Integration of Artificial Intelligence and IoT in Chemical Process Optimization and Water Quality Improvement in Treatment Plants

• Grant number: 25/03131-6
• Support Opportunities: Research Grants - Innovative Research in Small Business - PIPE
• Start date: November 01, 2025
• End date: October 31, 2027
• Field of knowledge: Engineering - Sanitary Engineering - Environmental Sanitation
• Principal Investigator: Amanda Eugênio de Castro
• Grantee: Amanda Eugênio de Castro
• Pesquisadores principais: Ana Carolina Moralles Barbosa Silva ; Brandon Silva de Araujo ; Jessika Campos de Sousa ; Luiz Octávio Martini Lima

Abstract

Physicochemical processes play a crucial role in water treatment, ensuring that quality standards are met. Determining the optimum coagulant dosage, which is influenced by factors such as pH, temperature, organic matter concentration, among others, is crucial to optimizing the efficiency of the water treatment process. However, conventional methods such as the widely used jar test have limitations due to their slow execution, making it difficult to respond quickly to changes in raw water quality. As a result, operators often adopt high preventive dosages, resulting in inadequate coagulation, increased costs, equipment wear and increased waste production. In phase 1 of the project, the potential of an IoT platform for controlling and dosing chemicals in Water Treatment Plants (WTPs) and Effluent Treatment Plants (ETEs) was evaluated. However, significant limitations were identified, such as the lack of instrumentation in WTPs, which compromised large-scale data collection, making it difficult to apply big data, train Artificial Intelligence (AI) algorithms, and perform multiparameter modeling. In addition, the need for more intuitive interfaces, clear organization of information, and greater attention to user needs was identified in order to optimize the process of interacting with the system. These factors restricted the development of an advanced system with predictive and proactive capabilities. Based on the lessons learned from this phase, phase 2 of the project proposes the development of a system that integrates IoT and AI to optimize physical-chemical processes, with an emphasis on accurate flocculant dosing. The solution will use in-line sensors coupled to a multiparameter monitoring platform. This approach will allow dynamic and efficient adjustment of coagulant dosage, reducing chemical waste, minimizing operating costs, and improving the quality of treated water. By directly impacting the efficiency and sustainability of treatment processes, the project has great potential to positively impact the sanitation sector, promoting the reduction of the environmental footprint throughout the entire production chain, including the manufacture of chemical products and sludge management. The implementation of this technological solution will bring a substantial advance in the modernization and sustainability of water treatment processes, with significant gains in terms of energy efficiency, reduction of chemical waste and water quality. (AU)