NOTE! This site uses cookies and similar technologies.

If you not change browser settings, you agree to it. Learn more

I understand

Learn more about cookies at : http://www.aboutcookies.org/Default.aspx?page=1

April 2, 2020 | Leaf: Verification of mixed-criticality embedded systems

Leaf cea list 250A tool for the analysis and formal verification of the time properties of embedded computers is currently being developed at CEA List. The goal is to control cohabitation and interactions by addressing the hierarchy between functions with different criticality levels.

Increasingly complex hardware architectures and command-control algorithms are creating new challenges for the aeronautics, automotive, and other industries. Not least of which is the widening gap between average performance and worst-case scenarios. This is particularly problematic for embedded systems, where functions with different criticality levels and a variety of performance requirements must live on the same computer. The response times for embedded systems must be well-defined and controlled. However, the computing resources are dimensioned according to the worst-case response times, which are significantly overestimated to provide a safety buffer.

A more detailed assessment of the computers' time properties (anomalies, memory contention, etc.) would enable more efficient use of the available computing resources and, in the process, make it possible to reduce the safety buffer. List turned to the University of California, Berkeley for its world-leading scientific research and targeted expertise in predictable architectures. Researchers at List subsequently started development work on Leaf, a tool that makes clever use of formal methods to very accurately reproduce, analyze, and verify program behaviors on a given computer.

Ultimately, Leaf should be able to provide information that will help manufacturers design and dimension their systems and, therefore, increase trust in system behavior.

Read article at: http://www.cea-tech.fr/

 

 

March 23, 2020 | Better supervision of medium-voltage electrical distribution networks

reseaux moyenne tension 250 cea listA system for detecting transitory faults on medium-voltage electrical distribution networks was developed to prevent power outages before they occur. It can detect and locate very early indicators of wear or damage to cables.

An ounce of prevention is worth a pound of cure! Cable manufacturer Nexans turned to CEA List to develop an innovative monitoring system for medium-voltage electrical distribution networks. The system detects extremely brief transitory faults (from tens of nanoseconds to a few microseconds in duration) that conventional monitoring systems cannot pick up.

Even very minor, extremely brief current and voltage fluctuations are detected by sensors installed at different points on the network. What makes the method developed by List so original is that an external GPS clock is used to regulate the timing of the autonomous sensors. Triangulation based on the mathematical principle of time reversal effectively leverages the time differences between when a signal is received by different sensors to determine the location of a fault to within several centimeters for networks up to 10 km long.

The information is automatically transmitted and saved to a database managed by Nexans, which operates the data processing systems created by the CEA. The system is currently able to provide information on the location, intensity, and frequency of faults. The researchers will soon be able to determine the type of fault detected, as well. Nexans is now exploring how to best make use of this new tool. One idea is to develop a mobile maintenance service that would be made available to Nexans customers a few weeks at a time.

Read article at http://www.cea-tech.fr/

 

 

March 20, 2020 | Phoebe: a complete dose-simulation code

phoebe cea list 250A powerful and modular new computing code was developed to simulate the dose of radiation received by patients during radiation therapy and medical imaging procedures.

Researchers at List, a CEA Tech institute, developed a particle transport code that uses the Monte Carlo method. The goal is to simulate the dose of radiation received by patients during radiation therapy and medical imaging. To ensure optimal control of the simulation code in both technical terms (physical models and geometries) and in legal terms (licensing), the researchers developed a new code called Phoebe (PHOton and Electron Beams).

They used the validated physical models from Penelope, the Monte Carlo code most often used in radiation therapy, as the starting point for the new code, which was written in a programming language that is more portable and modular. For instance, because of the programming language chosen, Phoebe can be used on all operating systems, from PCs to smartphones. And, because Phoebe is modular, new features can be integrated later on. In fact, Phoebe was recently given a model to simulate physical phenomena at a cellular scale used to determine how a patient will respond to radiation-therapy-enhancing nanoparticle injections. List engineers are currently developing new models to factor in the entire dose received by the patient during a radiation treatment, including in areas far away from the tumor.

Phoebe came through laboratory validation testing with flying colors and the technology is mature enough at this stage for the models to be made available to a broad community of users via an open source platform.

Read article at http://www.cea-tech.fr/

 

 

March 19, 2020 | Keep connected!

message covid 2020 CEA List 2020 250In accordance with government containment guidelines, dedicated to controlling the spread of COVID-19 and ensuring everyone's safety, the CEA has suspended its on-site activities until further notice.

During this exceptional period, the CEA List teams remain mobilized to continue working on our R&D projects for our industrial and research partnerships. As of Monday, March 16, we have put in place the digital resources necessary to ensure our collective missions and commitments.

Keep connected and good luck to everyone.

 

 

13 février 2020 | Labellisation Carnot : le CEA List de nouveau labellisé !

Logo CARNOT CEA LIST 2020À la suite de l'appel à labellisation lancé en juin 2019, le ministère de l'Enseignement supérieur, de la Recherche et de l'Innovation a annoncé, le 7 février 2020, les noms des 39 instituts de recherche labellisés « Institut Carnot ». Parmi eux, Le List, institut de CEA Tech, a obtenu le renouvellement de sa labellisation.

Le List : un institut labellisé Carnot depuis 2006

Le List est labellisé Carnot depuis 2006 pour la qualité et le professionnalisme de ses activités de recherche partenariale avec les acteurs économiques.

Ce type de dispositif permet de lancer des projets de recherche exploratoire dans de nouveaux domaines en rupture et ainsi d’évaluer de nouvelles idées en amont. Il permet de mettre en œuvre des stratégies de collaboration, sous forme de chaires, bourses ou projets et de réaliser des démonstrateurs technologiques.

Le List est donc labellisé de nouveau sous le nom de Carnot CEA List.

Pour en savoir plus sur le label Carnot : Institut Carnot : un label d’excellence

Site web du réseau des Carnot : https://www.instituts-carnot.eu/fr