Overview
Electrical machines in traction drives are usually supplied with three-phase inverters which are built in two-level topology. The DC battery voltage of electric buses is switched, and pulse-width modulated by the inverter with a carrier frequency in the kHz-range (typically with a maximum of 4kHz). Beside the conducting and switching losses of the inverter, the pulsed voltage causes additional losses in the machine which occur in the conductors of the stator and rotor as well as in the core lamination leading the magnetic field. Up to now it has not been achieved to pre-calculate the inverter-specific, additional machine losses accurately and with general validity for different operational parameters.
Content and goal of this project will be the significant increase of the overall efficiency of the drive by intelligent and coordinated operation and design. To achieve this, the development and validation of an advanced electric drive loss model will be carried out in order investigate the dependence of the total losses from different parameters. These investigations will be carried out for two- and three-level inverters. Due to several advantages of the three-level inverter topology it is expected that the three-level topology will also be used for traction drives in the medium term; three-level inverters will thus be investigated in the project. For an accurate calculation of the iron losses, the hysteresis model will be used which was developed in the FFG-funded project InWeMat. It will be specialised for pulsed voltages and modified for the use in concrete sections of an electric machine. The hysteresis model will then be integrated into existing machine loss models which are already able to calculate the temperature-dependent copper losses in both the stator and the rotor.
Inverter loss models will be adapted and combined with the machine model. This overall loss model for an electric drive will be used to investigate the influence of different operational parameters (switching frequency and pulse pattern of the inverter, the machine's point of operation), inverter topologies as well as machine designs onto the total losses. These findings will make it possible to select an inverter and machine design and develop a coordinated operational strategy which minimises the losses for a defined driving cycle of an electric bus. It is expected that these measures will reduce the total losses for the defined cycle by 20-30% compared to a drive with standard design and standard operation. In order to validate the simulations experimentally, prototypes will be developed. These will be an inverter which can be operated in both two- and three-level topology, and two testing machines with different iron lamination quality. Due to the pulsed voltage of the inverter, bearing currents emerge which can damage the bearings and lead to an early outage.
The influence of the operational parameters and inverter topology on the bearing currents of the machine will be determined experimentally and it will be assessed how these currents influence the lifetime of the machine. Those findings will also be incorporated into the operational strategy, as the drive should not only be highly energy-efficient but also resource-efficient by guaranteeing a long lifetime and high reliability