Litcius/Paper detail

Graph Cardinality Preserved Attention Network for Fault Diagnosis of Induction Motor Under Varying Speed and Load Condition

Yao Tang, Xiaofei Zhang, Guojun Qin, Zhuo Long, Shoudao Huang, Dianyi Song, Haidong Shao

2021IEEE Transactions on Industrial Informatics59 citationsDOI

Abstract

During the long-term operation of motors, their working conditions are changing due to the industrial demands or declining health status, and traditional diagnosis methods perform poorly in that case. This article proposes a fault diagnosis method based on graph cardinality preserved attention network (GCPAT), which can work under varying working conditions, and can be generalized to the transient state. Diagnosis results are obtained by analyzing signal-converting graphs, which are composed of nodes and edges. First, the vibration signals are converted into symmetrical snowflake images by symmetrized dot pattern (SDP) method. Second, SLIC is developed to make homogeneous super-pixels in SDP images as nodes, and form graphs according to color, texture, and distance features. Finally, the GCPAT is utilized to distinguish motor status. Compared with other state-of-art methods, the results show the out-performance of GCPAT under varying working conditions both in steady and transient state.

Topics & Concepts

Cardinality (data modeling)GraphComputer scienceTransient (computer programming)Fault (geology)HomogeneousVibrationPixelState (computer science)Induction motorArtificial intelligenceControl theory (sociology)Pattern recognition (psychology)AlgorithmMathematicsEngineeringData miningTheoretical computer scienceVoltageSeismologyElectrical engineeringGeologyCombinatoricsPhysicsOperating systemQuantum mechanicsControl (management)Machine Fault Diagnosis TechniquesMachine Learning in BioinformaticsIndustrial Vision Systems and Defect Detection
Graph Cardinality Preserved Attention Network for Fault Diagnosis of Induction Motor Under Varying Speed and Load Condition | Litcius