Control of sulphates in hazardous areas using artificial intelligence

• Grant number: 22/13720-0
• Support Opportunities: Research Grants - Innovative Research in Small Business - PIPE
• Start date: April 01, 2023
• End date: June 30, 2024
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computer Systems
• Agreement: SEBRAE-SP
• Principal Investigator: Eduardo José Sande e Oliveira dos Santos Souza
• Grantee: Eduardo José Sande e Oliveira dos Santos Souza
• Company: X Machina Fabricação de Equipamentos de Medida, Teste e Controle S/A (Filial)
• CNAE: Desenvolvimento e licenciamento de programas de computador não-customizáveis
• City: Vinhedo

Abstract

The main problem reported by Oxiteno is the lack of knowledge by the operation team of the fluid situation in the different stages of the process in classified areas (with associated risk) - today, collections are carried out that are analyzed by the quality laboratory, which arrive at a time later to the production team creating lags between analyzed and in-process fluids. The analyses carried out at the end of the process inform about the fulfillment of the product specification, but are unable to determine in which phase of the process the problems occur (whether with the inputs or in one of the phases). As mentioned above, Surfactant processes are complex and require online monitoring, Oxiteno produces 3 ton/h and has analyzes such as Insulphates and Dioxane that take 4 hours to complete and generate results, that is, possible process deviations in this interval may result in 12 ton of out of specification product. With this, external customers of the partner company can be impacted by the generation of non-conformities, if the product out of specification is not detected in time.In this way, through the implementation of XMachina technology, this problem can be eliminated. If the problem is detected only in the final tank, it is necessary to carry out blending costs of the product, which, in some cases, also impacts the market, leaving the availability of an approved product to meet the customer's needs. In addition, Oxiteno works with areas classified as EX (risks with flammable fluids) and, for this, it is necessary to use XMachina's technology specifically for this purpose, meeting the electrical classification standard for the area. In order to optimize analyzes that are already mandatory, Oxiteno will seek to validate the innovative technology developed by XMachina, through the implementation of the Fluid Homogeneity Monitor product between collections. At the end of the technology implementation, it is expected that Oxiteno will be able to: 1 - Monitor the homogeneity between fluid collections at different stages of the process in order to associate the quality problems of the final product with the occurrence of anomalies, either with the inputs or at some stage of the process; 2 - Identify monitoring profiles with analyzed anomalies in order to allow quick interventions by the operation team; 3 - Identify synchronicities between anomalies that occur in the process in order to assess cause and effect relationships between them; 4 - Develop closed-loop design in order to make automatic corrections of anomalies. The solution will be implemented at the Tremembé Unit in the Sulfation plant that was started up in 1991. In this plant, 70% sodium lauryl ether sulfate is produced, varying only the ethoxylated alcohol used. This project will be carried out with the product sodium lauryl ether sulfate 2 EO, which represents the largest volume of the plant, the highest occurrence of quality failures and is in a classified area - key point to validate XMachina's technology in an application never performed before and supplying Oxiteno's challenge of online assessments in an area that limits the use of non-certified equipment. With online monitoring, it is expected to increase the quality assurance of the process, due to the product quality control data in real time, which allows immediate action in case of process deviations, thus avoiding the generation of out-of-specification product, external complaints and need for reprocessing and rework. (AU)