Zujie Yang

Zujie Yang

Auto-generated profile, based on publicly available information

Zujie Yang

Hefei University of Technology

Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

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

Representative Publications

View All Publications
25 April 2022 in Measurement Science and Technology
Verified
Journal Article
Closed access
ZYZujie Yang
Zujie Yang
ZZZongzhen Zhang
Zongzhen Zhang
JWJinrui Wang
Jinrui Wang
BHBaokun Han
Baokun Han
HBHuaiqian Bao
Huaiqian Bao
SLShunming Li
Shunming Li
Random impact interference has always been an important subject in fault diagnosis. In the fast kurtogram (FK), kurtosis is a sparsity index used to locate optimal resonance frequency bands. However, the shortcomings of the kurtosis measure are a tendency to select the random impact with a large amplitude and a relatively weak sparse measure ability, which reduces the accuracy and anti-interference of the FK method. To overcome the effects of random impact interference, this paper proposes a novel bearing fault diagnosis method named the fast nonlinear Hoyergram (FNH), in which the nonlinear Hoyer index is employed to replace kurtosis to improve the fault representation capability under random impulse interference. Firstly, Z -score normalization and a generalized nonlinear sigmoid activation function are used for signal preprocessing, and the scale distribution of the signal is changed to weaken the impact of random impact interference. Secondly, the Hoyer index, which can be viewed as the normalized form of the l 2 / l 1 norm, is used to replace kurtosis to improve the sparsity measure capability. Thirdly, in the Hoyergram, the frequency band with the largest Hoyer value can be chosen as the best resonance frequency band for the squared envelope analysis. Finally, the proposed FNH is compared with the FK method through simulation and experimental signals, and the effectiveness of the proposed method is verified.

16 February 2023 in Measurement Science and Technology
Verified
Journal Article
Closed access
PDF file
ZYZujie Yang
Zujie Yang
ZZZongzhen Zhang
Zongzhen Zhang
JWJinrui Wang
Jinrui Wang
BHBaokun Han
Baokun Han
ZLZongling Liu
Zongling Liu
SXShuo Xing
Shuo Xing
The generalized nonlinear sparse spectrum (GNSS), as an improved fast kurtogram (FK) method, effectively suppresses the interference of abnormal signals through nonlinear preprocessing and sparse enhancement. However, the GNSS method inherits the shortcoming of the traditional FK method, using finite impulse response filters to process nonstationary signals, which limits the accuracy of fault extraction. Therefore, more precise filters should be developed to further improve the performance of fault features. Inspired by this, this paper introduces the dual-tree complex wavelet packet transform (DTCWPT) into the sparse spectrum, and proposes an enhanced generalized nonlinear sparse spectrum (EGNSS) for bearing fault diagnosis. Firstly, nonlinear preprocessing is performed on the input signal to weaken the interference of abnormal impacts. Secondly, the generalized pq-mean value of each subband obtained by DTCWPT is calculated. Finally, the sparse spectrum is constructed and the signal reconstruction is performed on the frequency band where the maximum generalized pq-mean value is located for the envelope analysis. Simulation signals and experimental bearing fault signals have been applied to demonstrate the superiority of the proposed EGNSS in improving fault performance and accuracy.

3 March 2023 in Measurement Science and Technology
Verified
Journal Article
Closed access
BHBaokun Han
Baokun Han
SXShuo Xing
Shuo Xing
JWJinrui Wang
Jinrui Wang
ZZZongzhen Zhang
Zongzhen Zhang
HBHuaiqian Bao
Huaiqian Bao
HMHao Ma
Hao Ma
Currently, most fault diagnosis methods can achieve desired results from a single signal source. However, a single sensor signal has limited features and adaptability to the working environment, which will greatly affect the diagnosis results. To overcome this weakness, a multichannel deep adaptive adversarial network (MCDAAN) based on fusing acoustic and vibration signals is proposed in this paper. The training process of MCDAAN primarily includes the following aspects. First, the acoustic and vibration signals extracted by the neural network feature extraction are fused after being adjusted by the convolutional block attention module in channel and spatial dimensions. Next, the fusion features of the source and target domains are measured by the Wasserstein distance. Finally, the fused features are classified by the label and domain classifiers. The proposed MCDAAN is tested using acoustic and vibration signals collected at ten transfer tasks. The results demonstrate that the diagnostic accuracy of the proposed MCDAAN can reach more than 99% in both groups of experiments. MCDAAN can accurately classify all kinds of fault samples, and the classification accuracy is superior to other comparison methods.

7 December 2022 in Measurement Science and Technology
Verified
Journal Article
Closed access
BHBaokun Han
Baokun Han
ZLZongling Liu
Zongling Liu
ZZZongzhen Zhang
Zongzhen Zhang
JWJinrui Wang
Jinrui Wang
HBHuaiqian Bao
Huaiqian Bao
BLBo Li
Bo Li
Machine learning is gradually applied to the fault diagnosis system of rotating machinery. However, the fault diagnosis system can only classify and identify the fault types previously trained by the model in the system. If the system is required to identify more types of faults, all the untrained new fault types and previously trained fault types need to be input into the model to retrain. Under the current background of big data, the upgrade time of fault types will be relatively long. To solve this problem, a parallel network model based on intrinsic component filtering (PICF) is proposed, in which each type of sample is trained separately, and then each type of training model is reduced in dimension, and finally the model we need is combined. The fault diagnosis framework based on the PICF is proposed. Firstly, the framework divides the input fault samples into training samples and test samples. Then the training samples are randomly segmented and input into the PICF training model, then the activation function is introduced to activate the training features and test features, and finally the softmax classifier is used for classification. The sparsity of order fault training in parallel network is discussed and the influence of sample segment number and nonlinear activation function on diagnosis is studied. Compared with other deep learning methods, the experiment results of the bearing and gearbox show that the proposed method can not only achieve higher fault classification accuracy under small sample training, but also update the model efficiently without reducing the diagnosis accuracy when increasing fault types.

31 July 2023 in Measurement Science and Technology
Verified
Journal Article
Closed access
BHBaokun Han
Baokun Han
SYShunxiang Yao
Shunxiang Yao
ZZZongzhen Zhang
Zongzhen Zhang
JWJinrui Wang
Jinrui Wang
ZYZujie Yang
Zujie Yang
YWYongchang Wei
Yongchang Wei
Variational mode decomposition (VMD) is widely used in the fault diagnosis of rotating equipment. The kurtosis index is typically utilized as the objective function to determine the optimal decomposition mode in conventional VMD. However, kurtosis is easily interfered with by abnormal impact signals, seriously affecting the accuracy of VMD in the fault extraction process. Therefore, the instability of the kurtosis index for anomalous disturbances should be addressed. In addition, the number of modal decompositions K and the penalty factor α are not wisely selected, exerting a significant influence on the decomposition effect of VMD. In this study, we propose an enhanced method for fault diagnosis of rotating machineries under abnormal impact interference, namely the improved nonlinear VMD method based on the l 1/2 norm (INVL). Initially, we employ the whale optimization algorithm to optimize the combination of parameters [ K , α ] with the aim of achieving the most favorable decomposition outcomes. Subsequently, the decomposed modes undergo preprocessing using the nonlinear tanh function to mitigate the impact of amplitude shocks. Furthermore, we utilize the sparse l 1/2 norm to select fault feature components, thereby evaluating the sparse distribution of modes. The modes exhibiting the lowest l 1/2-norm values are chosen for information reconstruction and envelope analysis. To assess the effectiveness of the proposed method, we conduct simulations and experimental verifications. The results demonstrate that the INVL method effectively suppresses abnormal shocks and successfully extracts bearing fault features in the presence of anomalous collision disturbances.

8 September 2026 in Journal of Vibration and Control
Verified
Journal Article
Closed access
ZYZu-Jie Yang
Zu-Jie Yang
YZYong-Bin Zhang
Yong-Bin Zhang
LXLiang Xu
Liang Xu
XZXiao-Zheng Zhang
Xiao-Zheng Zhang
Accurate characterization of material sound absorption is essential for acoustic engineering and material design. However, the traditional two-microphone impedance tube method is fundamentally constrained by the plane-wave assumption, resulting in significant performance degradation beyond the cutoff frequency. To overcome this limitation, this paper proposes a channel-prior attention-guided deep residual network (CPAM-DRN) for predicting the normal sound absorption coefficient (NSAC) directly from multimodal sound pressure fields. Unlike traditional modal decomposition or cancellation techniques, the proposed method explicitly exploits higher-order modal information as informative features for sound absorption characterization, enabling wide-band measurements without reducing tube diameter or increasing the number of microphones. The framework integrates physical acoustic propagation knowledge with data-driven learning to enhance prediction accuracy and robustness. A numerical impedance tube model is established to generate multimodal datasets for training, validation, and testing, enabling the network to learn the nonlinear mapping between pressure fields and NSAC. Simulation results demonstrate excellent predictive performance, achieving an R 2 of 0.9863 and an RMSE of 0.0024 under multimodal conditions, while the traditional method fails beyond the cutoff frequency. Experimental validation further shows that the predicted absorption coefficients in the 2300-6000 Hz range are highly consistent with measurements obtained using commercial B&K equipment, with an average deviation of less than 0.03. These results demonstrate the effectiveness and generalization capability of the proposed method for data-driven wide-band sound absorption characterization.