Intelligent Fault Diagnosis of Rolling Element Bearing Based on Convolutional Neural Network and Frequency Spectrograms

Effective fault diagnosis of rolling element bearing is vital for the reliability and safety of modern industry. Although traditional intelligent fault diagnosis technology such as support vector machine, extreme learning machines and artificial neural network might achieve satisfactory accuracy, expert knowledge and manual intervention are heavily relied on in the process of feature extraction and selection. In this paper, a novel fault diagnosis method based on deep learning is proposed for rolling bearing using convolutional neural networks (CNN) and frequency spectrograms. First of all, fast Fourier transform is used to extract frequency features from raw 1-D vibration signals and convert them into 2-D frequency spectrograms. Then, the extracted 2-D frequency spectrograms are inputted into the CNN model to achieve fault diagnosis of rolling bearing, which makes full use of the strong ability of CNN in image classification. Finally, a case study is carried out to demonstrate the proposed method. The results show that it can achieve higher accuracy than traditional methods. Moreover, its performance in stability is very good as well.