On-line multisensor monitoring of yogurt and Filmjolk fermentations on production scale
2004 (English)In: Journal of Agricultural and Food Chemistry, ISSN 0021-8561, E-ISSN 1520-5118, Vol. 52, no 3, 415-420 p.Article in journal (Refereed) Published
Near-infrared (NIR) spectrometry and electronic nose (EN) data were used for on-line monitoring of yogurt and filmjölk (a Swedish yogurt-like sour milk) fermentations under industrial conditions. The NIR and EN signals were selected by evaluation of principal component analysis loading vectors and further analyzed by studying the variability of the selected principal components. First principal components for the NIR and the EN signals were used for on-line generation of a process trajectory plot visualizing the actual state of fermentation. The NIR signals were also used to set up empirical partial least-squares (PLS) models for prediction of the cultures' pH and titratable acidity (expressed as Thorner degrees, °T). By using five or six PLS factors the models yielded acceptable predictions that could be further improved by increasing the number of reliable and precise calibration data. The presented results demonstrate that the fusion of the NIR and EN signals has a potential for rapid on-line monitoring and assessment of process quality of yogurt fermentation.
Place, publisher, year, edition, pages
American Chemical Society (ACS), 2004. Vol. 52, no 3, 415-420 p.
bioprocess monitoring, electronic nose, near-infrared spectroscopy, principal component analysis, partial least-squares (PLS) regression, process trajectory plot
Bioprocess Technology Food Engineering
IdentifiersURN: urn:nbn:se:liu:diva-48419DOI: 10.1021/jf0304876ISI: 000188830300003OAI: oai:DiVA.org:liu-48419DiVA: diva2:269315