ying wu

Dr. ying wu


Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

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

Representative Publications

View All Publications
13 March 2025 in Scientific Reports
Verified
Journal Article
Open access
PDF file
YWying wu
ying wu
ZZZhiyuan Zang
Zhiyuan Zang
XZXitao Zou
Xitao Zou
WLWentao Luo
Wentao Luo
NBNing Bai
Ning Bai
WDWei Dong
Wei Dong
The digital revolution in power systems has increased their complexity and interconnectivity, thereby exacerbating the risk of cyberattacks. To protect critical power infrastructure, there is an urgent need for an advanced intrusion detection system capable of capturing the intricate interactions within smart grids. Although traditional graph neural network (GNN)-based methods have exhibited substantial potential, they primarily rely on network data (e.g., IP addresses and ports) to construct the graph structure, failing to adequately integrate physical data from power grid devices. Moreover, these methods typically employ fixed activation functions in downstream deep networks, which limits the accurate representation of complex nonlinear attack patterns, thereby reducing detection accuracy. To address these challenges, this paper introduces GraphKAN, a novel intrusion detection framework that leverages graph attention network (GAT) and Kolmogorov–Arnold network (KAN) to enhance detection precision in smart grids. GraphKAN firstly constructs a graph-structured representation with power devices, information technology devices, and communication network devices as nodes, and integrates the physical connections and logical dependencies among infrastructure elements as edges, providing a comprehensive view of device interactions. Furthermore, the GAT module utilizes multi-head attention mechanisms to dynamically allocate node weights, extracting global features that encompass both feature information and interaction patterns. The KAN introduces learnable activation functions based on parameterized B-splines, enhancing the nonlinear expression of the global features extracted by GAT and significantly improving the detection accuracy of complex attack patterns. Experiments conducted on datasets obtained from Mississippi State University and Oak Ridge National Laboratory demonstrate that GraphKAN achieves detection accuracies of 97.63%, 98.66%, and 99.04% for binary, ternary, and 37-class intrusion detection tasks, respectively. These results represent substantial improvements over state-of-the-art models, including GA-RBF-SVM, BGWO-EC, and Net_Stack, with accuracy gains of 5.73%, 0.89%, and 3.52%, respectively. The findings underscore the efficacy of GraphKAN in enhancing intrusion detection accuracy in smart grids and its robust performance in complex attack scenarios.

31 May 2021 in Micromachines
Verified
Journal Article
Open access
PDF file
YZYuan Zhai
Yuan Zhai
YXYi Xiang
Yi Xiang
WYWeiqing Yuan
Weiqing Yuan
GCGang Chen
Gang Chen
JSJinliang Shi
Jinliang Shi
YWying wu
ying wu
High sensitivity detection of terahertz waves can be achieved with a graphene nanomesh as grating to improve the coupling efficiency of the incident terahertz waves and using a graphene nanostructure energy gap to enhance the excitation of plasmon. Herein, the fabrication process of the FET THz detector based on the rectangular GNM (r-GNM) is designed, and the THz detector is developed, including the CVD growth and the wet-process transfer of high quality monolayer graphene films, preparation of r-GNM by electron-beam lithography and oxygen plasma etching, and the fabrication of the gate electrodes on the Si3N4 dielectric layer. The problem that the conductive metal is easy to peel off during the fabrication process of the GNM THz device is mainly discussed. The photoelectric performance of the detector was tested at room temperature. The experimental results show that the sensitivity of the detector is 2.5 A/W (@ 3 THz) at room temperature.

30 June 2022 in Measurement Science and Technology
Verified
Journal Article
Closed access
ZYZhi Yang
Zhi Yang
YZYuan Zhai
Yuan Zhai
BDBo Deng
Bo Deng
HYHao Yin
Hao Yin
ZLZhuang Lu
Zhuang Lu
YWying wu
ying wu
To improve the linear range of measurement, a restrictor is usually added to the main flow channel of a capillary-tube-type thermal mass flowmeter. However, as the linear range of the flowmeter increases, the singularity will occur in the low mass flow velocity section of the flowmeter calibration data, which will lead to a significant decrease in the measurement accuracy in this section. To address this issue, first, this paper developed a microelectromechanical system (MEMS) thermal gas flowmeter based on the principle of capillary-tube-type thermal mass flow. Then, the effect of the bypass ratio on its performance was further investigated using quantitative analysis and was verified using Fluent numerical analysis. In addition, to explain the singularity, the bypass ratio-sensitivity relationship model was built. Based on this model, the bypass ratio can be changed by adjusting the restrictor parameters to optimize the performance of the MEMS thermal gas flowmeter. Moreover, six numerical simulation models of restrictors with different aperture diameters and aperture numbers were designed. The numerical analysis results show that within a certain range of aperture diameters and aperture numbers, the singularity is more easily weakened when the aperture diameter decreases or the aperture number increases. Finally, the restrictor with the optimum aperture number and suitable aperture diameter was selected as the test model for calibration, and the goodness of fit R 2 of the calibration results is improved from 0.9937 to 0.9972, the singularity in the low mass flow velocity section is significantly weakened, and the sensitivity is improved.

4 July 2022 in IEEE Access
Verified
Journal Article
Open access
ZYZhi Yang
Zhi Yang
YZYuan Zhai
Yuan Zhai
YXYi Xiang
Yi Xiang
JWJianquan Wu
Jianquan Wu
JSJinliang Shi
Jinliang Shi
YWying wu
ying wu
The success of convolutional neural networks (CNNs) benefits from the stacking of convolutional layers, which improves the model’s receptive field for image data but also causes a decrease in inference speed. To improve the inference speed of large convolutional network models without sacrificing performance indicators too much, a data-aware adaptive pruning algorithm is proposed. The algorithm consists of two parts, namely, a channel pruning method based on the attention mechanism and a data-aware pruning policy based on reinforcement learning. Experimental results on the CIFAR-100 dataset show that the performance of the proposed pruning algorithm is reduced by only 2.05%, 1.93% and 5.66% after pruning the VGG19, ResNet56 and EfficientNet networks, respectively, but the speedup ratios are 3.63, 3.35, and 1.14, respectively, and the comprehensive pruning performance is the best. In addition, the generalization ability of the reconstruction model is evaluated on the ImageNet dataset and FGVC Aircraft dataset, and the performance of the proposed algorithm is the best, which shows that the proposed algorithm learns data-related information in the pruning process, that is, it is a data-aware algorithm.

21 September 2021 in Chemistry and Technology of Fuels and Oils
Verified
Journal Article
Closed access
PDF file
BHBo Huang
Bo Huang
TPTao Peng
Tao Peng
CXChenyang Xia
Chenyang Xia
YZYuan Zhai
Yuan Zhai
JSJinliang Shi
Jinliang Shi
YWying wu
ying wu

30 April 2024 in Semiconductor Science and Technology
Verified
Journal Article
Closed access
FZFuzhong Zheng
Fuzhong Zheng
ZLZhonghao Li
Zhonghao Li
BHBo Huang
Bo Huang
HZHongbing Zhu
Hongbing Zhu
YWying wu
ying wu
The influence of trap effects on carrier transport characteristics in quantum dot (QD) thin films is the subject of study, aiming to provide a theoretical basis for the structural design and performance improvement of QD thin film optoelectronic devices. This study presents a specific mathematical description of capturing and releasing charges by traps, which includes the time-varying equation for captured charges. Utilizing the carrier hopping transport model, a system of partial differential equations is employed as the physical field, establishing hopping transport models that account for both shallow traps and a combination of shallow and deep traps. Simulations based on specific experimental samples reveal that the presence of traps introduces asymmetry in the diffusion motion of charge carriers, extending the duration of the photocurrent signal and resulting in the capture of charges, along with a reduction in the peak value of the current signal. The model also simulates carrier transport characteristics under the influence of repetitive light pulses, demonstrating distinct patterns in capturing and releasing charges for both shallow and deep traps.