A Neural Network for Damaging Wind Prediction

Abstract

A neural network is developed to diagnose which circulations detected by the National Severe Storms Laboratory's (NSSL) Mesocyclone Detection Algorithm (MDA) yield damaging wind. In particular, 23 variables characterizing the circulations are selected to be used as the input nodes of a feed-forward, supervised neural network. The outputs of the network… (More)

Topics

  • Presentations referencing similar topics