Machine learning to predict the behaviour of untested electronic components

The project will help users to know at the design stage whether a component is suitable for their instrument, thus saving test execution costs and time.

The National Accelerator Centre (CNA) and Alter Technology, acting as Aggregate Agent, are leading the project on Prediction of the Electrical Behaviour of Electronic Devices under Radiation (PRECEDER).

This is a knowledge transfer subproject, based on artificial intelligence, whose objective is to prepare a large database and develop Machine Learning techniques on a set of results, which allow predicting the behaviour of other untested electronic components based on experience.  

The CNA, a joint centre of the University of Seville, the Junta de Andalucía and the CSIC, is a reference for irradiation tests and the company Alter is an expert in the irradiation of electronic devices for the space sector. The evaluation of radiation behaviour is essential for the design and assembly of satellites, probes, robots, etc.  

The tool to be developed in this project will allow the prediction of behaviour, and therefore has a direct application in space and hostile environment projects. It will help users to know at the design stage whether a component is suitable for their instrument, thus saving test execution costs and time.
 
PRECEDER is part of the Project Innovative Ecosystem with Artificial Intelligence for Andalusia 2025 led by the Campus of International Excellence Andalucía TECH for the University of Seville and the University of Malaga to act with technology companies for the development of artificial intelligence technologies in all areas of the Smart Specialisation Strategy (RIS3) in Andalusia. Forty-nine knowledge transfer sub-projects are included in this initiative funded by the Andalusian Regional Government, through the Directorate General for Research and Knowledge Transfer of the Regional Ministry of Economy, Knowledge, Business and University, within the framework of the ERDF Operational Programme.