@article{StoettnerRauscherEndisch2022, author = {St{\"o}ttner, Julia and Rauscher, Andreas and Endisch, Christian}, title = {Pareto optimization of multilevel inverter structures regarding the DC magnitude, switching frequency and switching angles}, volume = {2022}, pages = {108259}, journal = {International Journal of Electrical Power \& Energy Systems}, number = {142, Part A}, publisher = {Elsevier}, address = {Amsterdam}, issn = {0142-0615}, doi = {https://doi.org/10.1016/j.ijepes.2022.108259}, year = {2022}, language = {en} } @article{RauscherHufnagelEndisch2022, author = {Rauscher, Andreas and Hufnagel, Manuel and Endisch, Christian}, title = {Pareto optimization of wavelet filter design for partial discharge detection in electrical machines}, volume = {2022}, pages = {112163}, journal = {Measurement}, number = {205}, publisher = {Elsevier}, address = {Amsterdam}, issn = {1873-412X}, doi = {https://doi.org/10.1016/j.measurement.2022.112163}, year = {2022}, abstract = {A reliable detection of partial discharges (PDs) in the insulation system of electrical machines is essential to prevent premature breakdowns. Wavelet filters (WFs) are often recommended for denoising this PD measurement data. However, filtering performance strongly depends on numerous settings, which impedes an optimum filter choice. We propose an approach to systematically select the best filter for denoising PD signals. First, representative PD pulses are determined from measurement data by an averaging and a singular value decomposition method. Next, realistic benchmark signals are generated for filter assessment. Further, all possible WF combinations from a set of different mother wavelet selection methods, decomposition trees, and thresholding approaches are evaluated. On the basis of two contradicting objectives, namely mean square error of noise intervals and pulse amplitude distortion, Pareto optimization is applied to choose optimal WF settings. The corresponding filters dominate several common WF proposals from previous literature.}, language = {en} } @article{RauscherKaiserDevarajuetal.2024, author = {Rauscher, Andreas and Kaiser, Johannes and Devaraju, Manoj and Endisch, Christian}, title = {Deep learning and data augmentation for partial discharge detection in electrical machines}, volume = {133}, pages = {108074}, journal = {Engineering Applications of Artificial Intelligence}, number = {Part A}, publisher = {Elsevier}, address = {Amsterdam}, issn = {1873-6769}, doi = {https://doi.org/10.1016/j.engappai.2024.108074}, year = {2024}, abstract = {Fault testing in the production line of automotive traction machines is essential to ensure the desired lifetime. Since repetitive partial discharges (PDs) caused by anomalies in the insulation system lead to premature breakdowns of electrical machines, a reliable PD detection is of great importance. This paper proposes deep learning (DL) methods to improve the discrimination of PD from background noise in comparison with the state-of-the-art amplitude based PD detection in the production line. First, a systematic data extraction and labeling procedure is introduced to obtain correctly labeled datasets from arbitrary PD measurements. In addition, datasets are enhanced with low signal-to-noise ratio PD pulses by applying a special data augmentation approach. 13 different convolutional, recurrent and fully connected neural networks are compared for various time-frequency representations of the input signals. Hyperparameters for input transform, network topology and solver are optimized for all 13 combinations to ensure a fair case study. As a result, the two-dimensional convolutional neural network with continuous wavelet transform achieves the best accuracy of around 99.76\% on a test dataset of PD signals originating from previously not utilized test objects. All DL models considered in this comparison outperform the state-of-the-art threshold-based PD classification. Even for PD events with an amplitude close to the noise level, the detection rate is still around 95\% for the best network. Furthermore, without applying the proposed data augmentation procedure, the DL models investigated are not able to distinguish small PD pulses from noise.}, language = {en} }