R-PODID

We are excited to share the news that Bewell Tech has is a part of the R-PODID EU research project, a co-funded initiative by CHIPS JU. The primary goal of R-PODID is to pioneer the development of an automated, cloudless, short-term fault-prediction system for electric drives, power modules, and power devices, with seamless integration into power converters.

Within the scope of the R-PODID project, Bewell is set to contribute by developing AI models specifically designed for on-device training, along with the creation of a custom Linux-based gateway.

The project's objectives include:

• establishing a methodology for generating fault-prediction models from sparse training sets or system simulations,

• integrating power electronics with embedded AI support,

• 24-hour fault-prediction capabilities for Gallium Nitride (GaN) and Silicon Carbide (SiC) based power converters.

• 24-hour fault-prediction and fault mitigation for electric drives, implement sensors for reliability prediction in power modules, and enhance overall system resilience.

With the collaborative efforts of 33 partners, R-PODID's groundbreaking innovations will be incorporated into power modules and applied across diverse use cases, spanning conveyor belts, industrial lighting, automotive traction inverters, and a heavy-duty testbed. This project marks a significant step forward in advancing predictive maintenance solutions and embedding cutting-edge technologies in the realm of power electronics.