PowerCare: Cognitive power electronics with
integrated health monitoring
PowerCare: Cognitive power electronics with
integrated health monitoring

Project

Motivation

Comprehensive electrification is key to decarbonizing industry and mobility. At the same time, there is a growing demand for powerful, reliable, and energy-efficient power electronics. Industrial motors account for around 30% of global electricity consumption and contribute significantly to greenhouse gas emissions. Conventional power modules are reaching their limits here—especially in terms of efficiency, miniaturization, and early failure detection.

Goals

  • Development of compact, energy-efficient power modules based on vertical GaN semiconductors
  • Integration of AI-supported real-time failure prediction directly into the motor control system
  • Early detection of malfunctions to prevent unplanned downtime
  • Reduction of CO₂ emissions in industry and mobility (potential: up to 124 Mt CO₂e/year)
  • Increased safety and optimization of maintenance and operating processes

Solution

PowerCare pursues a novel co-design approach combining power semiconductors, microcontrollers, and AI models. Vertical GaN power semiconductors are combined with an AI-enabled high-performance microcontroller that enables local status monitoring and real-time failure prediction.

The AI analyzes operating data directly from the power electronics—without additional sensors—and is trained with real-world mission profiles. The result is a miniaturized, energy-efficient power module that can be integrated directly into drives.

The project is being implemented jointly by several Fraunhofer Institutes and industry partners and will result in practical demonstrators and test platforms for industrial use by 2026.