Please use this identifier to cite or link to this item: http://www.alice.cnptia.embrapa.br/alice/handle/doc/1184556
Full metadata record
DC FieldValueLanguage
dc.contributor.authorALVES, R. G.
dc.contributor.authorLIMA, F.
dc.contributor.authorGUEDES, I. M. R.
dc.contributor.authorGIMENEZ, S. P.
dc.date.accessioned2026-02-24T07:26:35Z-
dc.date.available2026-02-24T07:26:35Z-
dc.date.created2026-02-23
dc.date.issued2026
dc.identifier.citationIEEE Internet of Things Journal, v. 13, n. 5, p. 9192-9210, Mar. 2026.
dc.identifier.issn2327-4662
dc.identifier.urihttp://www.alice.cnptia.embrapa.br/alice/handle/doc/1184556-
dc.descriptionVertical farming offers a controlled environment for food production in regions where land scarcity and environmental stress are prevalent. This study presents a bio-inspired optimization strategy for refining the spectral composition of red, green, and blue (RGB) light from light-emitting diode (LED) to enhance crop performance. A genetic algorithm (GA) was employed to iteratively adjust spectral ratios in 2.5-day intervals over a single 25-day practical growth cycle. The algorithm employed selection, crossover, and mutation operators, targeting a weighted-sum fitness function based on key morphological traits, including fresh weight, height, width, and the number of leaves. An experimental trial was conducted under controlled conditions, with identical light intensity and photoperiod for both RGB treatments and a cold-white LED reference. The primary finding is that the optimization process successfully converged on a stable compo sition of approximately 67% red, 13% green, and 20% blue, which is consistent with prior studies on photosynthetic efficiency. This convergence validates the GA’s ability to autonomously discover a scientifically backed recipe from a neutral baseline. A final statistical analysis of the individual plant traits revealed a complex, multiobjective response, with plant height being the most statistically responsive parameter, while differences in the final biomass and leaf count were not statistically significant under the tested conditions. These findings demonstrate the potential of evolutionary algorithms for solving complex, multi objective optimization problems in vertical farming, supporting the development of adaptive lighting strategies.
dc.language.isoeng
dc.rightsopenAccess
dc.subjectFazenda vertical
dc.subjectAgricultura em ambiente controlado
dc.titleOptimizing RGB lighting for lettuce growth in vertical farms with bio-inspired optimization algorithm.
dc.typeArtigo de periódico
dc.subject.thesagroCultivo Protegido
dc.subject.thesagroIluminação Artificial
riaa.ainfo.id1184556
riaa.ainfo.lastupdate2026-02-23
dc.contributor.institutionRAFAEL GOMES ALVES, CENTRO UNIVERSITÁRIO DA FEI; FÁBIO LIMA, CENTRO UNIVERSITÁRIO DA FEI; ITALO MORAES ROCHA GUEDES, CNPH; SALVADOR PINILLOS GIMENEZ, PONTIFÍCIA UNIVERSIDADE CATÓLICA DE SÃO PAULO.
Appears in Collections:Artigo em periódico indexado (CNPH)

Files in This Item:
File SizeFormat 
CNPH-42387-AP.pdf4,07 MBAdobe PDFView/Open
Show simple item record

FacebookTwitterDeliciousLinkedInGoogle BookmarksMySpace