Power transmission line fault identification method and device based on Grubrum angle field and residual network

• Main Authors: DUAN YU, LIU HUANQING, LIN FENGKAI, NAN DONGLIANG, ZHAO QI, WANG CHANG, WANG JIAN, ZHANG LU, ZHAO ZHENMIN, GAO XING, CHEN JUN, PENG YINZHANG, OUYANG JINXIN, TAN JINLONG
• Format: Patent
• Language: Chinese; English
• Published: 22.12.2023
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; ELECTRIC DIGITAL DATA PROCESSING; PHYSICS

The invention relates to a power transmission line fault identification method based on a Gramb angle field and a residual network, and the method comprises the steps: constructing a fault type and fault cause identification model, and comprises the steps: constructing an original identification model based on a multi-target residual network; acquiring three-phase fault recording data of a power transmission line fault sample, and converting the acquired three-phase fault recording data into a two-dimensional color graph based on a Grubrum angle field to serve as a fault sample image; classifying the fault sample graphs according to fault types and fault reasons, and performing Smote oversampling processing on small sample types to obtain a fault sample set; and training and testing the original identification model based on the fault sample set to obtain a final identification model. According to the method, one-dimensional fault recording data is converted into a two-dimensional color image through a GAF me