Classification of Superimposed Partial Discharge Patterns
Phase resolved partial discharge patterns (PRPD) are routinely used to assess the condition of power transformers. In the past, classification systems have been developed in order to automate the fault identification task. Most of those systems work with the assumption that only one source is active...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-04-01
|
| Series: | Energies |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1996-1073/14/8/2144 |
| _version_ | 1827695322334232576 |
|---|---|
| author | Benjamin Adam Stefan Tenbohlen |
| author_facet | Benjamin Adam Stefan Tenbohlen |
| author_sort | Benjamin Adam |
| collection | DOAJ |
| description | Phase resolved partial discharge patterns (PRPD) are routinely used to assess the condition of power transformers. In the past, classification systems have been developed in order to automate the fault identification task. Most of those systems work with the assumption that only one source is active. In reality, however, multiple PD sources can be active at the same time. Hence, PRPD patterns can overlap and cannot be separated easily, e.g., by visual inspection. Multiple PD sources in a single PRPD represent a multi-label classification problem. We present a system based on long short-term memory (LSTM) neural networks to resolve this task. The system is generally able to classify multiple overlapping PRPD by while only being trained by single class PD sources. The system achieves a single class accuracy of 99% and a mean multi-label accuracy of 43% for an imbalanced dataset. This method can be used with overlapping PRPD patterns to identify the main PD source and, depending on the data, also classify the second source. The method works with conventional electrical measuring devices. Within a detailed discussion of the presented approach, both its benefits but also its problems regarding different repetition rates of different PD sources are being evaluated. |
| first_indexed | 2024-03-10T12:23:47Z |
| format | Article |
| id | doaj.art-0f9e3ee9d4e347369d23caef505f9d1a |
| institution | Directory Open Access Journal |
| issn | 1996-1073 |
| language | English |
| last_indexed | 2024-03-10T12:23:47Z |
| publishDate | 2021-04-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Energies |
| spelling | doaj.art-0f9e3ee9d4e347369d23caef505f9d1a2023-11-21T15:11:30ZengMDPI AGEnergies1996-10732021-04-01148214410.3390/en14082144Classification of Superimposed Partial Discharge PatternsBenjamin Adam0Stefan Tenbohlen1Institute of Power Transmission and High Voltage Technology, University of Stuttgart, 70569 Stuttgart, GermanyInstitute of Power Transmission and High Voltage Technology, University of Stuttgart, 70569 Stuttgart, GermanyPhase resolved partial discharge patterns (PRPD) are routinely used to assess the condition of power transformers. In the past, classification systems have been developed in order to automate the fault identification task. Most of those systems work with the assumption that only one source is active. In reality, however, multiple PD sources can be active at the same time. Hence, PRPD patterns can overlap and cannot be separated easily, e.g., by visual inspection. Multiple PD sources in a single PRPD represent a multi-label classification problem. We present a system based on long short-term memory (LSTM) neural networks to resolve this task. The system is generally able to classify multiple overlapping PRPD by while only being trained by single class PD sources. The system achieves a single class accuracy of 99% and a mean multi-label accuracy of 43% for an imbalanced dataset. This method can be used with overlapping PRPD patterns to identify the main PD source and, depending on the data, also classify the second source. The method works with conventional electrical measuring devices. Within a detailed discussion of the presented approach, both its benefits but also its problems regarding different repetition rates of different PD sources are being evaluated.https://www.mdpi.com/1996-1073/14/8/2144partial dischargePDclassificationneural networksLSTM |
| spellingShingle | Benjamin Adam Stefan Tenbohlen Classification of Superimposed Partial Discharge Patterns Energies partial discharge PD classification neural networks LSTM |
| title | Classification of Superimposed Partial Discharge Patterns |
| title_full | Classification of Superimposed Partial Discharge Patterns |
| title_fullStr | Classification of Superimposed Partial Discharge Patterns |
| title_full_unstemmed | Classification of Superimposed Partial Discharge Patterns |
| title_short | Classification of Superimposed Partial Discharge Patterns |
| title_sort | classification of superimposed partial discharge patterns |
| topic | partial discharge PD classification neural networks LSTM |
| url | https://www.mdpi.com/1996-1073/14/8/2144 |
| work_keys_str_mv | AT benjaminadam classificationofsuperimposedpartialdischargepatterns AT stefantenbohlen classificationofsuperimposedpartialdischargepatterns |