Handheld near infrared spectrometer and machine learning methods applied to the monitoring of multiple process stages in industrial sugar production. (1st February 2022)

• Record Type: Journal Article
• Authors: Henrique da Silva Melo, Bruno; Figueiredo Sales, Rafaella; da Silva Bastos Filho, Lourival; Souza Povoas da Silva, Jorge; Gabrielle Carolino de Almeida Sousa, Aluska; Maria Camará Peixoto, Deborah; Pimentel, Maria Fernanda
• Is Part Of: Food chemistry. Volume 369(2022)
• Publication Date: 2022-02-01
• Subjects: Soft-sensor -- Process analytical technology -- Support vector machine -- Partial least squares -- Process monitoring -- Sugarcane; Food -- Analysis -- Periodicals; Food -- Composition -- Periodicals; 664
• DOI: 10.1016/j.foodchem.2021.130919
• Languages: English
• ISSNs: 0308-8146

Abstract

Highlights: A handheld NIR spectrometer was used to monitor brix and pol in sugar production. Calibration models of NIR spectra were developed using machine learning methods. SVR outperformed PLS, which was ascribed to the non-linearities in the data. For each parameter, two models only were required to cover all the factory samples. Results indicated the feasibility of using NIR data for soft-sensor development. Abstract: This work aimed to evaluate the performance of a handheld NIR spectrometer in developing calibration models to quantify brix and pol at various stages of an industrial sugar production process. Because of sample variability, collected over two harvesting seasons, NIR measurements were acquired either in transmittance or diffuse reflectance. For modelling purpose, partial least squares (PLS), also combined with variable selection techniques, and support vector machine regression (SVR) were investigated. SVR was applied to handle non-linearities within the data. In general, results illustrated the best performance of SVR, that yielded lower root mean square error of prediction (RMSEP) values for brix and pol for spectra acquisition in transmittance (0.59 and 0.69%w/w) and using diffuse reflectance (1.44 and 2.44%w/w), respectively. Results from models using spectra collected in transmittance were comparable to those reported in other works where benchtop instruments were used, highlighting the cheaper and simpler employment of the portable spectrometer.