Banca de QUALIFICAÇÃO: RAUAN DOUGLAS DA CUNHA BARROS

Uma banca de QUALIFICAÇÃO de MESTRADO foi cadastrada pelo programa.
STUDENT : RAUAN DOUGLAS DA CUNHA BARROS
DATE: 24/08/2026
TIME: 14:00
LOCAL: Vídeo conferencia
TITLE:

Integrated UV-vis and ATFM-LED Chemometrics for Environmental Compliance Monitoring of Produced Water in Offshore

KEY WORDS:

Produced water; Total oil and grease; UV-Vis spectroscopy; ATFM-LED; Chemometrics; Environmental compliance; Offshore monitoring.

PAGES: 80
BIG AREA: Ciências Exatas e da Terra
AREA: Química
SUBÁREA: Química Analítica
SUMMARY:

Produced water (PW) is the largest effluent stream generated during offshore oil and gas production, representing a major environmental and operational challenge due to its high volume, chemical complexity, and strict discharge requirements. As reservoirs mature and the water cut increases, rapid and reliable analytical strategies become essential to support treatment control, regulatory compliance, and offshore decision-making. In this study, UV-Vis spectroscopy and an Attenuated Fluorescence Measurement (ATFM) platform based on LED optical responses were combined with chemometric tools for the quantitative prediction and classification of total oil and grease (TOG) in produced water. The proposed workflow included data preprocessing, Partial Least Squares (PLS) regression, Data-Driven Soft Independent Modeling of Class Analogy (DD-SIMCA), and One-Class Genetic Algorithm (OGA) for variable selection. The PLS models showed satisfactory predictive performance for all optical systems evaluated, with R2Pred = 0,972 and RMSEP = 4,63 mg L−1 for the blue LED/turbidity response, R2Pred = 0,971 and RMSEP = 4,75 mg L−1 for the red LED response, and R2Pred = 0,950 and RMSEP = 5,96 mg L−1 for UV-Vis spectroscopy. DD-SIMCA models were used to distinguish compliant samples, below 29 mg L−1 , from non-compliant samples, above 30 mg L−1 , according to CONAMA Resolution No. 393/2007. For the ATFM models without variable selection, the fluorescence channel achieved 90% training sensitivity, 100% prediction sensitivity, and 86% specificity, while the turbidity channel achieved 85% training sensitivity, 100% prediction sensitivity, and 75% specificity. For UV-Vis data, the application of OGA improved DD-SIMCA specificity from 70.45% to 79.54%, while sensitivity remained unchanged. Overall, the integration of UV-Vis spectroscopy, ATFM-LED optical sensing, PLS, DD-SIMCA, and OGA provides a rapid, low-cost, solvent-free, and environmentally aligned strategy for primary screening of produced water, with strong potential for routine offshore monitoring and real-time environmental decision-making. 

COMMITTEE MEMBERS:
Interno - ***.475.614-** - CAMILO DE LELIS MEDEIROS DE MORAIS - UFRN
Presidente - 1714946 - KASSIO MICHELL GOMES DE LIMA
Externo à Instituição - WILDSON ARCANJO DE MORAIS