Vertical GaN power transistors combine the performance advantages of vertical wide bandgap (WBG) transistors with the cost advantages of established silicon technology. In due course, they may replace IGBTs to reduce energy conversion losses in many price-sensitive applications, ranging from data center power supplies to traction inverters for electric vehicles, establishing GaN as the semiconductor of choice beyond consumer electronics. At PowerCare, vertical GaN trench MOSFETs based on engineered substrates are being developed for 48 V applications with a rated current between 35 and 160 A per device. Higher voltage levels are also being evaluated.
Failure models for inverters and connected electric motors are developed and ported to a RISC-V-based power module for on-site execution. A PWM controller with RISC-V architecture is extended with hardware accelerators for the functions that are important for the models (e.g., FFT, filtering). The model size and execution speed are optimized for real-time requirements (e.g., quantization or pruning).
The failure model for electric motors takes into account changes in load current and, optionally, other sensor data (vibration, acoustics, instantaneous speed) that can be observed due to impending failures. Based on the detection of bearing damage, the development will cover other faults such as demagnetization or winding faults.
The failure model for transistors and inverters is developed based on data from service life tests and parameter measurements and enables the training of failure prediction.
A GaN-based and AI-enabled power module is being built, into which the trained failure models are integrated and executed locally, with the current and sensor data from the inverter and motor serving as input parameters.
PowerCare: Cognitive power electronics with