A method for identifying dielectric response parameters of stator windings insulation in pumped storage generator-motor based on modified SPDMD of depolarization current
•The paper proposes a dielectric-parameter method based on depolarization-current modal decomposition, selecting dominant modes.•The method automatically infers branch count in the EDM, minimizing prior knowledge and keeping physical relevance.•Adaptive filtering separates dominant relaxation modes...
Saved in:
| Published in: | Measurement : journal of the International Measurement Confederation Vol. 257; p. 118912 |
|---|---|
| Main Authors: | , , , , , |
| Format: | Journal Article |
| Language: | English |
| Published: |
Elsevier Ltd
15.01.2026
|
| Subjects: | |
| ISSN: | 0263-2241 |
| Online Access: | Get full text |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | •The paper proposes a dielectric-parameter method based on depolarization-current modal decomposition, selecting dominant modes.•The method automatically infers branch count in the EDM, minimizing prior knowledge and keeping physical relevance.•Adaptive filtering separates dominant relaxation modes from noise, enabling precise dielectric-parameter identification.•Robustness is verified via simulations and on-site tests by identifying depolarization currents under varied conditions.•It enables fast PDC-based insulation assessment of stator windings in pumped-storage plants, raising maintenance efficiency.
Accurately assessing the insulation condition of stator windings in pumped storage units is crucial for ensuring the safe and stable operation of power systems. However, the presence of substantial harmonics within the power plant renders traditional methods incapable of precisely diagnosing insulation condition of stator windings. To achieve accurate identification of dielectric response parameters, thereby enabling precise diagnosis of insulation condition of stator windings, this paper proposes a method for identifying the dielectric response parameters of stator windings insulation in pumped storage generator-motor based on a modified Sparse Promoting Dynamic Mode Decomposition (SPDMD) of depolarization current. Firstly, the conventional Dynamic Mode Decomposition (DMD) is modified by introducing a sparse correction process prior to modal decomposition, which effectively mitigates noise at the source and achieves better separation of modal components. Then, the Fast Iterative Shrinkage-Thresholding Algorithm (FISTA) is modified by incorporating an adaptive mechanism for automatically selecting iteration parameters based on the characteristics of the input data. This allows for data-driven sparse optimization, enabling the extraction of the dominant mode that characterizes the insulation relaxation behavior, while suppressing spurious modes induced by noise. Thirdly, dielectric response parameters are extracted by fitting the dominant mode to an extended Debye model (EDM), thereby enabling accurate parameter identification. Finally, the accuracy and applicability of the proposed approach are validated through both simulation experiments and on-site testing, demonstrating its effectiveness for insulation diagnosis under complex operating conditions. Simulation results show that the proposed method achieves 100 % accuracy in determining the number of EDM branches under SNR levels of 30–50 dB, with parameter identification errors (MAPE) under 10 %. In on-site tests, the method maintains R2 > 0.9918 and MAPE < 5 %, confirming its robustness and practical effectiveness. |
|---|---|
| ISSN: | 0263-2241 |
| DOI: | 10.1016/j.measurement.2025.118912 |