<?xml version="1.0" encoding="ISO-8859-1"?>
<metadatalist>
	<metadata ReferenceType="Thesis">
		<site>mtc-m21c.sid.inpe.br 806</site>
		<holdercode>{isadg {BR SPINPE} ibi 8JMKD3MGPCW/3DT298S}</holdercode>
		<identifier>8JMKD3MGP3W34R/44JMNDL</identifier>
		<repository>sid.inpe.br/mtc-m21c/2021/04.29.03.12</repository>
		<lastupdate>2021:09.27.15.11.49 urlib.net/www/2017/11.22.19.04 simone</lastupdate>
		<metadatarepository>sid.inpe.br/mtc-m21c/2021/04.29.03.12.04</metadatarepository>
		<metadatalastupdate>2022:04.03.23.15.44 sid.inpe.br/bibdigital@80/2006/04.07.15.50 administrator</metadatalastupdate>
		<secondarykey>INPE-18470-TDI/3118</secondarykey>
		<citationkey>Almeida:2021:EsCaAm</citationkey>
		<title>Uso de redes neurais para previsão de descargas elétricas nuvem-solo a partir de dados de radar: um estudo de caso na Amazônia</title>
		<alternatetitle>Use of neural networks to predict cloud-to-ground lightning from radar data: a case study in the Amazon</alternatetitle>
		<course>CAP-COMP-DIPGR-INPE-MCTI-GOV-BR</course>
		<year>2021</year>
		<date>2021-05-28</date>
		<thesistype>Dissertação (Mestrado em Computação Aplicada)</thesistype>
		<secondarytype>TDI</secondarytype>
		<numberofpages>208</numberofpages>
		<numberoffiles>1</numberoffiles>
		<size>44233 KiB</size>
		<author>Almeida, Adriano Pereira,</author>
		<committee>Santos, Rafael Duarte Coelho (presidente),</committee>
		<committee>Calheiros, Alan James Peixoto (orientador),</committee>
		<committee>Shiguemori, Elcio Hideiti,</committee>
		<committee>Rodriguez, Carlos Augusto Morales,</committee>
		<e-mailaddress>adriano.almeida@inpe.br</e-mailaddress>
		<university>Instituto Nacional de Pesquisas Espaciais (INPE)</university>
		<city>São José dos Campos</city>
		<transferableflag>1</transferableflag>
		<keywords>tempestades severas, raios, RADAR, análise exploratória de dados, inteligência artificial, severe storms, lightnings, exploratory data analysis, artificial intelligence.</keywords>
		<abstract>A região central da Bacia Amazônica é caracterizada por ter uma vegetação densa e condições atmosféricas favoráveis para o desenvolvimento de tempestades severas e com alta incidência de raios. Diante disso, foi desenvolvido neste trabalho um estudo de algumas parametrizações através de técnicas de aprendizado de máquina para prover subsídios para o desenvolvimento de sistemas de previsão de descargas elétricas a curto prazo, levando em consideração as condições locais dessa região. Essas parametrizações partiram de conclusões obtidas na análise exploratória dos dados de radar meteorológico e da seleção de variáveis por meio de classificações com a árvore de decisão. Além disso, foram criados modelos baseados na rede neural perceptron multicamadas (MLP) com diversas configurações de entrada e variações dos seus principais hiper-parâmetros, que tiveram como intuito avaliar a sensibilidade das variáveis em diferentes intervalos de tempos anteriores à previsão. As previsões foram feitas para 12 minutos à frente, utilizando os valores das variáveis em instantes anteriores à previsão: t-1 (12 minutos); t-2 (24 minutos); e t-3 (36 minutos). Embora a maioria das variáveis meteorológicas utilizadas possuam dependências temporais mais restritas, os modelos que utilizaram os três instantes de tempo, tiveram melhor desempenho. Junto com a degradação dos dados de entrada e informações sobre as características dos perfis de refletividade radar foi possível definir critérios que podem ajudar na melhorias das previsões de raios via redes neurais. ABTSRACT: Lightning associated with severe storms can cause several socio-economic problems. Due to its high convective activity, predominantly equatorial location, and large territorial extension, Brazil is one of the countries where the highest incidence of lightning strikes occurs globally. Studies show that the most significant cause of human fatalities related to storms in Brazil is associated with lightning and that a large part of them appear in rural areas. The material damages caused by these events are also very recurrent. Therefore, the interest in creating forecasting systems for this natural event is of paramount importance. However, several challenges are associated with the creation of lightning forecasting systems. Sometimes these challenges are related to the characteristics of the event itself, such as its high variability and its relationship with complex physical processes that are difficult to model computationally. The local factors that can influence the development of electrically active storms should also be considered. The central region of the Amazon Basin is characterized by dense vegetation and favorable atmospheric conditions for the development of severe storms and a high incidence of lightning. Therefore, a study of some parameterizations through machine learning techniques was developed in this work to provide subsidies for the development of lightning forecasting lightning in the short-term systems, considering the local conditions of the central region of the Amazon Basin. These parameterizations started from conclusions obtained in the exploratory analysis of meteorological radar data and the selection of variables through classifications with the decision tree. In addition, models based on the neural network multilayer layer perceptron (MLP) with different input configurations were created to evaluate the sensitivity of the variables at different time intervals before the forecast. The forecasts were made for 12 minutes ahead, using the previous times values of the variables: t-1 (12 minutes); t-2 (24 minutes); and t-3 (36 minutes). Although most meteorological variables used have more restricted temporal dependencies, the models that used three previous time instants performed better. Along with the degradation of the input data and information on the characteristics of the radar reflectivity profiles, it was possible to define criteria that can help to improve the lightning forecasts via neural networks.</abstract>
		<area>COMP</area>
		<language>pt</language>
		<targetfile>publicacao.pdf</targetfile>
		<usergroup>adriano.almeida@inpe.br</usergroup>
		<usergroup>pubtc@inpe.br</usergroup>
		<usergroup>simone</usergroup>
		<visibility>shown</visibility>
		<copyright>urlib.net/www/2012/11.12.15.10</copyright>
		<readpermission>allow from all</readpermission>
		<documentstage>not transferred</documentstage>
		<mirrorrepository>urlib.net/www/2017/11.22.19.04.03</mirrorrepository>
		<nexthigherunit>8JMKD3MGPCW/3F2PHGS</nexthigherunit>
		<nexthigherunit>8JMKD3MGPCW/46KUES5</nexthigherunit>
		<citingitemlist>sid.inpe.br/bibdigital/2013/10.12.22.16 5</citingitemlist>
		<hostcollection>urlib.net/www/2017/11.22.19.04</hostcollection>
		<agreement>autorizacao.pdf .htaccess .htaccess2</agreement>
		<lasthostcollection>urlib.net/www/2017/11.22.19.04</lasthostcollection>
		<supervisor>Calheiros, Alan James Peixoto,</supervisor>
		<url>http://mtc-m21c.sid.inpe.br/rep-/sid.inpe.br/mtc-m21c/2021/04.29.03.12</url>
	</metadata>
</metadatalist>