| Summary: | Geomagnetic storms signi.icantly impact technological systems such as satellites, navigation, and power grids, necessitating accurate classi.ication methods to mitigate these effects. Traditional methods often fall short in capturing the complex nature of these storms. This study investigates the use of statistical feature extraction techniques on the SYM-H index time series to enhance geomagnetic storm classi.ication. By extracting features such as mean, variance, skewness, kurtosis, variance intensity, and the number of peaks and troughs, the understanding of geomagnetic storm behaviour can be improved. In this study, we present that variance intensity and skewness are particularly effective in distinguishing between mild and severe geomagnetic storms, providing a more accurate classi.ication framework. ANOVA analysis was employed to reduce the feature set, con.irming the signi.icance of variance intensity and skewness for classi.ication purposes. The results indicate that severe storms exhibit higher variance intensity and more peaks and troughs, re.lecting their greater complexity compared to mild storms. These.indings suggest that advanced statistical feature extraction techniques, combined with rigorous feature selection through ANOVA, can signi.icantly enhance classi.ication models and resilience against geomagnetic disturbances, aiding in better preparedness and mitigation strategies for affected technological systems. © 2024 Institute of Physics Publishing. All rights reserved.
|
|---|