Meng Zhan, Chen Qing, Li Hongbin, See Chan Hwang
School of Electrical and Electronic Engineering, Huazhong University of Science and Technology, Wuhan 430074, China.
School of Engineering and The Built Environment, Edinburgh Napier University, Edinburgh, United Kingdom.
Rev Sci Instrum. 2020 Jan 1;91(1):014705. doi: 10.1063/1.5123438.
The internal insulation condition of capacitor voltage transformers (CVTs) is a key influence factor that affects their measurement performance and safe operation. However, the internal insulation would age along with long-time operation and degrade due to environmental factors, and once the insulation degradation grows, serious damage and even explosion may happen in CVTs; hence, it is necessary to monitor the internal insulation condition of CVTs, and the fault type and fault degree need to be identified. In this paper, a data-driven internal insulation condition identification method for CVTs is proposed. Both the amplitude and phase of the output voltage of CVTs are collected, and then, recognition models based on the combination of the output voltages and distribution topology of CVTs in substations are built. A possibilistic fuzzy clustering method is used to monitor the internal insulation condition of CVTs, and different types and different degrees of insulation faults could be identified effectively. Finally, the proposed method is verified in several cases; not only the preset typical faults in the method could be identified effectively but also the faults beyond the preset faults could be diagnosed.
