Echo state network model for analyzing solar-wind effects on the AU and AL indices

Abstract. The properties of the auroral electrojets are examined on basis a trained machine-learning model. relationships between solar-wind parameters and AU AL indices modeled with an echo state network (ESN), kind recurrent neural network. We can consider this ESN model to represent nonlinear effects inputs electrojets. To identify electrojets, we obtain various synthetic data by using artificial ESN. analyses show that mainly controlled speed in addition Bz interplanetary magnetic field (IMF) as suggested literature. results also indicate density effect is emphasized when high IMF near zero. This suggests some density.