Open AccessGIS partial discharge defect diagnosis system and method based on Extreme Learning Machine
Author(s) -
Qi Tang,
Guowei Li,
Junbo Wang,
Rongbo Luo,
Lihui Wu
Publication year - 2021
Publication title -
iop conference series. earth and environmental science
Language(s) - English
Resource type - Journals
eISSN - 1755-1307
pISSN - 1755-1315
DOI - 10.1088/1755-1315/632/4/042005
Subject(s) - partial discharge , extreme learning machine , computer science , artificial intelligence , principal component analysis , artificial neural network , identification (biology) , pattern recognition (psychology) , feature vector , feature (linguistics) , set (abstract data type) , data mining , voltage , engineering , electrical engineering , linguistics , philosophy , botany , biology , programming language
Partial discharge type identification is of great significance to the diagnosis of insulation faults in high-voltage power equipment. The partial discharge type recognition method based on deep learning map diagnosis has the disadvantages of large memory space and high hardware environment requirements. This paper presents a GIS discharge defect diagnosis system and method based on Extreme Learning Machine (ELM), which can be deployed on the edge layer equipment. The principal component analysis method is used to obtain the principal parameters that characterize the defects and form a sample feature data set. The sample feature data set is used to train the neural network model and continuously optimize it. Finally, the trained neural network model is used to identify the types of partial discharge defects. This method can effectively solve the problems that the partial discharge type identification method based on atlas diagnosis occupies large memory space and requires high hardware environment, which is convenient for the wide application of engineering.