Liang Sun

Liang Sun

Auto-generated profile, based on publicly available information

Liang Sun

Dalian University of Technology

Research Focus & Metrics

Academic Career Length

10Years (2017-Present)

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

Representative Publications

View All Publications
4 December 2023 in IEEE Sensors Journal
Verified
Journal Article
Closed access
GBGuoli Bai
Guoli Bai
WSWei Sun
Wei Sun
CCCong Cao
Cong Cao
DWDongfeng Wang
Dongfeng Wang
QSQingchao Sun
Qingchao Sun
LSLiang Sun
Liang Sun
Bearings are vital components of a rotary machine. Real-time fault diagnosis of bearings has great significance in the maintenance of equipment. Current vibration-based fault diagnosis methods rely on the usage of long time series data to reduce the influence of noise but often suffer from imbalances in datasets, significantly affecting diagnosis accuracy. In this paper, a fault diagnosis method based on Intertemporal Return Plot and data augmentation is proposed. The Intertemporal Return Plot is employed to transform one-dimensional time series data into two-dimensional images. The Wasserstein Generative Adversarial Network is employed to generate synthetic images for data argumentation. This approach helps reduce the influence of data imbalance, thereby improving the accuracy and convergence speed of the fault diagnosis. Comparisons with existing well-known methods are performed to demonstrate the effectiveness of the proposed method. The results show that the proposed method achieves a high accuracy of fault diagnosis for short time series data. An inverse transformation is also proposed to convert the images to time series data.

1 November 2020 in Aerospace Science and Technology
Verified
Journal Article
Closed access
WYWeigang Yao
Weigang Yao
SMSimão Marques
Simão Marques
TRTrevor Robinson
Trevor Robinson
CACecil Armstrong
Cecil Armstrong
LSLiang Sun
Liang Sun

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.

13 May 2021 in Journal of Computational Design and Engineering
Verified
Journal Article
Open access
LSLiang Sun
Liang Sun
CACecil G Armstrong
Cecil G Armstrong
TRTrevor T Robinson
Trevor T Robinson
DPDimitrios Papadimitrakis
Dimitrios Papadimitrakis
Automatic quadrilateral (quad) or hexahedral (hex) multiblock decomposition has been a topic of research for many years. The key challenges are to automatically determine where to place mesh singularities and how to generate a decomposition based on the mesh singularities to get the desired mesh orientation and distribution. In this work, a new idea of achieving these is proposed based on an auxiliary subdivision of the domain into smaller subdomains, followed by applying an equation, which calculates the net number of mesh singularities of a surface, to locate the quad mesh singularities. Under this idea, two different methods are presented based on the medial axis and the inward boundary offset. Both methods are conformal to the vertex classifications of the original domain, which guarantees a good mesh quality at the boundary. The mesh results are compared with a paving method and a cross-field method.

26 August 2024 in IEEE Transactions on Instrumentation and Measurement
Verified
Journal Article
Closed access
GBGuoli Bai
Guoli Bai
WSWei Sun
Wei Sun
GTGuangyao Tian
Guangyao Tian
Qingchao SunQingchao Sun
Qingchao Sun
LSLiang Sun
Liang Sun
The development of digital twins and smart bearings in bearing industry demands efficient computational methods for factors such as strain analysis. Symbolic regression offers a promising solution by efficiently deriving analytical expressions from limited data, offering low computational cost, rapid speed, high accuracy, and strong interpretability. In this paper, symbolic regression is applied to the strain calculation problem of radially loaded hollow cylindrical rollers. The resulting expressions are evaluated and selected based on physical insights obtained from theoretical analysis and physical experiments. Compared to other methods, the selected expression retains interpretability and generalization capabilities while achieving superior computational accuracy.

27 November 2017 in Engineering with Computers
Verified
Journal Article
Open access
PDF file
LSLiang Sun
Liang Sun
CTChristopher M. Tierney
Christopher M. Tierney
CACecil G. Armstrong
Cecil G. Armstrong
TRTrevor T Robinson
Trevor T Robinson
Hexahedral (hex) meshes can be generated for CAD models of complex thin-walled components by isolating thin-sheet and long-slender regions, quad meshing one of the bounding faces, and sweeping the quad mesh through the volume to create hex elements. Continuing the work in Sun et al. (Proc Eng 163:225–237, 2016 ), where an improved approach to thin-sheet identification was presented, an enhanced automatic method for long-slender region identification is proposed in this paper. The objective is to improve the efficiency and decomposition quality compared to existing long-slender region identification processes. Geometric measures such as the edge length and the face width are employed to generate sizing measures, which are in turn used to identify candidate long-slender regions. Careful consideration is given to the generation and positioning of the cutting faces required to isolate the long-slender regions by assessing a priori quad mesh quality on the wall faces. It is shown that a significant reduction in the number of degrees of freedom (DOF) can be achieved by applying anisotropic hex elements on the identified regions.