Aerospace electrification has to go through significant innovations regarding adjustable speed drives, which are needed to increase specific power and effectiveness of propulsion and power control systems. This involves electrical insulation design with high-electric field and, therefore, increased risk of accelerated insulation aging. This risk becomes dramatic if highly-energetic degradation phenomena, such as partial discharges, PD, incept in (unavoidable) insulation defects. However measuring partial discharges, that is, the property associated with the fastest electrical degradation mechanism in organic insulation, has to face an increasing amount of issues compared to sinusoidal AC voltage supply. Referring to power electronics for rotating machines drives, the main difficulty lies with the extraction of partial discharge pulses from the noise produced by switch commutation, especially when the slew rate is fast and the operating voltage is high, compared to the threshold for partial discharge inception. Since partial discharges tend to occur mostly during voltage rise and fall times, thus simultaneously to switching noise, poor techniques for noise rejection and partial discharge identification may affect significantly any quality, commissioning and diagnostic test performed on electrical devices and apparatus supplied by power electronic waveforms. This paper presents a novel approach to partial discharge and noise separation, which can work also for very fast switch rise times, and increased operating voltage and frequency. Based on experimental testing, this paper shows that considering the time domain, rather than the frequency domain on which noise filtering is commonly based, and resorting to signal variance evaluation, partial discharge pulses can be extracted effectively from commutation noise.
Cheetham P., Bosworth M., Montanari G.C., Ghosh R., Seri P. (2020). A novel approach for noise rejection and partial discharge identification in electric drives for aerospace. American Institute of Aeronautics and Astronautics Inc, AIAA [10.2514/6.2020-3599].
A novel approach for noise rejection and partial discharge identification in electric drives for aerospace
Montanari G. C.;Ghosh R.;Seri P.
2020
Abstract
Aerospace electrification has to go through significant innovations regarding adjustable speed drives, which are needed to increase specific power and effectiveness of propulsion and power control systems. This involves electrical insulation design with high-electric field and, therefore, increased risk of accelerated insulation aging. This risk becomes dramatic if highly-energetic degradation phenomena, such as partial discharges, PD, incept in (unavoidable) insulation defects. However measuring partial discharges, that is, the property associated with the fastest electrical degradation mechanism in organic insulation, has to face an increasing amount of issues compared to sinusoidal AC voltage supply. Referring to power electronics for rotating machines drives, the main difficulty lies with the extraction of partial discharge pulses from the noise produced by switch commutation, especially when the slew rate is fast and the operating voltage is high, compared to the threshold for partial discharge inception. Since partial discharges tend to occur mostly during voltage rise and fall times, thus simultaneously to switching noise, poor techniques for noise rejection and partial discharge identification may affect significantly any quality, commissioning and diagnostic test performed on electrical devices and apparatus supplied by power electronic waveforms. This paper presents a novel approach to partial discharge and noise separation, which can work also for very fast switch rise times, and increased operating voltage and frequency. Based on experimental testing, this paper shows that considering the time domain, rather than the frequency domain on which noise filtering is commonly based, and resorting to signal variance evaluation, partial discharge pulses can be extracted effectively from commutation noise.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.