Figure - available from: Journal of Vibroengineering
This content is subject to copyright.
Source publication
Aiming at achieving early fault diagnosis and tracking the degradation process of bearings, we propose a novel monitoring methodology using a spectrum searching strategy in this paper. Firstly, a vibration signal is collected with appropriate sampling frequency and length. Secondly, the structural information of spectrum (SIOS) on a predefined freq...
Similar publications
![Figure 1. Structure of an autoencoder [7]](profile/Meng-Rao-3/publication/330475014/figure/fig1/AS:724168685404160@1549666677753/Structure-of-an-autoencoder-7_Q320.jpg)
Rotating machines are widely used in industry and often work under harsh and varying speed conditions. Fault diagnosis under varying speed conditions is needed to prevent major shutdowns. This paper aims to develop an intelligent rotating machinery fault diagnosis strategy based on deep neural networks (DNNs) and order tracking (OT). The developed...
Citations
... In an environment of manufactures and industry no machine without vibration, the vibrations signal indicate there is an error of the machine itself or in some parts, procedure and classify the type of the failure are identified by vibration analysis, the vibration analysis is one of the most common methods for determining the type and severity of any faults in components of machines (like bearings and gears), as well as any maintenance options related to the machine [9]. However, the classification and detection of early faults of the health condition of bearings remain as confused matters across various industries [10]. so, Yanfei et.al presented a novel diagnosis model using the "complementary ensemble empirical mode decomposition" (CEEMD) with kernel support vector machines (kernel SVM) to estimate the terms of defect severity for the condition of bearings [11] [12]. ...
The rotary machines are one of the main equipment in many industries even in the modern era of Industry 4.0. The rotary machines are part of precision manufacturing processes to advanced manufacturing. Bearings, gearboxes, rotors, and more are the most basic components of these rotary machines. With the continued use of rotary machine in the plant, these components tend to present catastrophic failures that could lead to economic, man power, and environmental losses. Hence, predicting the failures and raising the alarm are mandatory part of standard operating process for the industrial automation. The failures in the rotary system can be predicted by analyzing the parameters like vibration, speed of rotary machine obtained through the industrial IoT sensors. So, in this paper, we have employed different types of machine learning algorithms to predict four different bearing failures: (a) bearing health conditions (HC), (b) inner race fault (IF), outer race fault (OF), and (d) ball baring fault (BF). The comparison of results obtained from different machine learning algorithms shows that the SVM algorithm has a very high accuracy for bearing conditions with ball fault (accuracy = 100%) while the KNN approach gives the best performance in predicting all three bearing failures as compared to SVM, decision trees, ensemble, and neural network. The KNN provides 81–96% accuracy in predicting four different types of faults.
... It can be observed from (18) that weight coefficients in w * are assigned to globally spectral lines to obtain a final HI in each file number. The motivations of HI construction in (18) are given as follows: in many previous works, some HIs for early fault detection and degradation trajectory tracking were generated based on the sum of fault characteristic frequency and its harmonics [26]- [28]. However, such methods only considered the contributions of partial informative frequencies to early fault detection and degradation trajectory tracking. ...
Convex hulls based maximum margin classification has been widely studied for machine fault diagnosis, while its exploration for machine degradation modeling is seldom reported. In this study, a sparse and flexible convex-hull representation for machine degradation modeling is proposed to realize degradation trajectory tracking and fault diagnosis at a same time. First, considering using vibration data as health monitoring signals, globally normal and abnormal spectral lines can be obtained based on the fast Fourier transform and they are, respectively, characterized as individually flexible convex hulls. Subsequently, a sparse and flexible convex-hull representation degradation model is constructed by simultaneously finding the closest pair of samples and its sparse regularization between normal and abnormal convex hulls. Finally, a health indicator can be developed for early fault detection and degradation trajectory tracking during a machine life cycle. Meanwhile, quick fault diagnosis can be realized by finding a difference between the optimal closest samples in a normal convex hull and an abnormal convex hull. Two experimental cases are used to show the effectiveness and superiority of the proposed model to recent existing works.
