Incipient fault diagnosis in power transformers by data-driven models with over-sampled dataset

Lopes, Sofia Moreira de Andrade; Flauzino, Rogério Andrade; Altafim, Ruy Alberto Corrêa

doi:10.1016/j.epsr.2021.107519

Cited by 38 publications

(18 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For geometric optimization calculations, a 5 × 5 × 1 k-point was adopted, and a more accurate k-point of 7 × 7 × 1 was used for electronic structure calculations [17]. Moreover, for electron orbit calculation in Dmol 3 , only 1 ×1 × 1 k points were available for selection [27].…”

Section: Methodsmentioning

confidence: 99%

“…There are still a considerable number of large and mediumsized transformers around the world that use transformer oil as a cooling and insulating medium [2]. Effectively detecting the operating status of oil-immersed transformers to discover insulation defects in advance and eliminate hidden troubles in time can help to ensure the stable running of a power system [3,4].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Adsorption and Sensing Properties of Dissolved Gas in Oil on Cr-Doped InN Monolayer: A Density Functional Theory Study

Qian

Jin²,

Wang³

et al. 2022

Chemosensors

View full text Add to dashboard Cite

Dissolved gas analysis (DGA) is recognized as one of the most reliable methods in transformer fault diagnosis technology. In this paper, three characteristic gases of transformer oil (CO, C2H4, and CH4) were used in conjunction with a Cr-decorated InN monolayer according to first principle calculations. The adsorption performance of Cr–InN for these three gases were studied from several perspectives such as adsorption structures, adsorption energy, electron density, density of state, and band gap structure. The results revealed that the Cr–InN monolayer had good adsorption performance with CO and C2H4, while the band gap of the monolayer slightly changed after the adsorption of CO and C2H4. Additionally, the adsorption property of the Cr–InN monolayer on CH4 was acceptable and a significant response was simultaneously generated. This paper provides the first insights regarding the possibility of Cr-doped InN monolayers for the detection of gases dissolved in oil.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Adsorption and Sensing Properties of Dissolved Gas in Oil on Cr-Doped InN Monolayer: A Density Functional Theory Study

Qian

Jin²,

Wang³

et al. 2022

Chemosensors

View full text Add to dashboard Cite

show abstract

“…The incipient faults are transient events that occur at random locations and are pre-emptive hidden faults before permanent faults occur. Incipient faults in the power distribution network usually occur in underground cables [5], [25], [26], transformer equipment [27], [28], distribution networks with high Distributed Energy Resources (DERS) penetration [29], and so on. There are many reasons for incipient faults, such as tree interference, animal contact and coil contact [14].…”

Section: A Faults In Power Distribution Systemsmentioning

confidence: 99%

Incipient Fault Detection in Power Distribution System: A Time–Frequency Embedded Deep-Learning-Based Approach

Luo

Cheng

et al. 2023

IEEE Trans. Instrum. Meas.

View full text Add to dashboard Cite

Incipient fault detection in power distribution systems is crucial to improve the reliability of the grid. However, the non-stationary nature and the inadequacy of the training dataset due to the self-recovery of the incipient fault signal, make the incipient fault detection in power distribution systems a great challenge. In this paper, we focus on incipient fault detection in power distribution systems and address the above challenges.In particular, we propose an ADaptive Time-Frequency Memory (AD-TFM) cell by embedding wavelet transform into the Long Short-Term Memory (LSTM), to extract features in time and frequency domain from the non-stationary incipient fault signals.We make scale parameters and translation parameters of wavelet transform learnable to adapt to the dynamic input signals.Based on the stacked AD-TFM cells, we design a recurrent neural network with ATtention mechanism, named AD-TFM-AT model, to detect incipient fault with multi-resolution and multi-dimension analysis. In addition, we propose two data augmentation methods, namely phase switching and temporal sliding, to effectively enlarge the training datasets. Experimental results on two open datasets show that our proposed AD-TFM-AT model and data augmentation methods achieve state-of-theart (SOTA) performance of incipient fault detection in power distribution system. We also disclose one used dataset logged at State Grid Corporation of China to facilitate future research.

show abstract

“…The authors of [ 12 ] proposed a data enhancement method for short-term voltage stability assessment (STVSA), which has achieved good results in processing small sample datasets. The authors of [ 13 ] used an over-sampled technology to balance the DGA data of the power transformer, and the effect was significantly improved compared with the original dataset. The above methods have achieved good expansion effects for unbalanced data.…”

Section: Introductionmentioning

confidence: 99%

Fault Voiceprint Signal Diagnosis Method of Power Transformer Based on Mixup Data Enhancement

Wan

Dong

Zhang

et al. 2023

Sensors

View full text Add to dashboard Cite

A voiceprint signal as a non-contact test medium has a broad application prospect in power-transformer operation condition monitoring. Due to the high imbalance in the number of fault samples, when training the classification model, the classifier is prone to bias to the fault category with a large number of samples, resulting in poor prediction performance of other fault samples, and affecting the generalization performance of the classification system. To solve this problem, a method of power-transformer fault voiceprint signal diagnosis based on Mixup data enhancement and a convolution neural network (CNN) is proposed. First, the parallel Mel filter is used to reduce the dimension of the fault voiceprint signal to obtain the Mel time spectrum. Then, the Mixup data enhancement algorithm is used to reorganize the generated small number of samples, effectively expanding the number of samples. Finally, CNN is used to classify and identify the transformer fault types. The diagnosis accuracy of this method for a typical unbalanced fault of a power transformer can reach 99%, which is superior to other similar algorithms. The results show that this method can effectively improve the generalization ability of the model and has good classification performance.

show abstract

Incipient fault diagnosis in power transformers by data-driven models with over-sampled dataset

Cited by 38 publications

References 18 publications

Adsorption and Sensing Properties of Dissolved Gas in Oil on Cr-Doped InN Monolayer: A Density Functional Theory Study

Adsorption and Sensing Properties of Dissolved Gas in Oil on Cr-Doped InN Monolayer: A Density Functional Theory Study

Incipient Fault Detection in Power Distribution System: A Time–Frequency Embedded Deep-Learning-Based Approach

Fault Voiceprint Signal Diagnosis Method of Power Transformer Based on Mixup Data Enhancement

Contact Info

Product

Resources

About