Developed by a team at the Autonomous University of Barcelona, the sensor - described as an electronic tongue - is claimed to have 82 per cent accuracy. The work is detailed in Food Chemistry.
‘The concept of the electronic tongue consists in using a generic array of sensors, in other words with generic response to the various chemical compounds involved, which generate a varied spectrum of information with advanced tools for processing, pattern recognition and even artificial neural networks,’ said Manel del Valle, the main author of the study
The array was formed with 21 ion-selective electrodes combining cationic/anionic sensors, and others with generic response.
Responses were evaluated using two different pattern recognition methods: principal component analysis, which allowed identifying some initial patterns, and linear discriminant analysis in order to achieve the correct recognition of sample varieties.
‘Using more powerful tools – supervised learning – and linear discriminant analysis did enable us to distinguish between the main categories of beer we studied: Schwarzbier, lager, double malt, Pilsen, Alsatian and low-alcohol…with a success rate of 81.9 per cent,’ Del Valle said in a statement.
To order the varieties, the scientists estimated alcohol content with a numerical model developed with an artificial neural network.
‘This application could be considered a sensor by software, as the ethanol present does not respond directly to the sensors used, which only respond to the ions present in the solution.’
The researchers believe the tools could eventually be incorporated into robotic systems to improve quality control for food and drink manufacturers.
Project to investigate hybrid approach to titanium manufacturing
What is this a hybrid of? Superplastic forming tends to be performed slowly as otherwise the behaviour is the hot creep that typifies hot...