Meteorological time series forecasting based on MLP modelling using heterogeneous transfer functions

Abstract : In this paper, we propose to study four meteorological and seasonal time series coupled with a multi-layer perceptron (MLP) modeling. We chose to combine two transfer functions for the nodes of the hidden layer, and to use a temporal indicator (time index as input) in order to take into account the seasonal aspect of the studied time series. The results of the prediction concern two years of measurements and the learning step, eight independent years. We show that this methodology can improve the accuracy of meteorological data estimation compared to a classical MLP modelling with a homogenous transfer function.
Document type :
Conference papers
Complete list of metadatas

Cited literature [10 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-00984948
Contributor : Cyril Voyant <>
Submitted on : Tuesday, April 29, 2014 - 6:17:28 AM
Last modification on : Thursday, January 11, 2018 - 6:16:28 AM
Long-term archiving on : Tuesday, July 29, 2014 - 11:05:30 AM

File

ICMSQUARE2014.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-00984948, version 1
  • ARXIV : 1404.7255

Collections

Citation

Cyril Voyant, Marie Laure Nivet, Christophe Paoli, Marc Muselli, Gilles Notton. Meteorological time series forecasting based on MLP modelling using heterogeneous transfer functions. International Conference on Mathematical Modeling in Physical Sciences 2014, Aug 2014, Madrid, Spain. ⟨hal-00984948⟩

Share

Metrics

Record views

335

Files downloads

183