Close

1. Identity statement
Reference TypeConference Paper (Conference Proceedings)
Sitemtc-m16b.sid.inpe.br
Holder Codeisadg {BR SPINPE} ibi 8JMKD3MGPCW/3DT298S
Repositorycptec.inpe.br/adm_conf/2005/10.31.20.53
Last Update2006:04.20.12.18.06 (UTC) administrator
Metadata Repositorycptec.inpe.br/adm_conf/2005/10.31.20.53.04
Metadata Last Update2022:03.26.18.01.22 (UTC) administrator
Secondary KeyINPE-13855-PRE/9037
Citation KeyGuarnieriPereChou:2006:SoRaFo
TitleSolar radiation forecast using artificial neural networks in south Brazil
FormatCD-ROM; On-line.
Year2006
Access Date2024, June 13
Secondary TypePRE CI
Number of Files1
Size3150 KiB
2. Context
Author1 Guarnieri, Ricardo André
2 Pereira, Enio Bueno
3 Chou, Sin Chan
Resume Identifier1
2 8JMKD3MGP5W/3C9JH2E
Group1 RSU-INPE-MCT-BR
2 DMA-INPE-MCT-BR
3 DMD-INPE-MCT-BR
Affiliation1 Instituto Nacional de Pesquisas Espaciais (INPE)
2 Instituto Nacional de Pesquisas Espaciais (INPE)
3 Instituto Nacional de Pesquisas Espaciais (INPE)
Author e-Mail Address1 ricardog@cptec.inpe.br
2 eniobp@cptec.inpe.br
3 chou@cptec.inpe.br
EditorVera, Carolina
Nobre, Carlos
e-Mail Addressricardog@cptec.inpe.br
Conference NameInternational Conference on Southern Hemisphere Meteorology and Oceanography, 8 (ICSHMO).
Conference LocationFoz do Iguaçu
Date24-28 Apr. 2006
PublisherAmerican Meteorological Society (AMS)
Publisher City45 Beacon Hill Road, Boston, MA, USA
Pages1777-1785
Book TitleProceedings
Tertiary TypePoster
OrganizationAmerican Meteorological Society (AMS)
History (UTC)2005-10-31 20:53:04 :: ricardog@cptec.inpe.br -> administrator ::
2005-11-11 21:54:28 :: administrator -> adm_conf ::
2005-12-15 23:54:02 :: adm_conf -> ricardog@cptec.inpe.br ::
2006-03-28 14:00:26 :: ricardog@cptec.inpe.br -> adm_conf ::
2006-03-29 21:02:41 :: adm_conf -> ricardog@cptec.inpe.br ::
2006-03-29 23:24:56 :: ricardog@cptec.inpe.br -> administrator ::
2006-04-18 21:15:36 :: administrator -> lise@dpi.inpe.br ::
2010-12-28 12:36:45 :: lise@dpi.inpe.br -> administrator ::
2010-12-29 15:58:31 :: administrator -> lise@dpi.inpe.br :: 2006
2010-12-29 16:06:10 :: lise@dpi.inpe.br -> administrator :: 2006
2010-12-29 18:54:17 :: administrator -> banon :: 2006
2011-01-02 17:15:09 :: banon -> administrator :: 2006
2022-03-26 18:01:22 :: administrator -> :: 2006
3. Content and structure
Is the master or a copy?is the master
Content Stagecompleted
Transferable1
Keywordssolar radiation forecast
solar energy
artificial neural networks
Eta model
SONDA project
AbstractIncident solar radiation studies have several applications. Besides other uses we can emphasize agriculture, illumination and heating of buildings and residences, and meteorological research. Nowadays, an important application is to supply information and reliable data that support the utilization of solar radiation as a renewable energy source, with reduced impacts on the environment. Forecasting the incident solar radiation allows us to know, in advance, the resources availability. This information is important to improve the use of different energy sources in cogeneration systems. As an attempt to get a better predictability for the incident solar radiation at the surface, Artificial Neural Networks (ANNs) were used. Meteorological data provided by weather forecast models, representing the conditions of the future atmosphere, can be employed as inputs for ANNs. With a correct training, these ANNs can provide estimates of solar radiation, since the incidence of radiation at surface is totally dependent on the atmospheric conditions. In this work we have trained ANNs (Multilayer Perceptron ANNs) using as inputs some meteorological data generated by the Eta-operational Model run in the Brazilian Centro de Previsão de Tempo e Estudos Climáticos (CPTEC/INPE). Further, we have supplied the calculated solar radiation incident in the top of atmosphere, as an additional input. We have also used measurements of solar radiation incident at surface, as the targets during the training-phase. These radiation measurements were supplied by the radiometric stations of the SONDA project (Sistema de Organização Nacional de Dados Ambientais para o Setor de Energia). During the training-phase, input and target data sets were presented to the ANN. The ANN establishes mathematical relations between the input variables, with the goal of reducing the error between the calculated output and the target. The training technique used in this work was Resilient Backpropagation. After training, a new set of data was used to test the ANN performance for solar radiation forecasting. In this test-phase the solar radiation calculated by ANNs was compared with the solar radiation measurements. A comparison between radiation forecasts directly provided by the Eta model and solar radiation measurements was also accomplished, in order to verify whether there is an increase in predictability with the use of ANNs. The forecasts were tested for two SONDA-project stations: Florianópolis (lat. 27.6ºS, long. 48.5ºW) and São Matinho da Serra (lat. 29.4ºS, long. 53.8ºW). Solar radiation forecasts from ANNs presented higher correlation coefficients and lower errors than the Eta model output for shortwave radiation on ground. The well-know bias observed in solar radiation forecasts by the Eta model was removed by the use of ANNs. The improvement in RMSE obtained with ANNs over the Eta model was higher than 30%, estimated with a skill-score.
AreaMET
TypeWeather analysis and forecasting
Arrangement 1urlib.net > BDMCI > Fonds > Produção anterior à 2021 > CRCRS > Solar radiation forecast...
Arrangement 2urlib.net > BDMCI > Fonds > Produção anterior à 2021 > DIDMD > Solar radiation forecast...
Arrangement 3urlib.net > BDMCI > Fonds > Produção até 2016 > DMA > Solar radiation forecast...
doc Directory Contentaccess
source Directory Content
ExtAbstract8ICSHMONew.doc 29/03/2006 20:24 706.5 KiB 
agreement Directory Contentthere are no files
4. Conditions of access and use
data URLhttp://urlib.net/ibi/cptec.inpe.br/adm_conf/2005/10.31.20.53
zipped data URLhttp://urlib.net/zip/cptec.inpe.br/adm_conf/2005/10.31.20.53
Languageen
Target File1777-1786.pdf
User Groupadministrator
ricardog@cptec.inpe.br
administrator
banon
Visibilityshown
Copy HolderSID/SCD
Read Permissionallow from all
5. Allied materials
Next Higher Units8JMKD3MGPCW/3EUFCFP
8JMKD3MGPCW/43SKC35
8JMKD3MGPCW/46JKC45
Citing Item Listsid.inpe.br/bibdigital/2021/01.01.17.20 2
sid.inpe.br/mtc-m21/2012/07.13.14.45.21 2
sid.inpe.br/bibdigital/2022/03.26.17.47 1
Host Collectioncptec.inpe.br/nobre/2005/06.02.21.14
cptec.inpe.br/walmeida/2003/04.25.17.12
6. Notes
Mark1
Empty Fieldsarchivingpolicy archivist callnumber contenttype copyright creatorhistory descriptionlevel dissemination documentstage doi edition identifier isbn issn label lineage mirrorrepository nextedition notes numberofvolumes orcid parameterlist parentrepositories previousedition previouslowerunit progress project readergroup rightsholder schedulinginformation secondarydate secondarymark serieseditor session shorttitle sponsor subject tertiarymark url versiontype volume


Close