Indian Journal of Science and Technology

Open Access Open Access  Restricted Access Subscription Access

Analysis of the Reflectivity in Meteorological Radars using Data Mining and Neural Networks


Affiliations
1 Department of Electronic Engineering , Engineering Faculty, Distrital Francisco Jose de Caldas University, Bogota D.C., Colombia
 

Objectives: The aim of this work is show the analysis of the data measured by weather radar used in data mining and fuzzy logic. Methods/Analysis: A decoding of the data measured by the meteorological radar was made, which was encrypted, then an analysis of this data was made using neural networks that are trained with 10 and 20 neurons, in each case the effectiveness of each one is checked. Findings: The results showed that neural networks are an excellent tool that allows eliminate erroneous information and then normalize it to the scale used according to the standard. Improvements: This knowledge is essential for the aviation industry to operate properly and without risks for passengers, crew and aircraft, it is also important to anticipate and/or avoid, if possible, catastrophes generated by weather events related to rainfall.

Keywords

Data Mining, Neural Networks, Polarimetric Variables, Reflectivity, Weather Radar
User

Abstract Views: 209

PDF Views: 0




  • Analysis of the Reflectivity in Meteorological Radars using Data Mining and Neural Networks

Abstract Views: 209  |  PDF Views: 0

Authors

Julian R. Camargo L.
Department of Electronic Engineering , Engineering Faculty, Distrital Francisco Jose de Caldas University, Bogota D.C., Colombia
Ernesto Gomez Vargas
Department of Electronic Engineering , Engineering Faculty, Distrital Francisco Jose de Caldas University, Bogota D.C., Colombia
Cesar A. Perdomo Ch.
Department of Electronic Engineering , Engineering Faculty, Distrital Francisco Jose de Caldas University, Bogota D.C., Colombia

Abstract


Objectives: The aim of this work is show the analysis of the data measured by weather radar used in data mining and fuzzy logic. Methods/Analysis: A decoding of the data measured by the meteorological radar was made, which was encrypted, then an analysis of this data was made using neural networks that are trained with 10 and 20 neurons, in each case the effectiveness of each one is checked. Findings: The results showed that neural networks are an excellent tool that allows eliminate erroneous information and then normalize it to the scale used according to the standard. Improvements: This knowledge is essential for the aviation industry to operate properly and without risks for passengers, crew and aircraft, it is also important to anticipate and/or avoid, if possible, catastrophes generated by weather events related to rainfall.

Keywords


Data Mining, Neural Networks, Polarimetric Variables, Reflectivity, Weather Radar



DOI: https://doi.org/10.17485/ijst%2F2018%2Fv11i19%2F174518