Lei Dong

Prof. Dr. Lei Dong

orcid

Research Focus & Metrics

Academic Career Length

8Years (2019-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
28 August 2019 in Materials
Verified
Journal Article
Open access
PDF file
RSRuimin Shi
Ruimin Shi
BWBukang Wang
Bukang Wang
ZYZhiwei Yan
Zhiwei Yan
ZWZongyan Wang
Zongyan Wang
LDLei Dong
Lei Dong
In order to explore the relationship between the surface topography parameters and friction properties of a rough contact interface under fluid dynamic pressure lubrication conditions, friction experiments were carried out. The three-dimensional surface topography of specimens was measured and characterized with a profile microscopy measuring system and scanning electron microscope. The friction coefficient showed a trend of decreasing first and then increasing with the increase in some surface topography parameters at lower pressure, such as the surface height arithmetic mean Sa, surface height distribution kurtosis Sku, surface volume average volume Vvv, and surface center area average void volume Vvc, which are the ISO 25178 international standard parameters. The effects of surface topographic parameters on friction were analyzed and the wear mechanism of the worn surface was presented. The wear characteristics of the samples were mainly characterized as strain fatigue, grinding, and scraping. The results provide a theoretical basis for the functional characterization of surface topography.

14 June 2022 in Entropy
Verified
Journal Article
Open access
PDF file
RSRuimin Shi
Ruimin Shi
BWBukang Wang
Bukang Wang
ZWZongyan Wang
Zongyan Wang
JLJiquan Liu
Jiquan Liu
XFXinyu Feng
Xinyu Feng
LDLei Dong
Lei Dong
Due to the influence of signal-to-noise ratio in the early failure stage of rolling bearings in rotating machinery, it is difficult to effectively extract feature information. Variational Mode Decomposition (VMD) has been widely used to decompose vibration signals which can reflect more fault omens. In order to improve the efficiency and accuracy, a method to optimize VMD by using the Niche Genetic Algorithm (NGA) is proposed in this paper. In this method, the optimal Shannon entropy of modal components in a VMD algorithm is taken as the optimization objective, by using the NGA to constantly update and optimize the combination of influencing parameters composed of α and K so as to minimize the local minimum entropy. According to the obtained optimization results, the optimal input parameters of the VMD algorithm were set. The method mentioned is applied to the fault extraction of a simulated signal and a measured signal of a rolling bearing. The decomposition process of the rolling-bearing fault signal was transferred to the variational frame by the NGA-VMD algorithm, and several eigenmode function components were obtained. The energy feature extracted from the modal component containing the main fault information was used as the input vector of a particle swarm optimized support vector machine (PSO-SVM) and used to identify the fault type of the rolling bearing. The analysis results of the simulation signal and measured signal show that: the NGA-VMD algorithm can decompose the vibration signal of a rolling bearing accurately and has a better robust performance and correct recognition rate than the VMD algorithm. It can highlight the local characteristics of the original sample data and reduce the interference of the parameters selected artificially in the VMD algorithm on the processing results, improving the fault-diagnosis efficiency of rolling bearings.

18 June 2025 in IEEE Transactions on Instrumentation and Measurement
Verified
Journal Article
Closed access
WRWeibo Ren
Weibo Ren
WZWang Zhijian
Wang Zhijian
ZCZhongxin Chen
Zhongxin Chen
SZShun Zhao
Shun Zhao
LDLei Dong
Lei Dong
XFXin Fan
Xin Fan
Due to the limitation of industrial conditions in production, raw sensor data are always shown as an unbalanced dataset, characterized by abundant normal operational data and scarce fault instances. This unbalance can degrade the performance of conventional fault diagnosis methods, leading to reduced accuracy and unstable model training. To address this challenge in bearing fault diagnosis, this paper proposes a Multi-Fault Feature Wasserstein Generative Adversarial Network (MFF-WGAN) to enhance diagnostic precision. First, the framework employs a multi-encoder denoising autoencoder architecture to mitigate noise interference in raw sensor data. Subsequently, the proposed MFF-WGAN integrates label information into its adversarial loss function to enable simultaneous generation of diverse fault categories, while incorporating inter-class feature discrepancies to refine sample quality. Finally, the developed multi-fault feature Wasserstein generation adversarial network is tested on the Case Western Reserve University bearing dataset and the laboratory bearing dataset. Computational results show that the proposed method can generate high-quality bearing samples with multiple faults effectively, which can obtain a higher diagnosis accuracy of 99.01% and 97.71% compared with the existing methods.

30 April 2022 in Lubricants
Verified
Journal Article
Open access
PDF file
RSRuimin Shi
Ruimin Shi
BWBukang Wang
Bukang Wang
JLJiquan Liu
Jiquan Liu
ZYZhiwei Yan
Zhiwei Yan
LDLei Dong
Lei Dong
With the increasing market demand for artificial hip joints, total hip joint replacement has gradually become an effective means of treating a series of hip joint diseases. In order to improve the service life of artificial hip joints, some new artificial hip joint materials, including polyetheretherketone (PEEK) and carbon fiber reinforced polyetheretherketone (CFR-PEEK), have been developed. In this paper, pin-on-plate wear tests under different cross-shear ratios and contact pressures were carried out to study the wear mechanism and worn surface topography of PEEK and CFR-PEEK. The experimental results showed that the wear of PEEK was associated with cross-shear, while CFR-PEEK was not. When the cross-shear ratio was 0.039 and contact pressure was 3.18 MPa, PEEK had poor wear resistance and its wear factor was about eight times that of ultra-high molecular weight polyethylene (UHMWPE). The wear resistance of CFR-PEEK had a significant advantage, since its wear factor was about 30% of that of PEEK. The wear factors of PEEK and CFR-PEEK increased as the contact pressure increased. The arithmetic average of the height amplitude of the surface, Sa, also increased gradually according to the topography of the worn surface. The wear mechanisms of PEEK and CFR-PEEK were scratching, plough cutting, and abrasion Since CFR-PEEK had good wear resistance and insensitivity to cross-shear motion, it is suitable for making artificial hip joints under low contact pressure condition.

5 March 2024 in Measurement Science and Technology
Verified
Journal Article
Closed access
PDF file
SCSenhao Chai
Senhao Chai
LDLei Dong
Lei Dong
WRWeibo Ren
Weibo Ren
YLYanfeng Li
Yanfeng Li
ZCZhongxin Chen
Zhongxin Chen
WZWang Zhijian
Wang Zhijian
Most existing deep learning methods consider the remaining useful life (RUL) prediction problem under a single failure mode and cannot solve the RUL prediction problem with multiple failure modes coexisting caused by component coupling in actual engineering systems. Thus, considering these issues, this paper proposes a novel tree network framework to address fault classification and RUL prediction in parallel, and the RUL prediction results are fused output, which are suitable for bearing RUL prediction with multiple faults. First, this paper develops a fault recognizer combining a frequency domain classifier and deep convolutional neural network to improve model selection accuracy. Secondly, this paper proposes a feature fusion algorithm based on the Gini coefficient, and the fused indicators are input into the RUL prediction sub-network for model training. Finally, the RUL sub-network prediction results are dynamically weighted and fused with the fault classification results to obtain the RUL based on SoftMax. The bearing dataset XJTU-SY is introduced to verify the efficiency of the proposed method, and computational results show that the developed framework can effectively predict RUL compared with other traditional methods, especially for RUL prediction under multiple failure modes.

26 February 2025 in Measurement Science and Technology
Verified
Journal Article
Closed access
YLYanfeng Li
Yanfeng Li
XDXia Duan
Xia Duan
WRWeibo Ren
Weibo Ren
LDLei Dong
Lei Dong
ZCZhongxin Chen
Zhongxin Chen
JWJunyuan Wang
Junyuan Wang
The two-stage Wiener process (WP) model has become a common method to describe the phased deterioration of bearings over time. However, this model ignores the correlation of feature data distribution structure and change points (CPs) between the two stages, as well as the limitations of maximum-likelihood estimation methods for WP model parameter estimation. Therefore, this paper proposes a remaining useful life prediction approach that integrates feature distribution CP identification and a martingale process. First, a two-step feature screening method adopting trend consistency and composite score is proposed to construct a health indicator, which accounts for the trend consistency of the same feature on different bearings and can avoid redundancy while containing sufficient degradation information. Next, a t-neighborhood granular mean-shift clustering method is proposed, which makes the divisibility of the feature distribution more obvious and can identify CPs sensitively, flexibly and stably. Finally, a martingale method is introduced so that the parameter estimation of the two-stage WP model depends on the entire degradation path, which overcomes the limitations of WP model parameter estimation and enables the model to better characterize the bearing degradation process.