Modeling Geomagnetospheric Disturbances with Sequential Bayesian Recurrent Neural Networks

Ouarbya, Lahcen and Mirikitani, Derrick T.. 2009. 'Modeling Geomagnetospheric Disturbances with Sequential Bayesian Recurrent Neural Networks'. In: 16th International Conference on Neural Information Processing - ICONIP 2009. Bangkok, Thailand 1 - 5 December 2009. [Conference or Workshop Item]

No full text available

Abstract or Description

Sequential Bayesian trained recurrent neural networks ( RNNs ) have not yet been considered for modeling the dynamics of magnetospheric plasma. We provide a discussion of the state-space modeling framework and an overview of sequential Bayesian estimation. Three nonlinear filters are then proposed for online RNN parameter estimation, which include the extended Kalman filter, the unscented Kalman filter, and the ensemble Kalman filter. The exogenous inputs to the RNNs consist of three parameters, b z , b 2, and b^2_y, where b , b z , and by represent the magnitude, the southward and azimuthal components of the interplanetary magnetic field ( IMF ) respectively. The three models are compared to a model used in operational forecasts on a severe double storm that has so far been difficult to forecast. It is shown that some of the proposed models significantly outperform the current state of the art.

Item Type:

Conference or Workshop Item (Paper)

Identification Number (DOI):


Geomagnetic Storms, Recurrent Neural Networks, Filtering

Departments, Centres and Research Units:



1 December 2009Published

Event Location:

Bangkok, Thailand

Date range:

1 - 5 December 2009

Item ID:


Date Deposited:

12 Apr 2012 12:12

Last Modified:

11 Jan 2022 10:56


Edit Record Edit Record (login required)