By Topic

Use of neural networks to predict lightning at Kennedy Space Center

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$33 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

4 Author(s)
D. Frankel ; KTAADN Inc., Newton Centre, MA, USA ; I. Schiller ; J. S. Draper ; A. A. Barnes

The authors describe an effort to construct and train neural net architectures to generate spatio-temporal maps of predicted probabilities of lightning over the Cape Canaveral Air Force Station/Kennedy Space Center (KSC) complex. The goal is to improve the precision and accuracy of lightning prediction so that the launch commit criteria may be relaxed while maintaining acceptable safety margins. Comparisons are made with earlier methods based on correlation of wind divergence with the later occurrence of a lightning flash. Using KSC meteorological data for wind, electric field, and lightning strikes and the total area divergence product, a prediction skill level surpassing the prior state-of-the-art has been attained. Synthesis of different types of data is one of the strengths of the artificial neural systems approach. Network predictive power can be expected to increase when other inputs (e.g. temperature, humidity, satellite data, and radar returns) are included in the input array. Predictive power is also expected to improve when many days of data are used in training

Published in:

Neural Networks, 1991., IJCNN-91-Seattle International Joint Conference on  (Volume:i )

Date of Conference:

8-14 Jul 1991