Intelligent Fault Diagnosis of Rolling Bearing via Deep-Layerwise Feature Extraction Using Deep Belief Network

Abstract

Rolling bearing is commonly used in rotating machinery and the rolling bearing fault diagnosis is of great significance to enhance the reliability of the rotating machinery. In this paper., an intelligent fault diagnosis method using deep belief network (DBN) via deep-Iayerwise feature extraction is proposed for rolling bearing fault identification. In this method, discrete wavelet packet transform is first used to calculate the original features from raw vibration signals. Due to information redundancy of the original features, the paper constructs a deep belief network with three hidden layers for deep-layerwise feature extraction and dimensionality reduction. Furthermore., the effectiveness of the proposed method is verified by two rolling bearing datasets and comparisons with the traditional intelligent fault diagnosis methods are also carried out. The result confirms that the proposed method is capable to detect the faults in rolling bearing and performs much better than the traditional intelligent fault diagnosis method.