ARAndreas Rauscher

Andreas Rauscher

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Andreas Rauscher has a total of 10 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Convolutional Neural Networks and Thresholding Approaches for Single and Multi-Sensor Detection of Partial Discharges in Electrical Machine Stators

    by Andreas Rauscher, Julian Braun, Rainer Hiemer, Marcelo Lobo Heldwein, Christian Endisch

    The repetitive occurrence of partial discharges (PDs) in electrical machines degenerates the insulation system and causes a premature breakdown. Consequently, PD tests at the end of the production line in accordance with the standard IEC/TS 61934 are mandatory for automotive hairpin stators to ensure that the repetitive PD inception voltage (RPDIV) meets the minimum requirements for a PD-free operation over the product lifetime. However, the test results strongly depend on the sensitivity of the utilized sensor and the PD detection algorithm. This paper investigates various single and multi-sensor-based approaches to increase the PD detection rate while ensuring robustness to noise interference. Different convolutional neural network (CNN) architectures are trained to classify between PD and noise. A profound case study for rating the detection methods is designed utilizing artificial multi-channel test signals closely representing real measurements. The CNN models outperform the conventional thresholding approach based on bandpassfiltered signals, particularly under low signal-to-noise conditions and for noise with pulse-shaped interferences. A sensor fusion with a logical OR operation increases the PD detection rate compared to the single-sensor evaluations of the Horn antennas and the high-frequency current transformer. All classifiers are successfully validated under different noise conditions and for stators not incorporated into the training process. A final RPDIV comparison with the state-of-the-art PD tester on real-world data verifies an increased detection sensitivity of the CNNs and the proposed fusion concepts, in particular for stators with small PD pulses around the inception level.

  • Investigation of Factors Influencing the End-of-Line Partial Discharge Testing of Electrical Machines

    by Andreas Rauscher, Peer Stenzel, Christian Endisch

    Recurrent partial discharges (PD) in the insulation system of electrical machines result in accelerated insulation degeneration and a premature breakdown. Determining the repetitive PD inception voltage (RPDIV) of stators for automotive traction machines at the end of the production line following IEC/TS 61934 is mandatory to ensure a PD-free operation over the service life. The RPDIV depends on various influencing factors, which impedes the comparability of RPDIV results from different test systems. This paper introduces a PD detection platform to systematically investigate the RPDIV differences between two exemplary commercial PD testers. The proposed fusion between the PD signals received by a Horn antenna and a high-frequency current transformer is a suitable reference with a higher PD detection capability than the state-of-the-art systems. Three main influencing factors are identified to explain the RPDIV differences of up to 15 % between the commercial PD testers. The first one is the sensitivity of the antennas for PD detection, which is affected by the antenna design, the relative position to the stator, and the frequency band utilized in the signal processing. Second, differences in the rise time of the high-voltage surge pulses between 220 and 670 ns do not result in a consistent overall RPDIV trend. However, the RPDIV tends to increase with a longer rise time for some stators. Finally, measuring the test voltage directly at the stator terminals or estimating the actual voltage by correction factors is essential to avoid RPDIV differences caused by an inaccurate voltage measurement.

  • Optimization of Sensor Setup and Filter Frequency for End-of-Line Partial Discharge Testing of Electrical Machines

    by Andreas Rauscher, Peer Stenzel, Christian Endisch

    Repetitive partial discharges (PDs) occurring within minor faults in the insulation system of an electrical machine accelerate insulation degeneration and cause premature failure. Therefore, reliable PD testing is crucial at the end of the production line for automotive traction machines to ensure the required service life. However, the PD detection capability and the test results depend on the sensitivity of the utilized sensors. This paper thoroughly assesses various sensor setups, including ultrahigh frequency (UHF) antennas and a high-frequency current transformer (HFCT), to identify the optimal test configuration for automotive hairpin stators. First, an optimization procedure is introduced to determine the most appropriate frequency range for each sensor to detect PD. In addition, an analysis of noise interference is conducted. Secondly, the signal-to-noise ratio (SNR) and the repetitive partial discharge inception voltage (RPDIV), as defined in IEC/TS 61934, are compared for the HFCT and different antennas at varying positions and distances. The HFCT has been found to have the highest sensitivity, resulting in lower RPDIV values in most test runs. Broadband UHF antennas provide a higher SNR than the antenna of the commercial PD tester, which only utilizes a narrow frequency band for PD detection. The sensitivity of all antennas under consideration increases when the distance to the test object is reduced. Synchronous acquisition with two identical antennas at different positions or a combined evaluation of the HFCT and an antenna promises a more reliable PD detection for a few test objects. Further, the choice of the sensor cannot systematically influence the RPDIV variation of several consecutive test runs, as the stochastic nature of PD occurrence causes this variation.