified training criterion, and the model structure (number of neurons and input terms) is evolved using a Multi-Objective Genetic Algorithm (MOGA). The set of goals and objectives used in the MOGA model optimisation reflect different requirements in the design: obtaining good generalisation ability, good balance between one-step-ahead prediction accuracy and model complexity, and good multi-step prediction accuracy. A number of experiments were carried out, whose results are presented, producing already a number of models whose predictive performance is satisfactory.",
keywords = "Electricity load demand, Modeling, Neural networks, Prediction, Radial basis functions",
author = "Ferreira, {Pedro M.} and Ruano, {Ant{\'o}nio E.} and Rui Pestana and L. K{\'o}czy",
year = "2009",
doi = "10.3182/20090921-3-TR-3005.00073",
language = "English",
isbn = "9783902661661",
volume = "2",
booktitle = "IFAC Proceedings Volumes (IFAC-PapersOnline)",
edition = "PART 1",

}

TY - GEN

T1 - Evolving RBF predictive models to forecast the Portuguese electricity consumption

AU - Ferreira, Pedro M.

AU - Ruano, António E.

AU - Pestana, Rui

AU - Kóczy, L.

PY - 2009

Y1 - 2009

N2 - The Portuguese power grid company wants to improve the accuracy of the electricity load demand (ELD) forecast within an horizon of 24 to 48 hours, in order to identify the need of reserves to be allocated in the Iberian Market. In this work we present some preliminary results about the identification of radial basis function (RBF) neural network (NN) ELD predictive models and about the performance of a model selection algorithm. The methodology follows the principles already employed by the authors in different applications: the NN models are trained by the Levenberg-Marquardt algorithm using a modified training criterion, and the model structure (number of neurons and input terms) is evolved using a Multi-Objective Genetic Algorithm (MOGA). The set of goals and objectives used in the MOGA model optimisation reflect different requirements in the design: obtaining good generalisation ability, good balance between one-step-ahead prediction accuracy and model complexity, and good multi-step prediction accuracy. A number of experiments were carried out, whose results are presented, producing already a number of models whose predictive performance is satisfactory.

AB - The Portuguese power grid company wants to improve the accuracy of the electricity load demand (ELD) forecast within an horizon of 24 to 48 hours, in order to identify the need of reserves to be allocated in the Iberian Market. In this work we present some preliminary results about the identification of radial basis function (RBF) neural network (NN) ELD predictive models and about the performance of a model selection algorithm. The methodology follows the principles already employed by the authors in different applications: the NN models are trained by the Levenberg-Marquardt algorithm using a modified training criterion, and the model structure (number of neurons and input terms) is evolved using a Multi-Objective Genetic Algorithm (MOGA). The set of goals and objectives used in the MOGA model optimisation reflect different requirements in the design: obtaining good generalisation ability, good balance between one-step-ahead prediction accuracy and model complexity, and good multi-step prediction accuracy. A number of experiments were carried out, whose results are presented, producing already a number of models whose predictive performance is satisfactory.

KW - Electricity load demand

KW - Modeling

KW - Neural networks

KW - Prediction

KW - Radial basis functions

UR - http://www.scopus.com/inward/record.url?scp=79955709618&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79955709618&partnerID=8YFLogxK

U2 - 10.3182/20090921-3-TR-3005.00073

DO - 10.3182/20090921-3-TR-3005.00073

M3 - Conference contribution

SN - 9783902661661

VL - 2

BT - IFAC Proceedings Volumes (IFAC-PapersOnline)

ER -