Rolling bearing fault detection using continuous deep belief network with locally linear embedding

Rolling bearing fault detection is of crucial significance to enhance the availability, the reliability and the security of rotating machinery. In this paper, a novel method called continuous deep belief network with locally linear embedding is proposed for rolling bearing fault detection. Firstly, a new comprehensive feature index is defined based on locally linear embedding to quantify rolling bearing performance degradation. Secondly, a continuous deep belief network (CDBN) is constructed based on a series of trained continuous restricted Boltzmann machines (CRBMs) to model vibration signals. Finally, the key parameters of the continuous deep belief network are optimized with genetic algorithm (GA) to adapt to the signal characteristics. The proposed method is applied to analyze the experimental bearing signals. The results demonstrate that the proposed method is more superior in stability and accuracy to the traditional methods.