Mirikitani, Derrick Takeshi and Ouarbya, Lahcen. 2009. 'Modeling Dst with Recurrent EM Neural Networks'. In: International Conference on Artificial Neural Networks - ICANN 2009. Limassol, Cyprus 14 - 17 September 2009. [Conference or Workshop Item]

Preview

Text 5thDraft.pdf - Accepted Version Available under License Creative Commons Attribution Non-commercial. Download (395kB) | Preview

Abstract or Description

Recurrent Neural Networks have been used extensively for space weather forecasts of geomagnetospheric disturbances. One of the major drawbacks for reliable forecasts have been the use of training algorithms that are unable to account for model uncertainty and noise in data. We propose a probabilistic training algorithm based on the Expectation Maximization framework for parameterization of the model, which makes use of a forward filtering and backward smoothing Expectation step, and a Maximization step in which the model uncertainty and measurement noise estimates are computed. The inputs to the network are based on three parameters of the interplanetary magnetic field (IMF ), b z , b 2 , and b y 2 , along with the D st index. Through numerical experimentation it is shown that the proposed model allows for reliable forecasts and also outperforms other neural time series models trained with the Extended Kalman Filter, and gradient descent learning.

Item Type:	Conference or Workshop Item (Paper)
Identification Number (DOI):	https://doi.org/10.1007/978-3-642-04274-4_100
Additional Information:	The final authenticated version is available online at https://doi.org/10.1007/978-3-642-04274-4_100.
Keywords:	Solar Wind, Root Mean Square Error, Interplanetary Magnetic Field, Geomagnetic Storm, Extended Kalman Filter
Departments, Centres and Research Units:	Computing
Dates:	Date Event 14 September 2009 Published
Event Location:	Limassol, Cyprus
Date range:	14 - 17 September 2009
Item ID:	30980
Date Deposited:	11 Jan 2022 10:34
Last Modified:	11 Jan 2022 16:48
URI:	https://research.gold.ac.uk/id/eprint/30980

View statistics for this item...

Edit Record (login required)