Yan Wang

Yan Wang

Auto-generated profile, based on publicly available information

Yan Wang

Big Data Research Center, University of Electronic Science and Technology of China, Chengdu 611731, China

Research Focus & Metrics

Academic Career Length

11Years (2016-Present)

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

Representative Publications

View All Publications
9 July 2018 in Applied Physics Letters
Verified
Journal Article
Closed access
PDF file
YWY. Wang
Y. Wang
MHM. Hao
M. Hao
ZXZ. Xu
Z. Xu
DQD. Qiang
D. Qiang
GCG. Chen
G. Chen
AVA. Vaughan
A. Vaughan
The ability to suppress space charge accumulation at high electric fields makes nanocomposites attract significant research interest as potential insulation materials in high-voltage direct current cable development. At present, the deep trap introduced by nanoparticles is frequently applied to be responsible for the observed space charge suppression in nanocomposites. However, the experimental results that support deep-trap formation have not been rigorously examined. We therefore propose herein a simple and more direct approach based on isothermal surface potential decay combined with pulsed electro-acoustic measurements to verify the presence of deep traps in silica-based blend polyethylene nanocomposites. The results indicate that the deep traps are indeed introduced by filling nanosilica and the space charge suppression observed in the nanocomposite with a low loading ratio is caused by deeply trapped charges in the sub-surface region of specimens.

23 July 2019 in Journal of Physics D: Applied Physics
Verified
Journal Article
Closed access
PDF file
DQDayuan Qiang
Dayuan Qiang
YWYan Wang
Yan Wang
XWXinyu Wang
Xinyu Wang
GCGeorge Chen
George Chen
TAThomas Andritsch
Thomas Andritsch
Nanocomposites that exhibit good insulation properties have already attracted a variety of research and their electrical properties are believed to be related to charge dynamics in bulk of materials. However, it is still unclear on how nanofiller loading ratios, surface treatment and resultant changes in morphology will influence the charge dynamics of nanocomposites. In this paper, we have clearly mentioned the influence of adding nanoparticles into epoxy resins and the characteristics of the movement of charges in the materials based on combining analysis on morphology, DC conductivity and space charge measurements. The presence of spherical nanoparticles (SiO 2 ) introduced additional traps in bulk, which impaired the charge injection and reduced the mobility of charge carriers in samples of low filler loading ratios (e.g. 0.5 wt%). However, in silica-based samples of higher filler loadings, more nanoparticles further caused a higher density of traps, which resulted in lower average distance between arbitrary traps/inter-particle surface distances and thus charge carriers required less energy when moving from one to another by hopping or the quantum tunnelling mechanism. The surface treatment of SiO 2 particles introduced deep traps which helped the separation of particles or related traps, and to some extent restricted the transport of charge carriers. In addition, hBN particles seem to act as barriers to charge injection and movement due to the layered structures and large numbers of resultant shallow traps in bulk. In terms of the moisture effect, the presence of water led to an obvious increase in charge injection and mobility, and resulted in the higher mobility of charge carriers in both base materials and within traps/particles of nanocomposites. The existence of water shells around spherical particles could contribute to a higher probability of the quantum tunnelling process and the formation of conductive percolation channels.

28 October 2024
Verified
Conference Paper
Open access
WHWenhao Hu
Wenhao Hu
WCWeilong Chen
Weilong Chen
WYWeimin Yuan
Weimin Yuan
YWYan Wang
Yan Wang
Shimin CaiShimin Cai
Shimin Cai
Yanru ZhangYanru Zhang
Yanru Zhang
Social media has emerged as a vital platform for communication, information sharing, and acquisition. Predictive analysis of social media data has wide applications, such as sentiment examination and social network analysis. However, existing work often directly utilizes social media data for training, neglecting the issue of mismatched text and images. This neglect can lead to confusion about the contents, thereby affecting the identification of trending topics and the accuracy of social media predictions. In this paper, an approach named Dual-Stream Pre-training Transformer (DSPT) is introduced to address this gap. In DSPT, we use a Visual-Language Model (VLM) and a Language Model (LM) to separately learn from image and text data, mitigating the impact of text-image mismatches. Moreover, to enhance the understanding of the model to social media data, we conduct incremental pre-training for both models. To achieve better feature interaction, we construct an integrated regression module combining LightGBM and CatBoost, jointly predicting the extracted feature embeddings. This dual-stream multimodal feature extraction method improves the performance of predictive tasks. Experimental results validate the effectiveness of our approach, demonstrating its potential and providing deeper insights into multimodal data mining in social media.

23 August 2018 in Journal of Physics D: Applied Physics
Verified
Journal Article
Closed access
PDF file
YWYan Wang
Yan Wang
DQDayuan Qiang
Dayuan Qiang
ZXZhiqiang Xu
Zhiqiang Xu
GCGeorge Chen
George Chen
AVAlun S Vaughan
Alun S Vaughan
Nanodielectrics have been expected to improve the electrical performance and are considered as dielectrics for the future. It has been recognized that the electrical performance is closely related to charge dynamics in the dielectrics material. However, the mechanism of charge dynamics in the interphase of nanodielectrics has not been fully understood, which causes difficulty in understanding the effect of nanoparticle loading ratios and electric fields applied on the electrical properties. Recently, a model based on the tunneling process with the presence of deep traps has been suggested as one conceivable candidate for explaining charge dynamics in nanodielectrics, but the related experiment results are not discussed with the tunnelling process. In this paper, the measurements including isothermal surface potential decay and space charge are conducted for the blend polyethylene incorporated with the untreated silica nanocomposites. According to the experimental observation compared with the unfilled blend polyethylene, the electrical properties of nanocomposites with high loading ratios of 2 wt%, 5 wt% and 10 wt% are worsened, such as facilitated space charge accumulation and injection, and faster charge carriers transport. On the other hand, regarding the nanocomposites with the low loading ratios of 0.5 wt%, slow transport of charge carriers, and suppressed space charge accumulation and injection were observed. The effect of the lower and higher electric field on the electrical properties of the nanocomposites was similar for the low and high loading ratios. The tunnelling process associated with deep traps can effectively explain these observed phenomena of nanocomposites, it is therefore suggested for further explaining the electrical properties and charge dynamics in the nanodielectrics.

30 March 2023 in Applied Sciences
Verified
Journal Article
Open access
PDF file
YGYachun Gao
Yachun Gao
JGJia Guo
Jia Guo
CFChuanji Fu
Chuanji Fu
YWYan Wang
Yan Wang
Shimin CaiShimin Cai
Shimin Cai
Region of interest (ROI) segmentation is a key step in computer-aided diagnosis (CAD). With the problems of blurred tissue edges and imprecise boundaries of ROI in medical images, it is hard to extract satisfactory ROIs from medical images. In order to overcome the shortcomings in segmentation from the V-Net model or the level set method (LSM), we propose in this paper a new image segmentation method, the VLSM-Net model, combining these two methods. Specifically, we first use the V-Net model to segment the ROIs, and set the segmentation result as the initial contour. It is then fed through the hybrid LSM for further fine segmentation. That is, the complete segmentation of the V-Net model can be obtained by successively combining the V-Net model and the hybrid LSM. The experimental results conducted in the public datasets LiTS and LUNA show that, compared with the V-Net model or LSM alone, our VLSM-Net model greatly improves the sensitivity, precision and dice coefficient values (DCV) in 3D image segmentation, thus validating our model’s effectiveness.

4 November 2024 in IEEE Access
Verified
Journal Article
Open access
YWYan Wang
Yan Wang
LDLianbing Deng
Lianbing Deng
Infrared imaging emerges as a promising technique for vision tasks within environments characterized by lowor obscured visibility. However, the scarcity of infrared datasets, particularly those comprising paired infrared-visible images, addresses significant challenges for the training of high-performance deep learning models. This paper investigates the application of generative adversarial networks (GANs) for data augmentation in infrared imaging. Our work encompasses a range of GAN models including Pix2Pix, CycleGAN, StyleGAN3, and the proposed Cycle-aided model, which incorporates vision-aided loss to enhance both model training and stability. The results demonstrate that CycleGAN encounters stability issues and is prone to mode collapse, leading to less satisfactory performance. By contrast, the Cycle-aided model that leverages pre-trained models to substantially improve both the discriminative and generative capabilities of GANs, evidenced by a 51% increase in peak signal-to-noise ratio (PSNR), a 43% increase in structural similarity index (SSIM) and an 18% decrease in the Fréchet inception distance (FID) over CycleGAN. These improvements underscore the potential of GANs to improve data augmentation practices for infrared imaging in low-visibility environments. The insights gained also pave the way for future research aimed at developing architectures and training strategies of GANS for fully exploiting the unique properties of data augmentation.