Open AccessRETRACTED: Identifying non-contact defects in fault bearing
Author(s) -
Maram Akila,
T. S. Mouleswaran,
Preeja Pradeep,
Prasanna Ramakrisnan
Publication year - 2021
Publication title -
iop conference series. materials science and engineering
Language(s) - English
Resource type - Journals
eISSN - 1757-899X
pISSN - 1757-8981
DOI - 10.1088/1757-899x/1166/1/012059
Subject(s) - bearing (navigation) , support vector machine , vibration , computer science , artificial intelligence , accelerometer , fault (geology) , pattern recognition (psychology) , data mining , engineering , physics , quantum mechanics , seismology , geology , operating system
Early diagnosis of bearing failures can save time, effort, and money on rotating machine maintenance. On this test, a non-touch type vibration pickup was designed and refined to capture vibration data for bearing fitness tracking under tight load and pace variations, avoiding the bodily connection of vibration pickup to the system tool. The signal was denoised and fault analysis was performed using a Hilbert rework. Principal Component Analysis (PCA) was used to reduce the dimensionality of the extracted capabilities, and then the chosen capabilities for lowering the amount of enter capabilities and discovering the maximum premier function set, the Sequential Floating Forward Selection (SFFS) method was used to rank them in order of significance. Finally, Support Vector Machines (SVM) and Artificial Neural Networks (ANN) were used to determine and classify the numerous faults in bearings. A comparison of SVM and ANN efficacy was carried out. The results reveal that vibration signatures from advanced non-touch sensors (NCS) correspond well with accelerometer data collected under the same conditions. The classification accuracy obtained by combining the advanced NCS with various sensors mentioned in the literature is comparable to that obtained by using the advanced NCS alone. The proposed method could be utilised to detect automated popularity system faults and issue early warnings, preventing unwelcome and unplanned device shutdowns due to bearing failure.