Multitime Scale Thunderstorm Monitoring System With Real-Time Warning and Imaging

As an important factor in fine thunderstorm detections, a multitime scale thunderstorm monitoring, warning, and imaging system is proposed in this article. The first computing phase involves a decomposition, classification, denoising, and reconstruction of the atmospheric electric field signals (AEFSs), collected by a self-made 3-D AEF apparatus, based on autocorrelation characteristics and fuzzy
$C$
-means (FCM). Second, FCM classifies the equally divided AEFS components. A scale reconstruction rule is put forward and applied to obtain multitime scale AEF branch data, according to the component temporal continuity in the same class. A corresponding scale correction strategy is then proposed. Thunderstorm point charge coordinate results are calculated by using branch data, and noise points contained in these results are removed. Finally, the curve fitting of denoised coordinate results is performed to image the point charge moving path. Empirical results confirm that the proposed system effectively warns and images thunderstorms, as well as provides a valid reference for multiscale thunderstorm monitoring.