The optimization of power modules is made complex by the parasitic elements of the circuits (inductances, capacitances, resistance) which, when subjected to switching of high currents and voltages with high speeds (di/dt, dV/dt ), cause overvoltages or current oscillations which can be damaging to the system and components (accelerated aging, early failures, breakdown, thermal runaway, etc.)

The interest of the proposed PhD is to go beyond the usual methodology aimed at using probes or sensors (invasive and therefore which disrupt the circuit that we seek to characterize) by developing a non-invasive method ('Near Field Scanning » or NFS) making it possible to map the electric and magnetic fields near circuits and components (resolution <1mm, bandwidth >GHz). This field mapping is thus an image of the currents and voltages of the circuit, which will be necessary to process by inverse physical modeling, in order to go back to the real currents and voltages in the circuit. The PhD work therefore aims to develop and implement a hybrid approach which aims to couple a near-field characterization to 2D or 3D electromagnetic simulations in order to de-convolve the measured signal and provide information allowing the different current paths to be evaluated.
The longer-term prospects are to set up a non-invasive characterization tool coupled with simulation, in order to be able to characterize power modules and more generally power circuits by measuring their currents and voltages, even electromagnetic emissions (EMI for “Electro-magnetic Interferences”) which is another major problem in power circuits.

This PhD will take place at CEA Grenoble, within a mixed team from CEA-LETI and CEA-LITEN bringing together experts in power electronics, in collaboration with IRT St Exupéry which will provide NFS expertise.

Références :
- C. Lanneluc, P. Perichon and D. Bergogne, 'DC-Bus capacitors influence in a GaN Motor Drive Inverter,' PCIM Europe 2019; International Exhibition and Conference for Power Electronics, Intelligent Motion, Renewable Energy and Energy Management, Nuremberg, Germany, 2019, pp. 1-8.
- S. Serpaud, A. Boyer, S. B. Dhia and F. Coccetti, “Performance Charaterisation of the Dec Capa Network using the NFS Measurement”; EMC-Europe, September 2023, Krakow, Poland.