Effects of UVC treatment on re-milled semolina dough and data — driven analysis of leavening process

• Published in: Food and bioproducts processing 2020-01, Vol.119, p.31-37
• Main Authors: Campagna, R., Romano, A., Raiola, A., Masi, P., Toraldo, G., Cavella, S.
• Format: Article
• Language: English
• Subjects: Bread dough; Colour; Data; Data acquisition; Dough; Dynamics; Effects; Evolution; Evolution-observation models; Food structure; Image acquisition; Image analysis; Image processing; Irradiation; Leavening kinetic; Mathematical analysis; Mathematical models; Moisture content; Particle filter; Procedures; Stochasticity; Studies; Ultraviolet radiation; Water content
• ISSN: 0960-3085; 1744-3571
• DOI: 10.1016/j.fbp.2019.10.009

•UV-treated semolina was evaluated as ingredient in bakery product development.•A prediction method allowed to monitor dough volume variations.•The impact of the UV treatment on dough re-milled- semolina was studied.•UV-treatment for 5–15min had a positive effect on dough volume ratio.•This data processing technique is particularly useful for missing information. The aims of the present work were to evaluate the impact of ultraviolet treatment of re-milled semolina on structural development of dough during mixing and leavening and to define a mathematical modelling for leavened doughs. The re-milled semolina was treated with UVC irradiation for 5, 15, 30 and 120min. The UVC treatment caused some changes of moisture content, colour and dough farinographic indices. The leavening kinetic was investigated by recording over time the variation of the dough volume by means of Image Analysis procedure. The mathematical procedure in this study was to model the evolution of a macroscopic morphology descriptor (volume) by a data-driven approach; the Particle Filter (PF) methodology incorporates the direct observations of the evolutionary dynamics in order to predict the data evolution. A specific differential problem is calibrated on the volume data of semolina doughs. An interesting feature of the proposed scheme is its ability to forecast variable states even when little information is available. More in detail, the volume variations are represented with a stochastic differential model and the dynamics of dough leavening are described by an evolution-observation scheme. The approach suggests the possibility of overcoming the costly data managing and image acquisitions.