{"id":3386,"date":"2025-07-16T10:14:23","date_gmt":"2025-07-16T08:14:23","guid":{"rendered":"https:\/\/www.astrum-it.de\/?post_type=projekt&p=3386"},"modified":"2025-07-16T10:25:10","modified_gmt":"2025-07-16T08:25:10","slug":"sustainet-innovate","status":"publish","type":"projekt","link":"https:\/\/www.astrum-it.de\/en\/projekt\/sustainet-innovate\/","title":{"rendered":"SUSTAINET-inNOvAte"},"content":{"rendered":"

Background<\/strong><\/p>\n

Advancing digitalization in areas such as medical technology, industrial automation and energy supply is steadily increasing their importance. The aim of the major European project SUSTAINET is to develop new communication technologies that increase the resilience, energy efficiency and sustainability of networks.<\/p>\n

As a CELTIC-NEXT flagship project, SUSTAINET pools the expertise of over 80 European research institutions, universities and companies. The project is divided into the three pillars SUSTAINET-Advance, SUSTAINET-inNOvAte and SUSTAINET-guarDian and has a volume of over 67 million euros.<\/p>\n

 <\/p>\n

The task<\/strong><\/p>\n

Within this framework, the SUSTAINET-inNOvAte sub-project focuses on researching and demonstrating a reliable and fail-safe communication infrastructure and communication networks that enable secure, seamless and energy-efficient connectivity.<\/p>\n

\"\"<\/p>\n

Challenge for ASTRUM IT<\/strong><\/p>\n

Within the project, ASTRUM IT is focusing on increasing the reliability of high-speed optical transceivers through predictive maintenance, thereby contributing to increasing the reliability of high-speed transceivers. By using machine learning (ML), potential failures can be predicted, which enables proactive maintenance and minimizes downtimes.<\/p>\n

The core tasks of ASTRUM IT include:<\/strong><\/p>\n

AI-supported anomaly detection: selection and optimization of methods to identify fault patterns.
\nData analysis and processing: evaluation and pre-processing of raw data to develop a robust framework.
\nPrediction of remaining useful life (RUL): Development and validation of algorithms for estimating the remaining useful life of transceivers.
\nIntegration and visualization: Integration of the framework into the development environment and presentation of the RUL estimates via a dashboard for real-time monitoring.
\nThese data-driven approaches drastically reduce the probability of hardware failures and extend the lifetime of optical transceivers.<\/p>\n

Project partner:<\/strong><\/p>\n