الفهرس 
INTRODUCTIONOnly 14 pages are availabe for public view
 |  | from | 79 |  |  |
| | | from | 79 | | |
Abstract
Rolling element bearings (REBs) are vital components in the rotating machinery which take part in many industries. Early faults in the bearing can hardly be detected using conventional vibration signal analysis techniques especially when working in a noisy environment. This issue has been addressed by many researchers to develop more advanced techniques that can process the raw vibration signal and alleviate the noise and other interferences in the signal. State of art techniques are developed for this purpose such as wavelet transform, minimum entropy de-convolution (MED), empirical mode decomposition (EMD) and local mean decomposition (LMD). Variational mode decomposition (VMD) method is recently introduced as a signal processing technique for fault identification of rolling element bearings. However, the selection of VMD parameters namely the number of modes (k) and the quadratic penalty factor (α) still represents a challenge to obtain proper decomposition modes with the most relevant and denoised fault information. This thesis presents a framework using sailfish optimization (SFO) algorithm and Gini index (GI) as a criterion to adaptively select the optimum VMD parameters for each bearing fault signal. The proposed algorithm is tested using three case studies of faulty bearings, and the most appropriate mode containing fault information is automatically extracted based on maximum GI values. The obtained results indicate a high efficiency of the proposed method in extracting fault feature and in exclusion of noise effect as compared to conventional fixed-parameter VMD, local mean decomposition (LMD), and empirical mode decomposition (EMD). In addition, REBs working in a variable speed environment are exposed to frequency smearing in the vibration spectrum. It made it necessary to combine the VMD with the computed order tracking (COT) technique to overcome this phenomenon and resample the signal in equal angles instead of time so that the vibration signal can easily be analyzed