چکیده مقاله
A key indicator in bearing fault detection is the vibration pattern, which can reveal bearing anomalies In this study, these patterns are derived from data collected at the Bearing Data Center of Case Western Reserve University Fault detection is achieved using ARMA autoregressive moving average models ARMA models are well established for the detection and prediction of time series and are employed here for fault detection In this study, ARMA models with orders 6,6 , 9,9 , and 12,12 were trained on 1,990 healthy bearing vibration samples at 1,772 rpm These models were then used to detect faults at 0 178 mm 0 007 in , 0 355 mm 0 014 in , and 0 711 mm 0 028 in in bearing balls at 1,779 rpm The RMS root mean square error between the model forecasted data and the actual vibration data for eight points was calculated Faults around 0 711 mm 0 028 in could be detected by large errors approximately 200% , while errors in the 10 15% range are often associated with faulty bearings with fault sizes between 0 178 mm 0 007 in and 0 355 mm 0 014 in Errors below 6% are typically indicative of healthy bearings, indicating effective fault detection
کلیدواژهها
نویسندگان
Sheyda Panahi
School of Mechanical Engineering, Shiraz University
Meysam Jokar
School of Mechanical Engineering, Shiraz University
شیوه ارجاع
Panahi, Sheyda and Jokar, Meysam,1404,Bearing Fault Detection Using ARMA Models,The First National Conference on Applied Technologies in Mechanical Engineering (ATME2025),Ahvaz
ارائهشده در
مجموعه مقالات اولین کنفرانس ملی فناوری های کاربردی در مهندسی مکانیک27 آبان 1404 · اهواز