Dongfeng Wang

Dongfeng Wang

Auto-generated profile, based on publicly available information

Dongfeng Wang

College of Mechanical Engineering, Zhejiang University of Technology, Hangzhou, China

Research Focus & Metrics

Academic Career Length

21Years (2006-Present)

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

Representative Publications

View All Publications
3 February 2022 in Lubricants
Verified
Journal Article
Open access
PDF file
JXJuan Xu
Juan Xu
SDShiyu Duan
Shiyu Duan
WCWeiwei Chen
Weiwei Chen
DWDongfeng Wang
Dongfeng Wang
Yuqi FanYuqi Fan
Yuqi Fan
In recent years, the development of deep learning-based remaining useful life (RUL) prediction methods of bearings has flourished because of their high accuracy, easy implementation, and lack of reliance on a priori knowledge. However, there are two challenging issues concerning the prediction accuracy of existing methods. The run-to-failure sequential data and its RUL labels are almost inaccessible in real-world scenarios. Meanwhile, the existing models usually capture the general degradation trend of bearings while ignoring the local information, which restricts the model performance. To tackle the aforementioned problems, we propose a novel health indicator derived from the original vibration signals by combining principal components analysis with Euclidean distance metric, which was motivated by the desire to resolve the dependency on RUL labels. Then, we design a novel self-attention augmented convolution GRU network (SACGNet) to predict the RUL. Combining a self-attention mechanism with a convolution framework can both adaptively assign greater weights to more important information and focus on local information. Furthermore, Gated Recurrent Units are used to parse the long-term dependencies in weighted features such that SACGNet can utilize the important weighted features and focus on local features to improve the prognostic accuracy. The experimental results on the PHM 2012 Challenge dataset and the XJTU-SY bearing dataset have demonstrated that our proposed method is superior to the state of the art.

12 June 2022 in Lubricants
Verified
Journal Article
Open access
PDF file
ZZZhe Zhang
Zhe Zhang
Yanshuang WangYanshuang Wang
Yanshuang Wang
JLJianghai Lin
Jianghai Lin
DWDongfeng Wang
Dongfeng Wang
The grease film thickness was measured in fully flooded elastohydrodynamic lubrication, and the influence of rolling speed, load, consistency, base oil type and thickener type on grease film thickness was analyzed. A new calculation model for grease film thickness was established. The results show that the grease film thickness increases with the increasing rolling speed, and then levels off with the amount of thickener in the contact region reaching an equilibrium. The degree of grease film enhancement comparing to its base oil will depend on thickener type and consistency. The larger the atmospheric viscosity and pressure-viscosity coefficient of the base oil, the higher the film thickness of the greases with the same thickener. The grease film thicknesses with the same base oil and different thickeners are determined by the size of thickener particles at the same consistency or concentration. The larger the consistence of the grease, the larger the effective viscosity of the grease at the contact and the thicker the grease film thickness whose base oil has the same type and viscosity along with the same type of thickener. The calculated values by the new model are in good agreement with the measured values.

29 March 2023 in Materials
Verified
Journal Article
Open access
PDF file
DWDongfeng Wang
Dongfeng Wang
JYJulong Yuan
Julong Yuan
HLHu Lai
Hu Lai
Binghai LyuBinghai Lyu
Binghai Lyu
The friction and wear performance of high-performance bearings directly affects the accuracy and maneuverability of weapons and equipment. In this study, high-speed, high-temperature, and heavy-load durability experiments of weapon bearings were carried out, and their wear properties (i.e., surface wear, metamorphic layer, scanning electron microscopy/energy-dispersive spectroscopy (SEM/EDS), residual stress, and retained austenite) were analyzed in multiple dimensions. The results showed the following: (1) The experimental temperature of the serviced front-end bearing is always lower than that of the rear bearing. (2) The metamorphic layer of the serviced rear bearing (i.e., inner ring, outer ring, rolling body, and cage) > the metamorphic layer of the serviced front-end bearing > the metamorphic layer of the unserviced bearing. (3) The rolling body of the rear bearing at high experimental temperatures contains not only elemental O, but also elemental P and Sr. (4) In the EDS analysis of the rolling elements, with the migration from the “ball edge” to the “ball center”, the elemental C in the rolling elements of serviced or unserviced bearings decreases slowly, while the elemental Fe content increases slowly.

30 April 2024 in IEEE Sensors Journal
Verified
Journal Article
Closed access
LSLiang Sun
Liang Sun
CCCong Cao
Cong Cao
GBGuoli Bai
Guoli Bai
ZZZhidan Zhong
Zhidan Zhong
WSWei Sun
Wei Sun
DWDongfeng Wang
Dongfeng Wang
Sensors are essential for the prognosis and health management of equipment. The fusion of multiple sensors data could improve the accuracy of fault diagnosis. Current data fusion methods focus on high-quality sensor data. However, in practical applications, constraints such as cost or space may limit the use of multiple high-quality sensors. Therefore, it is necessary to explore how to effectively use all available sensor data, even those containing some useful information but less effective when used individually. This paper presents a novel method for sensor data fusion, utilizing the highest-performing sensor data as a back bone and complementing it with preliminary data-level fusion data through multi-stage adaptive feature complementation. Comparative results validate the effectiveness and superiority of this method. The proposed method also demonstrates strong predictive capability by using data from a single period as training samples to forecast future data states.

1 January 2006 in Acta Physica Sinica
Verified
Journal Article
Open access
PDF file
WDWang Dong-Feng
Wang Dong-Feng
HPHan Pu
Han Pu
Proportional-integral-derivative(PID) control algorithm has the property of simplicity and practicability, but it is difficult to be tuned to control a complex nonlinear system. An advanced particle swarm optimization (PSO) was used to optimize the PID controller for controlling chaotic systems. Computer simulation of its application to several chaotic systems has been done. The controlled chaotic systems are the Hénon chaos, Duffing chaos, six-roll UC mill chaos, Nagumo-sato neuron chaos, Chen's chaos, and permanent magnet synchronous motor chaos. Th chaos control simulation results show that: 1) The output feedback based PID control for chaotic systems is effective; 2) using PID to control complex chaotic systems is feasible; and 3) using PSO to optimize the PID parameters for chaos control is effective and simple in programming.

3 June 2023 in Lubricants
Verified
Journal Article
Open access
PDF file
XPXiaoxu Pang
Xiaoxu Pang
DZDingkang Zhu
Dingkang Zhu
MQMing Qiu
Ming Qiu
DWDongfeng Wang
Dongfeng Wang
XWXinlong Wang
Xinlong Wang
When considering the problem of a vertical magnetic levitation bearing system, the rotor eccentric fall is more likely to cause the failure of the protective bearing. In this paper, a rotor drop collision model and a protective bearing dynamics model are constructed. It compares and analyzes the evolution of collision force values of the rotor eccentric drop as well as the non-eccentric drop. Further, this paper discusses the law of influence of three factors, rotor quality, rotational speed, and axial protection clearance, on the collision characteristics of the protected bearing in eccentric and non-eccentric cases. It has also experimentally verified this characteristic of rotor speed. The results show that compared with the non-eccentric condition, the axial impact force and radial impact forces of the rotor in the eccentric condition increase by 14% and 114%, respectively. Compared with the non-eccentric condition, with the increase in rotor quality, the axial and radial impact force increase by 68% on average, and the axial depth amplitude of the rotor increases by 350%. With the increase of rotor speed, the axial impact force without an eccentric drop is basically unchanged; the axial impact force of an eccentric drop increases slightly, and the radial impact force increases by 110%. With the increase of axial protection clearance, the radial displacement vibration of the rotor axis increases; the average increase of the maximum axial force is 120 N, and the average increase of the maximum radial force is 100%.