Publication:
A neural network approach to predict hurricane intensity in the North Atlantic basin

dc.contributor.advisor Ramírez-Beltrán, Nazario D.
dc.contributor.author Veneros-Castro, Anthony
dc.contributor.college College of Engineering en_US
dc.contributor.committee Hernández, William
dc.contributor.committee Gonzáles, Jorge E.
dc.contributor.committee Vasquez Espinosa, Ramon E.
dc.contributor.department Department of Industrial Engineering en_US
dc.contributor.representative Ierkic-Vidmar, Mario
dc.date.accessioned 2018-11-28T17:29:11Z
dc.date.available 2018-11-28T17:29:11Z
dc.date.issued 2004
dc.description.abstract Upper air information and artificial neural networks (ANN) are used to predict hurricane intensity in the North Atlantic basin. Competitive neural network is used to identify analog storms to the current hurricane. Once the analog hurricanes are identified the historical NCEP reanalysis data are used along of each storm tracks to develop a set of climatology, persistence and synoptic variables. Persistence, climatological and synoptic observations of the analog hurricanes and the current storm are combined to create a training set which is used to generate nonlinear transformations and an optimization algorithm is used to identify the variables that are best correlated with storm intensity. The best variables obtained from the optimization algorithm are used to train a neural network which used Levenberg-Marquardt algorithm as a learning rule. Preliminary results show that the proposed prediction scheme is a potential tool to increase the accuracy in predicting hurricane intensity. en_US
dc.description.abstract Redes neuronales artificiales e información atmosférica son utilizadas para predecir la intensidad de los huracanes en la parte norte del Océano Atlántico. Un proceso para identificar huracanes históricos que sean análogos al huracán actual es implementado usando una red neuronal competitiva. Una vez identificado los huracanes análogos, información histórica proveniente de NCEP es usada para crear una serie de variables sinópticas, climatologicas y persistentes a lo largo de la trayectoria de cada uno de los huracanes análogos. Estas variables son combinadas con las variables del huracán actual para crear un set de entrenamiento. Un algoritmo de optimización es implementado para identificar aquellas variables que tengan la mayor correlación con la intensidad. Estas luego son usadas para implementar una red neuronal que usa el algoritmo de Levenberg-Marquardt como regla de aprendizaje. Los resultados preliminares muestran que la metodología propuesta es una herramienta potencial en los esfuerzos por aumentar la precisión en la predicción de la intensidad de los huracanes. en_US
dc.description.graduationYear 2004 en_US
dc.identifier.uri https://hdl.handle.net/20.500.11801/1545
dc.language.iso English en_US
dc.rights.holder (c)2004 Anthony Veneros Castro en_US
dc.rights.license All rights reserved en_US
dc.subject Artificial neural networks en
dc.subject Hurricane intensity prediction en
dc.subject North Atlantic basin en
dc.subject Storm tracks en
dc.title A neural network approach to predict hurricane intensity in the North Atlantic basin en_US
dc.type Thesis en_US
dspace.entity.type Publication
thesis.degree.discipline Industrial Engineering en_US
thesis.degree.level M.S. en_US
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
ININ_VenerosCastroA_2004.pdf
Size:
1.94 MB
Format:
Adobe Portable Document Format
Description: