Multiple partial discharge source discrimination with multiclass support vector machines
作者：
Highlights：
• Different types of partial discharges are created with test objects in laboratory.
• Their frequency content depends on the type of discharge and other external factors.
• An SVM extracts characteristics from the power spectral density of the pulses.
• Noise, corona, internal and surface discharges have different characteristics.
• The differences are used to classify discharges and separate them from noise.
摘要
•Different types of partial discharges are created with test objects in laboratory.•Their frequency content depends on the type of discharge and other external factors.•An SVM extracts characteristics from the power spectral density of the pulses.•Noise, corona, internal and surface discharges have different characteristics.•The differences are used to classify discharges and separate them from noise.
论文关键词：Support vector machine,Partial discharges,Electric maintenance,Machine learning,Condition monitoring,Risk assessment
论文评审过程：Received 2 March 2015, Revised 8 February 2016, Accepted 9 February 2016, Available online 24 February 2016, Version of Record 9 March 2016.
论文官网地址：https://doi.org/10.1016/j.eswa.2016.02.014
