To achieve the goals set out in the Green Deal, large scale electrification of society will be needed. This includes massive development of renewable energy production from solar/wind/hydro, electrification of transport, decarbonization of industrial processes and efficient transmission of electric power. Power electronic converters is a key technology for conversion and control of electric power and will be crucial for successful electrification. The technology is rapidly evolving, providing much faster switching rates and higher voltages, in order to reduce complexity and increase efficiency compared to today’s solutions.
The faster switching and higher voltages result in steeper voltage shapes which introduce increased and different stresses internally in power components. These stresses initiate new and accelerate existing failure modes in the electric insulation of the power components, resulting in more severe ageing and deterioration of materials. The experience and knowledge built on conventional technology and stresses is not always relevant. Hence, new and more complex models and analysis on all levels from the material behaviour to component and system analysis are required to understand and predict the effects of faster power electronics.
The project will develop methods to measure and characterize the increased stresses power components are exposed to and. A novel method for measuring the electric field along non-linear field grading has been demonstrated and will be applied to investigate increased electric stresses in end-winding and termination due to fast voltage transients. SwoP will further investigate the acceleration of degradation, and initiation of new fault scenarios, that will emerge due to use of fast power electronic converters. The project has a special focus on how transformer and machine insulation is degraded from increased stresses, and long term breakdown studies have been started in 2024. This will ensure efficient and reliable electrification through increased implementation of fast power electronic converters.
The research will be conducted in cooperation with international industry within energy and electrification, and universities in Italy and Norway.
Electricity has been produced, transmitted, and consumed mainly in AC systems where the voltages vary sinusoidally at 50 or 60 Hz for more than a century. Design and testing of power components have been developed based on experience with AC. During the last decades power electronic converters (PECs) have been introduced. PECs are ideally suited to control and transform power – and permeate society for numerous applications: zero-emission transport, renewable energy, electrification, and transmission.
PECs function as on/off switches, resulting in sharp-edged square wave voltages with high repetition rate. These sharp voltage flanks introduce different stress distributions internally in components, that are challenging to model and predict. Today’s methods for modelling this are not accurate enough and new methods must be developed to map stresses accurately for components under PEC stress.
The increased stresses initiate new, or accelerate existing, failure modes in electric insulation. Partial discharges (PD), which are small electric sparks in voids and at defects in the electric insulation, is the dominating failure mechanism and is more severe under switched voltages. Understanding why and how is crucial for reliable electrification of society.
Qualification and testing are done under AC voltages because standardized methods are incompatible with switched voltages since high-frequency noise generated by PECs overlaps the frequency range used for conventional PD testing. Novel high-frequency methods for testing in factory and on-site must therefore be developed and communicated to standardization bodies.
The project will develop methods to measure and characterize the increased stresses on power components exposed to fast PECs. Acceleration of degradation, and initiation of new fault scenarios, that emerge due to PECs will be investigated and predictive models generated. This will ensure reliable components and grids contributing to electrification of society.