System based on fuzzy logic for diagnosis of water quality for the cultivation of Nile tilápia

• Autor(a) principal: Castro Junior, Sergio Luis de
• Data de Publicação: 2022
• Outros Autores: Lamarca, Daniel Sá Freire, Kraetzer, Thiago Lorente, Balthazar, Glauber da Rocha, Caneppele, Fernando de Lima
• Tipo de documento: Artigo
• Idioma: por
• Título da fonte: Research, Society and Development
• Texto Completo: https://rsdjournal.org/index.php/rsd/article/view/26933
• topic: Fish farming; Aquaculture; Mathematical modeling.; Piscicultura; Acuicultura; Modelo matematico.; Piscicultura; Aquicultura; Modelagem matemática.

Aquaculture is characterized as a system for the production of plants and animals in a controlled aquatic environment, being generally applied in a sustainable way on farms for presenting a fast economic return. Accurate and quick information about water quality is essential to guarantee both the survival of fish and their correct feed conversion. In this context, the objective of this article was to develop a decision support system, based on the theory of fuzzy sets, for the evaluation of water quality conditions and their influence on the productivity and ambience of Nile tilapia. The execution of this work was divided into three stages: a) bibliographic survey of the water quality parameters, considering their influence on the productive performance of the fish; b) the use of the results of the previous phase, together with contributions from experts, for the development of a fuzzy inference system for diagnosing water quality in breeding tanks; c) use the fuzzy system previously elaborated for analysis of a database representing a commercial tank. The results obtained in the present work were shown to be adequate for the classification of water quality for Nile tilapia, using the fuzzy modeling process. The classifications determined by the fuzzy model were similar to the classification given by the Boolean model. However, the divergences found between the models are relevant as small oscillations observed in the input variables (temperature and pH) do not indicate sudden changes in the model's output variable (water quality), in the case of the fuzzy model.