| 1. Identity statement | |
| Reference Type | Conference Paper (Conference Proceedings) |
| Site | mtc-m16b.sid.inpe.br |
| Holder Code | isadg {BR SPINPE} ibi 8JMKD3MGPCW/3DT298S |
| Repository | sid.inpe.br/mtc-m15@80/2006/08.04.12.50 (restricted access) |
| Last Update | 2006:09.18.14.24.54 (UTC) administrator |
| Metadata Repository | sid.inpe.br/mtc-m15@80/2006/08.04.12.50.31 |
| Metadata Last Update | 2022:03.26.18.03.53 (UTC) administrator |
| Secondary Key | INPE-14193-PRE/9311 |
| Citation Key | GuarnieriPereChan:2006:NeNeAk |
| Title | Neural networks aks applied to solar resources forecast  |
| Format | Papel |
| Year | 2006 |
| Secondary Date | 20060918 |
| Access Date | 2025, May 09 |
| Secondary Type | PRE CI |
| Number of Files | 1 |
| Size | 3578 KiB |
| 2. Context | |
| Author | 1 Guarnieri, Ricardo André 2 Pereira, Enio Bueno 3 Chan, Sin Chou |
| Resume Identifier | 1 2 8JMKD3MGP5W/3C9JH2E |
| Group | 1 DMA-INPE-MCT-BR 2 DMA-INPE-MCT-BR 3 DMD-INPE-MCT-BR |
| Affiliation | 1 Instituto Nacional de Pesquisas Espaciais (INPE), Centro de Previsão de Tempo e Estudos Climáticos (CPTEC) 2 Instituto Nacional de Pesquisas Espaciais (INPE), Centro de Previsão de Tempo e Estudos Climáticos (CPTEC) 3 Instituto Nacional de Pesquisas Espaciais (INPE), Centro de Previsão de Tempo e Estudos Climáticos (CPTEC) |
| e-Mail Address | atus@cptec.inpe.br |
| Conference Name | EGU General Assembly. |
| Conference Location | Vienna, Austria |
| Date | Apr. 02-07 |
| Book Title | Proceedings |
| Tertiary Type | Poster Session |
| Organization | EGU |
| History (UTC) | 2006-11-13 18:27:14 :: estagiario -> administrator :: 2008-06-25 01:34:30 :: administrator -> estagiario :: 2010-05-11 16:56:27 :: estagiario -> administrator :: 2022-03-26 18:03:53 :: administrator -> marciana :: 2006 |
| 3. Content and structure | |
| Is the master or a copy? | is the master |
| Content Stage | completed |
| Transferable | 1 |
| Content Type | External Contribution |
| Keywords | solar energy subtropical countries solar irradiance agriculture meteorological artificial neural |
| Abstract | Solar energy is one of the most important sources of energy that should be increasingly inserted into the energy matrixes of a large amount of countries, chiefly in tropical and subtropical countries. Although some countries are already partially supplying their energy demands using solar energy, mainly because the reduced environmental damage and also due to the fact that it is a renewable source, this number is yet very reduced. There is a worldwide demand from the energy sector for accurate forecasts of solar energy (and wind as well) so as to manage co-generation systems and energy dispatch in transmission lines. Solar irradiance forecast is also important for agriculture, meteorological studies, and other human activities. However, forecasting solar irradiation, even one day in advance, is a complicated task. Part of the difficulties arises from the solar radiation dependence on clouds and meteorological conditions which intrinsically involves non-linear processes. Other difficulties are related with the inaccuracy of weather forecasts by numerical models, due to the complexity of the non-linear processes involved, and also due to the difficulties of achieving optical properties for the future state of the atmosphere. The Eta model is the current operational mesoscale weather forecast model in the Brazilian Center of Weather Forecast and Climate Studies (CPTEC/INPE). The model output for shortwave radiation incidence at the Earth surface presents a considerable bias, probably related to deficiencies in the parameterization of the radiation scheme. Aiming to obtain a more accurate and reliable solar radiation forecast, artificial neural networks (ANNs) have been used. These ANNs (multilayer perceptron backpropagation training) have been trained with former Eta forecasts outputs, calculated solar radiation at the top of atmosphere, and solar radiation measurements from two ground-based stations of SONDA/INPE Project: Florianópolis and São Martinho da Serra. The main purpose of this work is to present and evaluate the performance of ANNs with the goal of forecasting incident solar radiation. It will be presented some improvements obtained with the use of this tool over the forecast of solar radiation provided directly by the Eta model. Some results have shown that ANNs improve slightly the prediction, reducing bias and the root mean square error (RMSE), and increasing the correlation coefficient between forecasts and observations. ANNs forecasts have shown an improvement of about 30% (RMSE reduction) over Eta solar radiation outputs. In conclusion, with this methodology (ANNs based on Eta outputs) we are able to produce better solar radiation forecasts that can be used by the national energy sector for several energy-related studies from renewable energy supply to electric energy distribution. |
| Area | MET |
| Arrangement 1 | urlib.net > BDMCI > Fonds > Produção anterior à 2021 > DIDMD > Neural networks aks... |
| Arrangement 2 | urlib.net > BDMCI > Fonds > Produção até 2016 > DMA > Neural networks aks... |
| doc Directory Content | access |
| source Directory Content | there are no files |
| agreement Directory Content | there are no files |
| 4. Conditions of access and use | |
| Language | en |
| Target File | neural.guarnieri.EGU.pdf |
| User Group | administrator estagiario |
| Visibility | shown |
| Copy Holder | SID/SCD |
| Read Permission | deny from all and allow from sem and allow from restrição |
| Update Permission | transferred to estagiario |
| 5. Allied materials | |
| Next Higher Units | 8JMKD3MGPCW/43SKC35 8JMKD3MGPCW/46JKC45 |
| Host Collection | cptec.inpe.br/walmeida/2003/04.25.17.12 |
| 6. Notes | |
| Notes | Publicado o Abstracts em Geophysical Research Absctracts , 8 , p.00733 , SREF-ID: 1607-7962/gra/EGU06-A-00733, EGU 2006 |
| Empty Fields | archivingpolicy archivist callnumber copyright creatorhistory descriptionlevel dissemination doi edition editor electronicmailaddress identifier isbn issn label lineage mark mirrorrepository nextedition numberofvolumes orcid pages parameterlist parentrepositories previousedition previouslowerunit progress project publisher publisheraddress readergroup rightsholder schedulinginformation secondarymark serieseditor session shorttitle sponsor subject tertiarymark type url versiontype volume |
| 7. Description control | |
| e-Mail (login) | marciana |
| update | |