Entity

Time filter

Source Type


Qin W.-L.,Beihang University | Zhang W.-J.,Beihang University | Lu C.,Science and Technology on Reliability and Environment Engineering Laboratory
Vibroengineering Procedia | Year: 2015

This paper presents a rolling bearing fault diagnosis approach based on the combination of Ensemble Empirical Mode Decomposition (EEMD), Information Entropy (IE) and Random Forests (RF). The horizontal and vertical vibration signals of the bearings are utilized as the input of the method. First, the signals, after preprocess, are decomposed into certain number of intrinsic mode functions (IMF) using EEMD. Second, the IEs of the IMFs are calculated as the features for further fault diagnosis. Third, the selected features are adopted to train the random forests model using 10-fold cross validation. Fourth, the trained RF model is used to conduct bearing fault diagnosis. To verify the effectiveness of the proposed approach, three types of faults including inner-ring fault, outer-ring fault and rolling element fault are considered and data from two individual experiments are used. The results demonstrate that the approach has desirable diagnostic performance both for cylindrical roller bearing and deep groove ball bearing. © JVE International Ltd. Source


Guo J.-B.,Beihang University | Guo J.-B.,Science and Technology on Reliability and Environment Engineering Laboratory | Du S.-H.,Beihang University | Wang X.,Beijing Institute of Environment Features | And 2 more authors.
Xi Tong Gong Cheng Yu Dian Zi Ji Shu/Systems Engineering and Electronics | Year: 2015

The failure propagation in dynamic systems is driven by discrete component fault events, continuous processes as well as their interactions. This hybrid feature of the fault propagation brings about the difficulty in fault cognition and modeling. Existing researches regard faults just as discrete events. Thus they only focus on analyzing how the discrete system failures are caused by random component failures. However, the continuous processes in fault propagation have always been ignored for the sake of simplifications in engineering, which leads to the inaccuracy of the description of dynamic system failures. This paper defines the hybrid failure propagation within two dimensions for the dynamic system, and analyzes its hybrid factors and propagation features. For accurate description of these hybrid features, a hybrid failure propagation modeling method is proposed based on the hybrid theory which is used to model the interaction of discrete events and continuous processes. The proposed method is applied to a temperature control system. The simulation results show the hybrid feature of the failure propagation, as well as the feasibility of the presented modeling method. ©, 2015, Chinese Institute of Electronics. All right reserved. Source


Du S.,Beihang University | Guo J.,Science and Technology on Reliability and Environment Engineering Laboratory | Guo J.,Beihang University | Zhao Z.,Beihang University | And 2 more authors.
Proceedings of 2014 Prognostics and System Health Management Conference, PHM 2014 | Year: 2014

Reliability Sensitivity Analysis aims at quantifying the influence of the change of the uncertain model inputs on the model reliability. The reliability sensitivity measures which are calculated through reliability sensitivity analysis will help a lot because they are used to identify the importance of the statistical characteristics of the uncertain parameters e.g. means, standard deviations, etc. In this paper, a new method is proposed for reliability sensitivity analysis of the mechanism. An airborne retractable mechanism is used to test our new method. Good agreement is observed between the results from the proposed method and the results calculated based on a large number of Monte Carlo simulations, which proves that the proposed method is applicable and efficient for the engineering cases. © 2014 IEEE. Source


Liu Z.,Science and Technology on Reliability and Environment Engineering Laboratory | Zeng S.,Science and Technology on Reliability and Environment Engineering Laboratory | Zeng S.,Beihang University | Guo J.,Science and Technology on Reliability and Environment Engineering Laboratory | Guo J.,Beihang University
Proceedings of 2014 Prognostics and System Health Management Conference, PHM 2014 | Year: 2014

The reliability analysis of systems may be very time-consuming by traditional sampling algorithms such as Monte Carlo especially for dynamic systems, because the states of dynamic systems are time-dependent. We address this main problem of dynamic systems with a new reliability analysis method based on stochastic reachability which has been studied in the field of control theory and control engineering. Randomness in dynamic systems will be described by this approach. It is involved from the following two aspects. One is to use jumps involving probability distributions when settling the time automata model. This adds the possibility to represent component failures. Another aspect is to put stochastic differential equation components inside the dynamic system models. By doing that, the impact of external disturbances such as environmental temperature can be taken into consideration. © 2014 IEEE. Source


Luo M.,Science and Technology on Reliability and Environment Engineering Laboratory | Zeng S.,Science and Technology on Reliability and Environment Engineering Laboratory | Zeng S.,Beihang University | Guo J.,Science and Technology on Reliability and Environment Engineering Laboratory | And 2 more authors.
Proceedings of 2014 Prognostics and System Health Management Conference, PHM 2014 | Year: 2014

This paper describes a reliability analysis approach which is called F-VMEA. Traditional FMEA is insufficient to consider detrimental variations which lead to soft failure, so, the proposed F-VMEA compensates for the deficiencies. Hence, the objective of this article is to show how to simultaneously consider failure and variation in whole product development process. F-VMEA starts with FMEA and processes in parallel with VMEA, hence can derive significantly cooperative effect for predicting system reliability finally. During the concept design phase, integrating FMEA with basic or enhanced VMEA is reasonable and effective for limited information. With the data, information and knowledge increased, the FMECA is combined with probabilistic VMEA in the detailed design stage. This approach not only concentrates on failure but also fully pays more attention on unwanted variation, then, the causes which lead to soft or hard failures can be roundly founded in the design stage. This paper outlines the detailed implementation process of F-VMEA The feasibility of this method is demonstrated by specific steps of implementation. Finally, the jet pipe electrohydraulic servo valve is used to verify the F-VMEA method. © 2014 IEEE. Source

Discover hidden collaborations