Fault diagnosis of bearing based on WSN and data fusion algorithm of PCA-RBF
Abstract：In order to reduce data quantity and wireless sensor network(WSN) load of bearing fault diagnosis system used by wireless sensor and increase accuracy of fault diagnosis, this research proposes a WSN devices based on monitor system, the principal component analysis (PCA) and radial basis (RBF) artificial neural networks as a data fusion diagnosis algorithm. Firstly a 3-layer data fusion model based on LEACH is established, and then dimension reduc- tion of sensor data is operated by cluster header node, and lastly sink node accomplish decision- level fusion of data. Simulation results show that 3 member transmits 10 data packets respec- tively, while only 4 packets are remained after data fusion by sink node, so data fusion ratio is 86.7%, and accuracy of fault diagnosis is 85 %. The algorithm with a good recognition rate and high compression ratio can be well applied in fault monitoring of the equipment for coal mine.
Keywords：wireless sensor networkmechanical fault diagnosisdata fusionPCARBF neu- ral network
Publication Date：2012-01-01
Online Publishing Date：2025-08-15(First online date of this platform, not the publication date of the document)
Pages：8( 964-970,977 )