Bearings are some of the most critical industrial parts and are widely used in various types of mechanical equipment. Bearing health status can have a significant impact on the overall equipment performance, and bearing failures often cause serious economic losses and even casualties. Thus, estimating the remaining useful life (RUL) of bearings in real time is of utmost importance. This paper proposes a data-driven RUL prediction method for bearings based on Bayesian theory. First, time-domain features are extracted from the bearing vibration signal and data are fused to build a health indicator (HI) and a state model of bearing degradation. Then, according to Bayesian theory, a Bayesian model of state parameters and bearing life is established. The parameters of the Bayesian model are updated and bearing RUL is predicted by the Metropolis–Hastings algorithm. The method was validated by the XJTU-SY bearing open datasets and the prediction results are compared with the existing methods. Accuracy of the proposed method was demonstrated.
The deep neural network is widely applied in remaining useful life prediction because of its strong feature extraction ability. However, the prediction results of deep learning neural networks are often influenced by random noise and modeling parameters. Besides, the training process of the traditional neural network is time-consuming. In order to overcome these drawbacks, a novel bootstrap ensemble learning convolutional simple recurrent unit method is proposed for remaining useful life prediction. The simple recurrent unit is used to learn the time-series features of sensor data, which can effectively reduce the model parameters and boost the calculation speed. Moreover, the remaining useful life prediction uncertainty can be quantified with the prediction interval, which can be calculated by the ensemble learning convolutional simple recurrent unit model. The prediction performance of the ensemble learning convolutional simple recurrent unit model is demonstrated with the turbofan engine dataset. The experimental results show that the proposed ensemble learning convolutional simple recurrent unit model provides a prognosis framework with better prediction performance for quantifying RUL prediction uncertainty.
Turning tool is a critical part of numerical control machining, and its reliability directly affects machining efficiency and stability of the entire system. Turning tool life reliability analysis has major theoretical and practical significance. Laboratory test data or information monitoring are one of commonly employed methods for assessing mechanical performance. This information is conducive for determining mechanical property distribution and assessing the structural reliability. Therefore, experimental monitoring data is incorporated into turning tool life reliability analysis to obtain adequate evaluation results. Considering tool wear process uncertainty, reliability model based on Taylor tool life equation is proposed. Approximate Bayesian theory is introduced to update reliability model parameters. According to obtained parameter posterior samples, tool life reliability is analyzed via Monte Carlo simulation. Effectiveness of the proposed method is validated against experimental samples of turning tool wear. Results of tool life reliability analysis are in accordance with the actual working conditions, which provides a theoretical basis for the selection of numerical control machining process parameters and tool replacement strategies.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.