Cast-resin dry-type transformer partial discharge signal analysis using spectral correlated empirical mode decomposition method

Tang, Ya-Wen; Su, Chean-Cheng; Tai, Cheng‐Chi; Chen, Jiann-Fuh

doi:10.1109/i2mtc.2013.6555472

Cited by 2 publications

(1 citation statement)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recently, while the incidence of accidents due to manufacturing defects has decreased, there is an increase in degradation due to thermal, mechanical, and electrical caused by continuous loads [8], [9]. Therefore, research on PD for early detection and analysis of accidents caused by manufacturing defects or deterioration is crucial [10].…”

Section: Introductionmentioning

confidence: 99%

Classification of PRPD Pattern in Cast- Resin Transformers Using CNN and Implementation of Explainable AI (XAI) With Grad-CAM

Kim,

Jung,

Hwang

et al. 2024

IEEE Access

View full text Add to dashboard Cite

Cast-resin transformers are affected by deterioration due to manufacturing defects and continuous load. Studying PD, which is capable of detecting defects or degradation in advance, is important. With the rapid advancement of AI technologies, research on PD classification using CNN models is being actively conducted. However, due to the black box problem, it is impossible to explain the reasoning behind the learning outcomes. Therefore, relying solely on predictive outcomes of learning for PD classification raises issues of reliability. Recent studies in various fields are progressing with the application of XAI to address the black box issue of CNNs, aiming to identify the criteria used for making predictions. However, research on applying XAI in AI-based PD classification is currently insufficient. Therefore, further study on the implementation of XAI is necessary. In this paper, an excellent CNN model was applied to image classification for PD classification of cast-resin transformers, and the grad-cam model was used for XAI. This approach proposes a method for humans to comprehend the rationale behind the learning outcomes. The data used for training consists of artificial defects under laboratory conditions and noise measured in castresin transformers via UHF sensors. PD and noise classification due to defects was performed, and the reasons for successful and failed results were analyzed through XAI. Consequently, it was observed that the application of XAI to CNN models leads to the construction of a more reliable model. INDEX TERMSCast-resin transformer, PD, pattern classification, convolution neural network (CNN), explainable artificial intelligence (XAI), gradient weighted class activation mapping (Grad-CAM)

show abstract

Section: Introductionmentioning

confidence: 99%

Classification of PRPD Pattern in Cast- Resin Transformers Using CNN and Implementation of Explainable AI (XAI) With Grad-CAM

Kim,

Jung,

Hwang

et al. 2024

IEEE Access

View full text Add to dashboard Cite

show abstract

Typical Partial Discharge Signal Acquisition and Feature Extraction in High Voltage Switchgear

Zeng

2024

IJCSIT

View full text Add to dashboard Cite

With the progress of power system technology, how to effectively evaluate the reliability of high-voltage switchgear equipment has become an important part of power grid power supply and industrial distribution. Switchgear is used in power system power generation, transmission, power distribution, power conversion and consumption play a role in switching, control or protection, voltage level in 3.6kV-550kV electrical products, usually contains a variety of switches, circuit breakers, contactors and other components. In the process of use, due to various reasons, partial discharge may occur. Partial discharge refers to a partial discharge phenomenon in insulating media, which is usually caused by defects in the medium or excessive local electric field intensity. In switchgear, PD may be caused by aging of equipment, deterioration of insulation materials, dust accumulation and other reasons. Pd generates discharge current and discharge heat, which may damage the insulation performance of the equipment or even cause a fire. Therefore, it is very important to detect and deal with the partial discharge phenomenon in the switchgear in time. Therefore, timely detection of partial discharge of equipment can deal with partial discharge problems by checking equipment, cleaning equipment, timely replacement of aging components, and other methods to avoid losses caused by equipment failure due to partial discharge. Due to the complex distribution site environment and the interference of wireless communication signals from other equipment, the real high-frequency signals of partial discharge will be interfered, which will affect the accuracy of insulation performance evaluation of electrical equipment in the switchgear. Therefore, how to correctly collect partial discharge signals and how to identify the characteristics of partial discharge signals will be an important part of partial discharge detection in switchgear. At present, the commonly used partial discharge signal feature recognition is mainly based on the typical signal simulated by Matlab and artificially added noise for processing. Due to the wide frequency band of local discharge and the complex field environment, the real collected waveform will be different from the typical discharge signal. The original signal is filtered and detected for processing and analysis.

show abstract

Cast-resin dry-type transformer partial discharge signal analysis using spectral correlated empirical mode decomposition method

Cited by 2 publications

References 8 publications

Classification of PRPD Pattern in Cast- Resin Transformers Using CNN and Implementation of Explainable AI (XAI) With Grad-CAM

Classification of PRPD Pattern in Cast- Resin Transformers Using CNN and Implementation of Explainable AI (XAI) With Grad-CAM

Typical Partial Discharge Signal Acquisition and Feature Extraction in High Voltage Switchgear

Contact Info

Product

Resources

About