Open AccessA fault diagnosis method based on the Support Vector Machine in rod pumping systems
Author(s) -
Juanni Li,
Jun Peng Shao
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/2125/1/012004
Subject(s) - sucker rod , support vector machine , fault (geology) , field (mathematics) , computer science , pattern recognition (psychology) , artificial intelligence , engineering , control engineering , mathematics , mechanical engineering , geology , seismology , pure mathematics
Monitoring the working status of the sucker rod pump is an important part in petroleum engineering. With the development of artificial intelligence technology, more methods have been applied to the fault diagnosis of rod pumping systems. An evolutional fault diagnosis method based on Support Vector Machine (SVM) in sucker rod pumping systems is proposed. Fourier descriptors and Light Field compression algorithm are used in this method to extract the graphic features of the indicator diagram. SVM is used to build fault classification model. This method is verified experimentally through data of indicator diagrams and the results show that it has a shorter training time and higher accuracy.