Kun Yu

• PhD
• Lecturer at China University of Mining and Technology

In industrial scenarios, the source domain (SD) data typically encompasses condition monitoring (CM) data from all machines within a workshop or factory setting, while the target domain (TD) data may only include CM data from one or a small number of machines. The intelligent diagnostic method based on partial domain adaptation (PDA) represents a p...

Intelligent fault diagnosis plays an important role in maintaining the safe and reliable operation of rotating machinery. However, the data collected in real engineering scenarios may be severely insufficient, which presents challenges to the intelligent fault diagnosis methods. To address this problem, this paper introduces a metric-based meta lea...

With the development of deep learning methods, the data-driven fault diagnosis methods have attracted a great deal of interest. However, as for the data-driven fault diagnosis methods, technology has to overcome various difficulties in the practical industrial scenarios, such as variable working conditions, insufficient effective samples, and envir...

Cross-domain machinery fault diagnosis aims to transfer enriched diagnosis knowledge from a labeled source domain to a new unlabeled target domain. Most existing methods assume that the prior information on the fault modes of the target domain is known in advance. However, in engineering practice, prior knowledge of fault modes is rare in a new dom...

Cross-domain fault diagnosis based on transfer learning has been popularly developed to overcome inconsistent data distribution-caused degradation of diagnostic performance. However, the existing methods are typically suffering from class imbalance of domains and lacking sufficient fault data because it is difficult to obtain the real industrial da...

Unsupervised cross-domain fault diagnosis for rotating machinery is of great practical significance for real-world industrial scenarios; however, most existing methods are developed based on the vibration signal from a single sensor. With the increasing complexity of industrial systems, multisensor collaborative monitoring has played an important r...

The greedy-step $Q$ -learning (GQL) can effectively accelerate the $Q$ -value updating process. However, since it is an improved version of $Q$ -learning, the problem of $Q$ -value overestimation also exists. Since there are in total two max operators used to iteratively calculate $Q$ -value in GQL, many existing solutions to reduce the...

Fault diagnosis of rolling bearings has attracted extensive attention in industrial fields, which plays a vital role in guaranteeing the reliability, safety, and economical efficiency of mechanical systems. Traditional data-driven fault diagnosis methods require obtaining a dataset of full failure modes in advance as the training data. However, thi...

Intelligent fault diagnosis methods based on deep learning have attracted significant attention in recent years. However, it still faces many challenges, including complex and variable working conditions, noise interference, and insufficient valid data samples. Therefore, a novel deep transfer learning bearing fault diagnosis model is designed in t...

Although cross-domain fault diagnosis has received much attention in intelligent mechanical fault diagnosis, most existing methods only achieve knowledge transfer within the data from a single sensor. For a complex industrial system, multiple sensors are usually required to monitor its operating conditions coordinately. In such a situation, the fau...

Fault diagnosis of rolling bearing is essential to guarantee production efficiency and avoid catastrophic accidents. Domain adaptation is emerging as a critical technology for the intelligent fault diagnosis of rolling bearing. Most existing solutions learn domain-invariant features by statistical moment matching, adversarial training, or fusing tw...

Since the sample data after one exploration process can only be used to update network parameters once in on-policy deep reinforcement learning (DRL), a high sample efficiency is necessary to accelerate the training process of on-policy DRL. In the proposed method, a submartingale criterion is proposed on the basis of the equivalence relationship b...

Reassigned method can provide a high energy-concentrated time-frequency representation (TFR) result for the frequency-varying signal with linear group delay by reassigning the TF coefficients among both the time and frequency directions, whilst the TFR result provided by RM can’t be reconstructed since the reassigned operators in RM are conducted o...

Remaining useful life (RUL) prediction of rolling bearings is of paramount importance to various industrial applications. Recently, intelligent data-driven RUL prediction methods have achieved fruitful results. However, the existing methods heavily rely on the quality and quantity of the available data. For some critical bearings in industrial scen...

Generative adversarial networks (GANs) have demonstrated superior performances in image generation. In recent years, various improvements of network structure and learning theory related to GANs have undergone numerous advancement. Among these improvement techniques, the asymmetric training on the generator and discriminator networks has been widel...

Due to the changeable operating conditions of rotating machinery, the feature distributions of fault are usually changed. Most current cross-domain intelligent fault diagnosis methods only achieve global domain alignment, while ignoring the class discrepancy, resulting in the misclassification of the target domain samples near the class boundary. I...

Strong frequency-varying signal always undergoes a rapid frequency change within a short time duration. Converting its one-dimensional waveform into the two-dimensional time-frequency (TF) plane is convenient for us to understand the structure of the strong frequency-varying signal. Time-reassigned multisynchrosqueezing transform (TMSST) is a recen...

In current research works, a number of intelligent fault diagnosis methods have been proposed with the assistance of domain adaptation approach, which attempt to distinguish the health modes for target domain data using the diagnostic knowledge learned from source domain data. An important assumption for these methods is that the label information...

In this paper, the flexibility of bearing parts and localized defects are considered, and a method for calculating the radial time-varying stiffness of flexible cylindrical roller bearings (CRBs) with localized defects is proposed. The proposed method uses the finite element (FE) method to calculate the global flexibility of bearing parts and adopt...

Tooth fractures and cracks are common defects in gear transmission systems. Crack propagation paths of the spur gears are acquired using the fracture mechanics theory. Based on the simulated crack propagation path, the flexible body model of the cracked gear system is established in multibody dynamic software ADAMS. The vibration signals of cracked...

Limited condition monitoring data are recorded with label information in practice, which make the fault identification task more challenging. A semi-supervised learning (SSL) approach can be employed to increase the identification performance of the classifiers under such situation. In this study, a three-stage SSL approach using data augmentation...

Conventional intelligent diagnostic model is built on the foundation that the training data and testing data are recorded under the same operating condition, which neglects the fact that the operating condition of the rotating machinery usually varies. The feature distribution of the recorded data in one operating condition may be inconsistent with...

Deep learning has been widely used nowadays to achieve an automated fault diagnosis of rolling bearings. However, most of deep learning based bearing fault diagnosis methods are based on the assumption that the recorded samples are labeled data, though most of field data are recorded without label information. To address this issue, an effective se...

Transient impulsive feature detection is of vital importance in fault diagnosis of rolling bearing. However, the transient impulsive feature of rolling bearing is always heavily buried in the noise contaminated signal, which makes it difficult to be detected. Robust principal component analysis (PRCA) is an effective approach to exploit the underly...

In order to simulate the rubbing dynamics of the multi-blades/flexible casing, the finite element model (FEM) of the blisk-oval casing system with elastic supports is established using the self-programmed beam-shell-spring hybrid elements in combination with two types of self-programmed interfacial coupling elements (ICEs). The rotating effects suc...

In order to simulate the rubbing characteristics of the single blade-casing system with flexible supports, two types of the self-programmed elements ( i.e., two-node Timoshenko beam and spring elements) are adopted to establish corresponding finite element models (FEMs) in sequence. Then a node to surface contact algorithm is programmed to build th...

Though general parameterized time-frequency analysis is able to improve the energy concentration of time-frequency (TF) representation by the matched parameterized kernel, restricted by the structure of window function in short term Fourier transform, the energy diffusion phenomenon always occurred in the instantaneous frequency (IF). Synchrosqueez...

A rotating pre-twisted and inclined cantilever beam model (RPICBM) with the flapwise-chordwise-axial-torsional coupling is established with the Hamilton principle and the finite element (FE) method. The effectiveness of the model is verified via comparisons with the literatures and the FE models in ANSYS. The effects of the setting and pre-twisted...

Time-frequency analysis (TFA) technique is an effective approach to capture the changing dynamic in a nonstationary signal. However, the commonly adopted TFA techniques are inadequate in dealing with signals having a strong non-stationary characteristic or multi-component signals having close frequency components. To overcome this shortcoming, a ne...

An artificial intelligent bearing fault and hierarchical severity diagnosis framework is proposed in this study. The framework utilizes a combined deep belief networks (DBNs) and Dempster–Shafer (D-S) theory fault diagnosis scheme and adopts a two-stage approach in classifying (1) bearing fault conditions and (2) fault severities. The combined faul...

In the areas of measurement, instrumentation and sensing across science and engineering, the recorded signal is typically manifested as a strong non-stationary characteristic and the amplitude of each component is varied among one another. It is necessary to propose an effective signal processing algorithm that could clearly reflect the fast varyin...

Empirical wavelet transform (EWT) is an adaptive wavelet based analysis which can be used to extract useful amplitude modulated-frequency modulated (AM-FM) mono components from a bearing vibration signal. However, the pre-requisite segmentation method on the Fourier support of a signal without a rigorous theoretical principle has limited the applic...

A new fault diagnosis technique for rolling element bearing using multi-scale Lempel-Ziv complexity (LZC) and Mahalanobis distance (MD) criterion is proposed in this study. A multi-scale coarse-graining process is used to extract fault features for various bearing fault conditions to overcome the limitation of the single stage coarse-graining proce...

Angular contact ball bearing (ACBB) is widely used in rotating machinery for its great externalload capacity. The operating condition of ACBB is easily affected by surrounding mechanicalcomponents. The variations of contact angles and load distributions can sufficiently reflect theoperating condition of ACBB. A localized defect on the raceway will...

This paper employs a combined ensemble empirical mode decomposition (EEMD) and singular value decomposition (SVD) technique to extract useful fault features from the condition monitoring data of rolling element bearings. The fault features is then classified by a Fuzzy clustering method, Gath-Geva (GG) clustering, to obtain the cluster center and m...

Rolling element bearings play a crucial role in determining the overall health condition of a rotating machine. An effective condition-monitoring program on bearing operation can improve a machine’s operation efficiency, reduce the maintenance/replacement cost, and prolong the useful lifespan of a machine. This chapter presents a general overview o...