Sistemas fuzzy para a estimativa da produtividade no plantio do arroz
Carregando...
Data
2022-10-13
Autores
Presumido Junior, Marcio
Título da Revista
ISSN da Revista
Título de Volume
Editor
Universidade Estadual Paulista (Unesp)
Resumo
O arroz irrigado (Oryza sativa L.) é a principal fonte alimentar e energética para mais da metade da população mundial e desempenha papel estratégico em níveis econômicos e sociais. É o segundo cereal mais produzido no mundo com 758,8 milhões de toneladas colhidas, instaladas em 162,6 milhões de hectares. No Brasil, destacam-se duas técnicas de cultivares de arroz: o sistema irrigado que demanda manejo, preparo do solo, adubação, sementes e alto custo de investimento no preparo da terra e insumos e, o sistema sequeiro, que necessita de poucos insumos constituindo baixo custo de investimento inicial. Este estudo apresentou como objetivo geral o desenvolvimento de uma ferramenta que permite a rápida avaliação de produtividade da cultura do arroz. A ferramenta foi desenvolvida apoiada na Lógica Fuzzy (ou Lógica Difusa), com um sistema baseado em regras. O presente estudo é composto por duas partes: uma revisão bibliográfica e um estudo de campo, por meio da aplicação de questionários e do uso da modelagem baseada em lógica fuzzy. Como objetivos específicos, o referido estudo apresentou um panorama geral sobre os sistemas de preparo do solo e de plantio do arroz no Brasil; desenvolveu sistemas fuzzy para estimar a produtividade do arroz com base na produtividade de grãos e estimar a produtividade do arroz com base na altura de plantas; espiguetas férteis por panícula; espiguetas estéreis por panícula; colmos e panículas m2 ; perfilhamento útil e massa de 100 grãos. Como principal resultado, observa-se que os sistemas fuzzy conseguiram definir a produtividade do arroz em diferentes dosagens, e ainda que cada fertilizante tenha suas qualidades em diferentes dosagens. Sendo assim, o conhecimento do produtor atinge diretamente nos seus resultados de sua cultura
Irrigated rice (Oryza sativa L.) is the main source of food and energy for more than half of the world's population and plays a strategic role in economic and social levels. It is the second most produced cereal in the world with 758.8 million tons harvested on 162.6 million hectares. In Brazil, there are two techniques of rice cultivars: the irrigated system, which requires management, soil preparation, fertilization, seeds, and high cost of investment in land preparation and inputs, and the rainfed system, which requires few inputs constituting a low cost of initial investment. The general objective of this study was to develop a tool that allows a quick assessment of the rice crop productivity. The tool was developed based on Fuzzy Logic, with a rule-based system. This study consists of two parts: a literature review and a field study, through the application of questionnaires and the use of fuzzy logic-based modeling. As specific objectives, this study presented an overview of the systems of soil preparation and planting of rice in Brazil; developed fuzzy systems to estimate the productivity of rice based on grain yield and estimate the productivity of rice based on plant height; fertile spikelets per panicle; sterile spikelets per panicle; stalks and panicles m2 ; useful tillering and mass of 100 grains. As a main result, it is observed that the fuzzy systems were able to define the rice yield at different dosages, and also that each fertilizer has its qualities at different dosages. Thus, the producer's knowledge directly affects the results of his crop.
Irrigated rice (Oryza sativa L.) is the main source of food and energy for more than half of the world's population and plays a strategic role in economic and social levels. It is the second most produced cereal in the world with 758.8 million tons harvested on 162.6 million hectares. In Brazil, there are two techniques of rice cultivars: the irrigated system, which requires management, soil preparation, fertilization, seeds, and high cost of investment in land preparation and inputs, and the rainfed system, which requires few inputs constituting a low cost of initial investment. The general objective of this study was to develop a tool that allows a quick assessment of the rice crop productivity. The tool was developed based on Fuzzy Logic, with a rule-based system. This study consists of two parts: a literature review and a field study, through the application of questionnaires and the use of fuzzy logic-based modeling. As specific objectives, this study presented an overview of the systems of soil preparation and planting of rice in Brazil; developed fuzzy systems to estimate the productivity of rice based on grain yield and estimate the productivity of rice based on plant height; fertile spikelets per panicle; sterile spikelets per panicle; stalks and panicles m2 ; useful tillering and mass of 100 grains. As a main result, it is observed that the fuzzy systems were able to define the rice yield at different dosages, and also that each fertilizer has its qualities at different dosages. Thus, the producer's knowledge directly affects the results of his crop.
Descrição
Palavras-chave
Oryza sativa L, Agronegócio, Sistema de Inferência