Energies, Vol. 18, Pages 4991: Partial Discharge Characteristics of Typical Defects in Oil-Paper Insulation Based on Photon Detection Technology

Energies, Vol. 18, Pages 4991: Partial Discharge Characteristics of Typical Defects in Oil-Paper Insulation Based on Photon Detection Technology

Energies doi: 10.3390/en18184991

Authors:
Zhengyan Zhang
Yong Yi
Ji Qi
Qian Wang
Weiqi Qin
Xianhao Fan
Chuanyang Li

As a key equipment in the power system, the insulation state of oil-immersed transformer is directly related to the safe and stable operation of the power grid. To explore the feasibility of optical detection methods for detecting transformer insulation defects and further analyze the trend of partial discharge optical signal characteristics under typical oil-paper insulation defects in transformers, this paper proposes a method for detecting insulation defects in transformers based on photon detection technology. This method can not only reflect the periodicity and phase characteristics of photon signals, but also exhibits higher sensitivity compared to the traditional PRPD method. Firstly, the study builds an experimental platform for optoelectronic combined transformer partial discharge based on photon detection technology and carries out partial discharge simulation experiments on four typical insulation defect models through the step-up method to collect their pulse current signals and photon signals. Then, a phase-resolved photon counting (PRPC) method is proposed to analyze the signals during the development of partial discharges. Finally, the optical signal characteristics of the four defect models are extracted for comparative analysis. The results show that the optical signals of partial discharges can effectively respond to the generation and development process of partial discharges inside the transformer, and different types of insulation defects and development stages can be clearly distinguished according to the phase distribution characteristics and characteristic parameters of the optical signals. This study verifies the effectiveness of photon detection technology and provides a new effective tool for the detection of transformer oil-paper insulation defects.

More From Author

Energies, Vol. 18, Pages 4997: A Comparison of Deep Recurrent Neural Networks and Bayesian Neural Networks for Detecting Electric Motor Damage Through Sound Signal Analysis

Energies, Vol. 18, Pages 4994: Short-Term Forecasting of Unplanned Power Outages Using Machine Learning Algorithms: A Robust Feature Engineering Strategy Against Multicollinearity and Nonlinearity

Leave a Reply

Your email address will not be published. Required fields are marked *