A field-portable gas analyzer with an array of six semiconductor sensors. Part 2: Identification of beer samples using artificial neural networks

✍ Scribed by Peter W. Alexander; Lucy T. Di Benedetto; D. Brynn Hibbert

• Publisher: John Wiley and Sons
• Year: 1998
• Weight: 129 KB
• Volume: 2
• Category: Article
• ISSN: 1086-900X
• DOI: 10.1002/(sici)1520-6521(1998)2:3<145::aid-fact3>3.0.co;2-u

✦ Synopsis

A method is described based on a new portable, multisensor gas analyzer applying an artificial neural network to discriminate between six beer brands. The gas analyzer as previously reported employed six different tin-oxide semiconductor sensors, and the detection method was based on headspace analysis of the vapor above beer samples. The artificial neural network (ANN) used in this study was a three-layer network, standard back-propagation algorithm. The network was trained with the use of 553 cycles in at a learning rate 7.11 min of 1.0 and training tolerance of 0.1. A Macintosh PowerBook 1400cs with PowerPCீ 603e at 133-MHz clock frequency and 128-kilobyte Level Two writethrough cache memory on a processor system bus was used to train the ANN. This study indicates that the portable, multisensor analyzer is able to discriminate among beers and thus may be used to monitor beer product quality in industrial processes, having the advantage of portability and low cost for use in sites remote from chemical laboratories. Further applications in food technology are in the testing of foods and beverages for quality and shelf life.