Condition monitoring data of ball bearing based on wavelet analysis model with machine learning techniques

<ns3:p>This research paper cites the outcomes of complex Morlet wavelet analysis, artificial intelligence techniques, thermography analysis, and SKF-Machine condition advisors for monitoring the conditions of deep groove ball bearings under several faults in different components. The complex Morlet wavelet approach is presented as a crucial technique to specify a complete picture regarding the condition of the deep groove ball bearing in conjunction with Fast Fourier Transform (FFT) analyzer vibration data in the initial stage of the inception of the fault. Subsequently, this study implements two Artificial Intelligence (AI) techniques, artificial neural networks (ANN) and support vector machines (SVM), for the identification and classification of bearing faults more effectively in a multi-fault environment. Finally, onsite condition monitoring tools such as thermal imaging cameras and machine condition advisors are found to be potential tools for the indication of defects in deep groove ball bearings in a cost-effective manner and to validate the result after a certain level of severity of the fault. The experimental results showed that the proposed system using a complex Morlet wavelet and support vector machine is effective.</ns3:p>