You are in the accessibility menu

Please use this identifier to cite or link to this item: http://acervodigital.unesp.br/handle/11449/71968
Full metadata record
DC FieldValueLanguage
dc.contributor.authorRoveda, Sandra Regina Monteiro Masalskiene-
dc.contributor.authorBondança, Ana Paula Maia-
dc.contributor.authorSilva, João Guilherme Soares-
dc.contributor.authorRoveda, José Arnaldo Frutuoso-
dc.contributor.authorRosa, André Henrique-
dc.date.accessioned2014-05-27T11:24:50Z-
dc.date.accessioned2016-10-25T18:30:22Z-
dc.date.available2014-05-27T11:24:50Z-
dc.date.available2016-10-25T18:30:22Z-
dc.date.issued2010-11-25-
dc.identifierhttp://dx.doi.org/10.1109/FUZZY.2010.5584172-
dc.identifier.citation2010 IEEE World Congress on Computational Intelligence, WCCI 2010.-
dc.identifier.urihttp://hdl.handle.net/11449/71968-
dc.identifier.urihttp://acervodigital.unesp.br/handle/11449/71968-
dc.description.abstractDue to growing urbanization and industrialization, the environment is suffering from pollution of rivers, degradation of soils and deteriorated air quality. Quality indices appear to be useful to evaluate the conditions of these media. The aim of this study was the development of a water quality index using a fuzzy inference system, since such an approach has proved advantageous in addressing problems that are subjective by nature or for which the data are uncertain. The methodology employed was based on this inference system, and considered the nine water quality parameters employed by CETESB (Companhia de Tecnologia de Saneamento Ambiental, São Paulo, Brazil) to evaluate water quality. After assessment of the data using the index, a comparison was made with the WQI (Water Quality Index), which is used for the monitoring of various water bodies, including in the study region. The results obtained using the index developed on the basis of fuzzy inference were found to be more useful than those derived from the method currently used by CETESB, since losses and/or omissions concerning individual parameters were minimized. © 2010 IEEE.en
dc.language.isoeng-
dc.sourceScopus-
dc.subjectFuzzy inference systems-
dc.subjectInference systems-
dc.subjectQuality indices-
dc.subjectWater quality indexes-
dc.subjectWater quality parameters-
dc.subjectWaterbodies-
dc.subjectAir quality-
dc.subjectArtificial intelligence-
dc.subjectFish-
dc.subjectFuzzy inference-
dc.subjectFuzzy systems-
dc.subjectQuality assurance-
dc.subjectQuality control-
dc.subjectWater quality-
dc.subjectRiver pollution-
dc.titleDevelopment of a water quality index using a fuzzy logic: A case study for the sorocaba riveren
dc.typeoutro-
dc.contributor.institutionUniversidade Estadual Paulista (UNESP)-
dc.description.affiliationSao Paulo State University (UNESP), Sorocaba, SP 18087-180-
dc.description.affiliationUnespSao Paulo State University (UNESP), Sorocaba, SP 18087-180-
dc.identifier.doi10.1109/FUZZY.2010.5584172-
dc.identifier.wosWOS:000287453602073-
dc.rights.accessRightsAcesso restrito-
dc.relation.ispartof2010 IEEE World Congress on Computational Intelligence, WCCI 2010-
dc.identifier.scopus2-s2.0-78549261727-
dc.identifier.orcid0000-0002-2042-018Xpt
Appears in Collections:Artigos, TCCs, Teses e Dissertações da Unesp

There are no files associated with this item.
 

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.