Analysis of the use of artificial intelligence applied to coagulant dosage on water treatment

• Grant number: 22/03675-8
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: June 01, 2022
• End date: May 31, 2023
• Field of knowledge: Physical Sciences and Mathematics - Probability and Statistics - Statistics
• Principal Investigator: Adriano Bressane
• Grantee: Ana Paula Garcia Goulart
• Host Institution: Instituto de Ciência e Tecnologia (ICT). Universidade Estadual Paulista (UNESP). Campus de São José dos Campos. São José dos Campos , SP, Brazil

Abstract

Coagulation is among the main operations of a water treatment process. Traditionally, coagulant dosage is based on the jar test. However, the performance of this test is subject to operator experience, involves operational costs, and, as it is reactive, it does not allow real-time prediction, which makes it inappropriate in conditions of substantial variation in raw water quality. In this context, empirical models based on artificial intelligence have been evaluated as a promising alternative. Despite advances in the scientific field, there is a gap to which this project seeks to contribute: a comparative analysis of different algorithms applicable to the regression task, in order to determine which may represent the most promising approach for coagulant dosage purposes. As a research hypothesis, it is considered that a fuzzy system should provide the best performance, due to its ability to deal with uncertainties, when compared to other artificial intelligence alternatives. (AU)

Scientific publications

(References retrieved automatically from Web of Science and SciELO through information on FAPESP grants and their corresponding numbers as mentioned in the publications by the authors)

BRESSANE, ADRIANO; GOULART, ANA PAULA GARCIA; MELO, CARRIE PERES; GOMES, ISADORA GURJON; LOUREIRO, ANNA ISABEL SILVA; NEGRI, ROGERIO GALANTE; MORUZZI, RODRIGO; DOS REIS, ADRIANO GONCALVES; FORMIGA, JORGE KENNETY SILVA; DA SILVA, GUSTAVO HENRIQUE RIBEIRO; et al. A Non-Hybrid Data-Driven Fuzzy Inference System for Coagulant Dosage in Drinking Water Treatment Plant: Machine-Learning for Accurate Real-Time Prediction. WATER, v. 15, n. 6, p. 14-pg., 2023-03-01. (22/03675-8)