Periodic Reporting for period 2 - R-PODID (Reliable Powerdown for Industrial Drives)
Período documentado: 2024-09-01 hasta 2025-08-31
R-PODID aims to develop the technologies and methodologies required to improve the reliability in power electronics circuits and systems based on III/V-semiconductor devices like gallium-nitride (GaN) and silicon-carbine (SiC). GaN and SiC devices will be the standard devices for power electronic systems of the future thanks to their electric and thermal performance, but they suffer reliability limitations. Moreover, periodic shutdowns of industrial plants for maintenance purposes set a considerable industrial cost, while a modern AI-based prediction of the failure may lower the exercise costs and trigger novel business models. R-PODID targets all the above scenarios by developing models, circuits, and systems for improving the reliability of power electronics systems.
The project's basic idea is to incorporate and combine artificial intelligence (AI) algorithms with state-of-the-art physical models to realize compact failure models that can be deployed on resource-constrained embedded systems commonly used in power applications, allowing for short-term fault prediction and real-time residual lifetime estimation. To achieve the final goal, the project has to face interdisciplinary challenges and set ambitious objectives.
Failure analysis is typically characterized by a reduced training dataset, therefore novel methodologies for developing AI models from sparse datasets must be conceived. Advanced AI learning techniques and combination with compressive sensing, will be investigated for this purpose. AI models are usually run in the cloud, on very powerful servers. However, the applications foreseen in this project require the local deployment of the failure model on embedded devices, which are usually resource-constrained. This is especially true in the automotive context, which is targeted by two Use Cases in R-PODID. The development of efficient methodologies for the deployment of the AI, or the AI-enhanced, model into the embedded device is another objective of the R-PODID project. The project will target the most used embedded architectures: a RISC-V-based system with safety capabilities for industrial applications and an ARM-cortex MCU with ASIL-D certifications for automotive applications. Finally, the AI-based models require a constant update of information about the current behavior of the monitored system, therefore specific and performing sensors must be developed to gather a great amount of data with the highest possible level of accuracy, given the critical nature of the applications. Moreover, in the case of federated learning over large industrial plants, secure and reliable communication systems must be defined. The last challenge is the integration of all the above-described methodologies and technologies into power-converter architectures without impinging the converter's original performance.
The achievement of the R-PODID goal will allow for an automated, cloudless, short-term fault prediction for electric drives, power modules, and power devices, that can be integrated into power converters. Thereby, electrical and mechanical faults of machines and of the power converters driving them will become predictable within a limited prediction horizon of 12-24 hours. This will enable a power-saving shutdown of a larger number of production machines during idle times, because a looming failure during the next power-on cycle can be reliably foreseen. It will also enable reliable mitigation of dangerous faults in critical applications like automotive.
The project's basic idea is to incorporate and combine artificial intelligence (AI) algorithms with state-of-the-art physical models to realize compact failure models that can be deployed on resource-constrained embedded systems commonly used in power applications, allowing for short-term fault prediction and real-time residual lifetime estimation. To achieve the final goal, the project has to face interdisciplinary challenges and set ambitious objectives.
Failure analysis is typically characterized by a reduced training dataset, therefore novel methodologies for developing AI models from sparse datasets must be conceived. Advanced AI learning techniques and combination with compressive sensing, will be investigated for this purpose. AI models are usually run in the cloud, on very powerful servers. However, the applications foreseen in this project require the local deployment of the failure model on embedded devices, which are usually resource-constrained. This is especially true in the automotive context, which is targeted by two Use Cases in R-PODID. The development of efficient methodologies for the deployment of the AI, or the AI-enhanced, model into the embedded device is another objective of the R-PODID project. The project will target the most used embedded architectures: a RISC-V-based system with safety capabilities for industrial applications and an ARM-cortex MCU with ASIL-D certifications for automotive applications. Finally, the AI-based models require a constant update of information about the current behavior of the monitored system, therefore specific and performing sensors must be developed to gather a great amount of data with the highest possible level of accuracy, given the critical nature of the applications. Moreover, in the case of federated learning over large industrial plants, secure and reliable communication systems must be defined. The last challenge is the integration of all the above-described methodologies and technologies into power-converter architectures without impinging the converter's original performance.
The achievement of the R-PODID goal will allow for an automated, cloudless, short-term fault prediction for electric drives, power modules, and power devices, that can be integrated into power converters. Thereby, electrical and mechanical faults of machines and of the power converters driving them will become predictable within a limited prediction horizon of 12-24 hours. This will enable a power-saving shutdown of a larger number of production machines during idle times, because a looming failure during the next power-on cycle can be reliably foreseen. It will also enable reliable mitigation of dangerous faults in critical applications like automotive.
All Work Packages (WPs) and Use Cases (UCs) are currently operational. Interdependencies between WPs have been clearly identified and are being actively monitored, resulting in no critical issues to date.
All the technical WPs have delivered high-quality intermediate results, contributing to the successful achievement of Milestone 2.
WP1 has produced a substantial volume of datasets, and initial versions of AI models are now available for testing.
WP2 has developed several power converters, utilizing either GaN or SiC technologies, for integration into the UCs’ demonstrators.
WP3 concentrated on the integration of demonstrators in their initial deployment phase.
All preliminary demonstrators are now ready for testing under WP4, which will inform the necessary refinements for final demo versions.
Main Results:
Delivered an end-to-end pipeline: GaN aging datasets from dedicated stress setups, refined PMSM and wide-band HF fault models, a nine-phase motor dataset, and an EMC-compliant SiC drive bench with PFRA.
Device/module models were validated, including a 40 GHz gate-resistor equivalent, GaN PCM v0 built and integrated (v1 entering production), and a UC2a-aligned SiC inverter with verified electro-thermal and reduced-order models plus FOC/pole-pair control.
AI capabilities include fault-detection pipelines and RUL models (3-phase sensor degradation, GaN-HEMT with uncertainty), optimized for edge via compression, low-rank/work-reuse/accelerators, and deployed on Arduino Nano, RISC-V MCUs, HIL, and Ultrascale workflows.
Field-ready connectivity and sensing were matured: IIoT add-on v2.0 (better encoding/recovery) with BLE→Ethernet→5G chain in progress; IDPS tested and deploying; sensors validated (Hall up to 250 Arms/6000 rpm, Vds monitoring, improved shunts, Pt temperature detectors).
Four V1 demonstrators are assembled, functionally validated, and entering AI test campaigns; the AI-in-PWM ASIC path was dropped due to area/memory limits, and resources reallocated to higher-impact integration and V2 upgrades.
In UC1, the consortium opted to split the power converter design into two distinct versions: version 0 (v0) will be used for monitoring motor degradation, while version 1 (v1) will focus on tracking GaN device degradation.
All the technical WPs have delivered high-quality intermediate results, contributing to the successful achievement of Milestone 2.
WP1 has produced a substantial volume of datasets, and initial versions of AI models are now available for testing.
WP2 has developed several power converters, utilizing either GaN or SiC technologies, for integration into the UCs’ demonstrators.
WP3 concentrated on the integration of demonstrators in their initial deployment phase.
All preliminary demonstrators are now ready for testing under WP4, which will inform the necessary refinements for final demo versions.
Main Results:
Delivered an end-to-end pipeline: GaN aging datasets from dedicated stress setups, refined PMSM and wide-band HF fault models, a nine-phase motor dataset, and an EMC-compliant SiC drive bench with PFRA.
Device/module models were validated, including a 40 GHz gate-resistor equivalent, GaN PCM v0 built and integrated (v1 entering production), and a UC2a-aligned SiC inverter with verified electro-thermal and reduced-order models plus FOC/pole-pair control.
AI capabilities include fault-detection pipelines and RUL models (3-phase sensor degradation, GaN-HEMT with uncertainty), optimized for edge via compression, low-rank/work-reuse/accelerators, and deployed on Arduino Nano, RISC-V MCUs, HIL, and Ultrascale workflows.
Field-ready connectivity and sensing were matured: IIoT add-on v2.0 (better encoding/recovery) with BLE→Ethernet→5G chain in progress; IDPS tested and deploying; sensors validated (Hall up to 250 Arms/6000 rpm, Vds monitoring, improved shunts, Pt temperature detectors).
Four V1 demonstrators are assembled, functionally validated, and entering AI test campaigns; the AI-in-PWM ASIC path was dropped due to area/memory limits, and resources reallocated to higher-impact integration and V2 upgrades.
In UC1, the consortium opted to split the power converter design into two distinct versions: version 0 (v0) will be used for monitoring motor degradation, while version 1 (v1) will focus on tracking GaN device degradation.
The project is well-positioned towards the achievement of the main objectives. Four preliminary demo platforms have been realized to demonstrate the key ideas and results of the project. The technologies and methodologies developed in R-PODID will demonstrate the potential impact of the cognitive power converters, but further development beyond the project is required for a full uptake.
R-PODID is expected to delivers commercial, environmental, scientific, and social impact by deploying AI‑driven predictive maintenance across new and legacy assets. Commercially, it enables subscription‑based replacement‑part services that improve production planning, boost efficiency, and cut inventory. New methods for optimizing assembly lines target a market >$23B and 162M motors by 2026; with motors using >70% of industrial electricity and 6–10% of drives idling in downtime, the project addresses major cost and energy losses. By advancing wide‑bandgap (WBG) power electronics and reliability knowledge, it catalyzes new products and growth for SMEs and large firms, reinforcing Europe’s semiconductor position; strong industrial participation ensures applicability. Socially, universities are training a new workforce in power electronics, AI, and sensing via theses, PhDs, and early‑career involvement. Impact is demonstrated in use cases: predictive maintenance in legacy lines (UC1) and automotive demonstrators (UC2a/UC2b), with outreach to precision automated manufacturing. Scientifically, diverse universities and RTOs drive interdisciplinary innovation through joint publications, workshops, and new collaborations.
R-PODID is expected to delivers commercial, environmental, scientific, and social impact by deploying AI‑driven predictive maintenance across new and legacy assets. Commercially, it enables subscription‑based replacement‑part services that improve production planning, boost efficiency, and cut inventory. New methods for optimizing assembly lines target a market >$23B and 162M motors by 2026; with motors using >70% of industrial electricity and 6–10% of drives idling in downtime, the project addresses major cost and energy losses. By advancing wide‑bandgap (WBG) power electronics and reliability knowledge, it catalyzes new products and growth for SMEs and large firms, reinforcing Europe’s semiconductor position; strong industrial participation ensures applicability. Socially, universities are training a new workforce in power electronics, AI, and sensing via theses, PhDs, and early‑career involvement. Impact is demonstrated in use cases: predictive maintenance in legacy lines (UC1) and automotive demonstrators (UC2a/UC2b), with outreach to precision automated manufacturing. Scientifically, diverse universities and RTOs drive interdisciplinary innovation through joint publications, workshops, and new collaborations.