Yun Sha

Dr. Yun Sha


Research Focus & Metrics

Academic Career Length

20Years (2007-Present)

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

Representative Publications

View All Publications
1 June 2007 in 2007 IEEE Intelligent Vehicles Symposium
Verified
Conference Paper
Closed access
YSYun Sha
Yun Sha
ZGZhang Guo-Ying
Zhang Guo-Ying
YYYang Yong
Yang Yong
Road detection is one of the most important branches of road following. In this paper we propose a classification-based road detection algorithm by boosting. To fully utilize potential region feature correlations and improve the accuracy of classification, this algorithm introduces the feature combination method into road detection. First, an over-completed feature set is constructed on several linear and non-linear combined functions. Second, a correlation feature set is selected from the over-completed feature set by feature selection algorithm. Then, the boosting, the support vector machine and the random forest classifiers are used to evaluate the correlation feature set and the raw feature set. The results of the experiment shows the performance of boosting classifier based on the correlation feature set provides the best outcome.

23 September 2024 in Sensors
Verified
Journal Article
Open access
PDF file
LJLiu Jiayi
Liu Jiayi
YSYun Sha
Yun Sha
WZWenchang Zhang
Wenchang Zhang
YYYong Yan
Yong Yan
LXLiu Xue-Jun
Liu Xue-Jun
Anomaly detection in industrial control system (ICS) data is one of the key technologies for ensuring the security monitoring of ICSs. ICS data are characterized as complex, multi-dimensional, and long-sequence time-series data that embody ICS business logic. Due to its complex and varying periodic characteristics, as well as the presence of long-distance and misaligned temporal associations among features, current anomaly detection methods in ICS are insufficient for feature extraction. This paper proposes an anomaly detection method named TFANet, based on time–frequency fusion feature attention encoding. Considering that periodic variations are more concentrated in the frequency domain, this method first transforms the time-domain data into the frequency domain, obtaining both amplitude and phase data. Then, these data, together with the original time-series data, are used to extract features from two perspectives: long-term temporal changes and long-distance associations. Finally, the six features learned from both the time and frequency domains are fused, and the feature weights are calculated using an attention mechanism to complete the anomaly classification. In multi-classification tasks on three ICS datasets, the proposed method outperforms three popular time-series models—iTransformer, Crossformer, and TimesNet—across five metrics: accuracy, precision, recall, F1 score, and AUC-ROC, with average improvements of approximately 19%, 37%, 31%, 35%, and 22%, respectively.

1 September 2011 in Applied Mechanics and Materials
Verified
Journal Article
Closed access
PDF file
YSYun Sha
Yun Sha
ZGZhi Jun Gao
Zhi Jun Gao
FXFei Xu
Fei Xu
Thin-walled structures of future hypersonic flight vehicles will encounter complex loadings and exhibit obvious nonlinear responses. The thermal loads from high speed flow or engine jet flow can cause thermal buckling of thin-walled structures, such as Thermal Protection System (TPS). If the structures are loaded with intense acoustic loads simultaneously, large deflection nonlinear response, including snap-through, can be induced. Snap-through will give rise to large amplitude stress cycles and non-zero mean stress, which can lessen the fatigue life markedly. Starting from Hooker’s Law with thermal components, the large deflection governing equations of motion for simply-supported plate under thermo-acoustic loadings are derived. The partial differential equation (PDE) of motion which is difficult to solve is then transformed with Galerkin’s method to the system of ordinary differential equations (ODE) under modal coordinates. The displacement responses under different combinations of temperature increments and sound pressure levels are calculated by employing Runge-Kutta method. Typical thermo-acoustic responses are predicted: 1) random vibration around pre-buckled equilibrium position, 2) persistent snap-through between post-buckled positions, 3) intermittent snap-through, 4) vibration around one of the two post-buckled positions. By dividing the restoring force term in the equation into linear term and nonlinear one, the evolutions of each term are obtained to illustrate the mechanism of thermo-acoustic response and the contributions of each force, including shear force, thermal force and membrane force. Thus a further insight into thermo-acoustic response has been achieved.

5 September 2022 in Applied Sciences
Verified
Journal Article
Open access
PDF file
YSYun Sha
Yun Sha
ZCZhaoyu Chen
Zhaoyu Chen
LXLiu Xue-Jun
Liu Xue-Jun
YYYong Yan
Yong Yan
CDChenchen Du
Chenchen Du
RHRanran Han
Ranran Han
The scarcity of attack samples is the bottleneck problem of anomaly detection of underlying business data in the industrial control system. Predecessors have done a lot of research on temporal data generation, but most of them are not suitable for industrial control attack sample generation. The change patterns of the characteristics of the underlying business data attack samples can be divided into three types: oscillation type, step type, and pulse type. This paper proposes an adaptive industrial control attack sample expansion algorithm based on GAN, which expands the three types of features in different ways. The basic network structure of data expansion adopts GAN. According to the characteristics of oscillation type changes, momentum is selected as the optimizer. Aiming at the characteristics of step type changes, the Adam optimization method is improved. For pulse type features, attack samples are generated according to the location and length of the pulse. Compared with previous time-series data generation methods, this method is more targeted for each feature and has higher similarities.

30 September 2021 in Studies in Informatics and Control
Verified
Journal Article
Closed access
LXLiu Xue-Jun
Liu Xue-Jun
HWHao Wang
Hao Wang
XZXiaoni Zhang
Xiaoni Zhang
HLHaiying Luan
Haiying Luan
YSYun Sha
Yun Sha
YYYong Yan
Yong Yan

1 February 2021 in Journal of Physics: Conference Series
Verified
Conference Paper
Open access
PDF file
WZWen Zhou
Wen Zhou
XCXueying Cao
Xueying Cao
XLXiaoming Li
Xiaoming Li
YSYun Sha
Yun Sha
JCJianping Chen
Jianping Chen
XKXiangmin Kong
Xiangmin Kong
The research of intrusion anomaly detection in industrial control system is facing the problem of few attack samples. This paper proposes an attack data sample generation algorithm based on dual discriminant model. Firstly, a few attack samples are generated by multiple attacks based on the original data set. Secondly, the sparse matrix algorithm is used to expand the data set, and the original data features are distributed to each extended data to ensure the authenticity of the seed data. Finally, a large number of attack samples are generated by the dual discriminant model to complete the sample expansion. The experimental results of Mississippi SCADA natural gas pipeline data set and an industrial control system data set show that, based on keeping the data in a reasonable range, the algorithm can obtain an effectively expanded sample data set. SVM, Random Forest and XGBOOST are used to classify the dataset. The results show that AUC index is better than the negative sample data set generated by traditional GAN algorithm, and is similar to the initial negative sample index.