Xiaoyang Zheng

Xiaoyang Zheng

Auto-generated profile, based on publicly available information

Xiaoyang Zheng

College of Computer Science and Technology, Qingdao University, Qingdao, China

Research Focus & Metrics

Academic Career Length

7Years (2020-Present)

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

Representative Publications

View All Publications
20 December 2022 in IEEE Transactions on Cybernetics
Verified
Journal Article
Closed access
ZXZongtang Xu
Zongtang Xu
YMYumei Ma
Yumei Ma
ZPZhenkuan Pan
Zhenkuan Pan
XZXiaoyang Zheng
Xiaoyang Zheng
Bearing fault diagnosis of electrical equipment has been a popular research area in recent years because there are often some faults during continuous operation in production due to the harsh working environment. However, the traditional fault signal processing methods rely on highly expert experience, and some parameters are difficult to be optimized by machine-learning methods. Thus, the satisfactory recognition accuracy of fault diagnosis cannot be achieved in the above methods. In this article, a new model based on the spiking neural network (SNN) is proposed, which is called deep the spiking residual shrinkage network (DSRSN) for bearing fault diagnosis. In the model, attention mechanisms and soft thresholding are introduced to improve the recognition rate under a high-level noise background. The higher recognition accuracy is obtained in the proposed model which is tested on the fault signal dataset under different noise intensities. Meanwhile, the training time is about treble as fast as the training time of the artificial neural network, which is reflecting the high efficiency of SNN.

18 December 2020
Verified
Conference Paper
Closed access
XZXiaoyang Zheng
Xiaoyang Zheng
JWJinliang Wu
Jinliang Wu
ZYZeyu Ye
Zeyu Ye
Existing fault diagnosis methods based on Deep Learning require complicated and cumbersome preprocessing procedures heavily relying on signal processing and manual feature extraction techniques, which fails to maximize the superiority of the adaptive representation capability of Deep Learning. This paper proposes a novel end-to-end fault diagnosis approach combined convolutional neural network (CNN) with Bidirectional Long Short-Term Memory (BiLSTM) and Attention Mechanism. Specifically, one dimensional CNN is utilized to perform local feature extraction and dimension reduction on the input signal data. Then, the BiLSTM network is adopted to extract high-level features from the above results. The miscellaneous features extracted are filtered out by using Attention Mechanism. Finally, the learned features are implemented to train Softmax classifier for identifying fault types. To evaluate the effectiveness of the proposed model, experiments are carried out on two different datasets by using ten-fold cross validation approach. The results demonstrate that the proposed model is capable to learn features adaptively with fewer steps while ensuring higher accuracy compared with other state-of-art fault diagnosis methods.

23 September 2022
Verified
Conference Paper
Open access
XZXiaoyang Zheng
Xiaoyang Zheng
YMYumei Ma
Yumei Ma
ZPZhenkuan Pan
Zhenkuan Pan
ZXZongtang Xu
Zongtang Xu
Cardiovascular disease is a serious threat to human health. Electrocardiogram (ECG) is of great significance in clinical diagnosis and follow-up treatment of arrhythmias. To further improve ECG signal classification performance, an arrhythmia detection algorithm based on Convolutional Neural Network (CNN) and channel attention mechanism was proposed. Firstly, one-dimensional ECG signals were converted to two-dimensional grayscale images as input data, which no longer required noise filtering and retained the original data information to the greatest extent. In addition, minority classes data could be enlarged by a special data enhancement method to solve the problem of data imbalance. The Squeeze-and-Excitation (SE) module was inserted to the model to focus on more important information. By training on eight heart rhythm classes from the MIT-BIH arrhythmia database, an overall accuracy of 99.53% was achieved on the SE-CNN algorithm. The experimental results show that the classification accuracy of arrhythmia is improved, while the detection efficiency ensured, which has practical application value and can help doctors to diagnose diseases quickly and accurately in clinical work.