Neural network prediction of AE data

Neural network (NN) models were constructed to study prediction of the AE index. Both solar wind (vBz) and previous observed AE inputs were used to predict AE data for different numbers of time steps ahead. It seems that prediction of the original unsmoothed AE data is possible only for 10 time steps (25 min) ahead. The predicted time series of the AE data for 50 time steps (125 min) ahead was found to be dynamically different from the original time series. It is possible that the NN model cannot reproduce the turbulent part of the power spectrum of the AE data. However, when using smoothed AE data the prediction for 10 time steps ahead gave an NMSE of 0.0438, and a correlation coefficient of 0.98. The predictive ability of the model gradually decreased as the lead time of the predictions was increased, but was quite good up to predictions for 30 time steps (75 min) ahead.

[1]  Toshiki Tajima,et al.  Neural net forecasting for geomagnetic activity , 1993 .

[2]  P. Wintoft,et al.  Prediction of geomagnetic storms from solar wind data with the use of a neural network , 1994 .

[3]  J. Takalo,et al.  Correlation dimension and affinity of AE data and bicolored noise , 1993 .

[4]  Geoffrey E. Hinton,et al.  Learning internal representations by error propagation , 1986 .

[5]  A. Kolmogorov A refinement of previous hypotheses concerning the local structure of turbulence in a viscous incompressible fluid at high Reynolds number , 1962, Journal of Fluid Mechanics.

[6]  H. Lundstedt,et al.  Prediction of geomagnetic storms from solar wind data using Elman Recurrent Neural Networks , 1996 .

[7]  J. Timonen,et al.  Structure function as a tool in AE and Dst time series analysis , 1995 .

[8]  G. Consolini,et al.  Multifractal structure of auroral electrojet index data. , 1996, Physical review letters.

[9]  S. Akasofu Vortical distribution of sunspots , 1985 .

[10]  M. Lehtokangas,et al.  Time Series Analysis and Prediction of AE and Dst Data , 1996 .

[11]  J. Timonen,et al.  Characteristic time scale of auroral electrojet data , 1994 .

[12]  Daniel N. Baker,et al.  Magnetospheric Impulse Response for Many Levels of Geomagnetic Activity , 1985 .

[13]  D. Marquardt An Algorithm for Least-Squares Estimation of Nonlinear Parameters , 1963 .

[14]  Daniel N. Baker,et al.  The relation between the northern polar cap and auroral electrojet geomagnetic indices in the wintertime , 1996 .

[15]  E. Friis-christensen,et al.  Magnetic activity in the polar cap—A new index , 1988 .

[16]  D. Rumelhart,et al.  Predicting sunspots and exchange rates with connectionist networks , 1991 .