News

New portable electronic device based on AI to measure the quality of flours

The quality of flour masses is determined by physical characteristics such as elasticity or extensibility, which determine the processing conditions and their final use. Researchers from the University of Barcelona (UB) have developed a new portable and low-cost electronic device, which uses artificial intelligence (AI) to measure in a few seconds the viscoelastic properties of flour and similar products. The project has received a Product grant of 149,954.53 euros, within the framework of the Knowledge Industry program of the Generalitat de Catalunya, with the aim of improving this prototype and promoting its transfer to the cereal production industry.

The RELAPP project is led by Antoni Monleón, professor in the Department of Genetics, Microbiology and Statistics of the Faculty of Biology and researcher of the Consolidated Research Group GRBIO and the Biost3 group of the UB, where Pere López, who is the entrepreneurial scientist, also participates . Researchers from the Mat Control Laboratory of the Faculty of Chemistry of the UB, directed by professor Àngels Sahuquillo, and from the Electronic and Biomedical Instrumentation group of the UPC (IEB-UPC), directed by Juan Ramos Castro, also participate.

The evaluation of the viscoelastic properties of flour masses in the food industry is carried out with techniques such as the Chopin alveograph or the farinograph, which offer highly precise and accurate results, but which present different limitations. “The characterization must be done in an analytical quality control laboratory with expensive, delicate devices that require specialized technical personnel and that do not allow their continuous use in a production line”, explains Antoni Monleón.

Faced with this problem, the project team has developed an innovative, compact and low-cost device that makes it possible to obtain on-site measurements of properties of flour masses such as elasticity, tenacity, extensibility, penetrability, among others, through motion sensors and electronic accelerometers. “RELAPP aims to become a portable or field device, which does not replace those used in the laboratory, but which provides approximate results of viscoelastic parameters on a wide spectrum of materials, with adequate reliability to be used as a screening method Professor Monleón emphasizes.

The sensors are controlled by specific software that makes all the appropriate calculations in order to obtain parameters of industrial interest. “The technology is based on the geometric analysis of the deformation of a soft material such as flour dough and applies an artificial intelligence model to predict the main rheological parameters used in the cereal industry” , explains the UB researcher.

Unlike standard devices, the prototype allows the measurement of the viscoelastic properties of wholemeal flours or flours with mixtures of other products such as seeds, additives, etc. In addition, it could also be used to “identify or predict parameters based on expert knowledge such as the amount of water for a good kneading or the quality of fermentation based on criteria of previous experience or measurements on other materials such as plastics”, adds Monleón.

Prototype validated in the laboratory and in industry

In 2019, the project already obtained a Llavor grant, also from the Knowledge Industry program, which allowed the construction of two prototypes that were first validated in the Mat Control laboratory of the UB – which has materials for flour quality control and wheat – and then under controlled conditions at the industrial company Farines Ylla (Vic).

With the new Product aid, the aim is to develop a robust and reliable pre-industrial prototype to automate all the measures described and achieve a proof of concept tested in industry.

Image of the prototype developed by researchers at the University of Barcelona.

Share this post:

Utilitzem cookies de tercers amb finalitats tècniques i analítiques. Si continua navegant vol dir que accepta la nostra política de cookies. Més informació,plugin cookies política de cookies.

ACEPTAR
Aviso de cookies