Yixiang Lu

Dr. Yixiang Lu


Research Focus & Metrics

Academic Career Length

14Years (2013-Present)

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

Representative Publications

View All Publications
4 November 2022 in IEEE Transactions on Multimedia
Verified
Journal Article
Open access
DZDawei Zhao
Dawei Zhao
QGQingwei Gao
Qingwei Gao
YLYixiang Lu
Yixiang Lu
DSDong Sun
Dong Sun
In the multi-view multi-label (MVML) classification problem, multiple views are simultaneously associated with multiple semantic representations. Multi-view multi-label learning inevitably has the problems of consistency, diversity, and non-alignment among views and the correlation among labels. Most of the existing multi-view multi-label methods for non-aligned views assume that each view has a common or shared label set, but because a single view cannot contain the entire label information, they often learn suboptimal results. Based on this, this paper proposes a non-aligned multi-view multi-label classification method that learns view-specific labels (LVSL), aiming to explicitly mine the information of view-specific labels and low-rank label structures in non-aligned views in a unified model framework. Furthermore, to alleviate insufficient available label information, we thoroughly explored the global and local structural information among labels. Specifically, first, we assume that there is structural consistency between the view and the label space and then construct the view-specific label model in turn. Second, to enrich the original label space information, we mine the consistent information of multiple views and the low-rank correlation information hidden among multiple labels. Finally, the contribution weight of each view is combined with learning the complementary information among the views in the decision-making stage, and extend the model to handle nonlinear data. The results of the proposed method compared with existing state-of-the-art algorithms on several datasets validate its effectiveness.

22 March 2021 in IEEE Transactions on Instrumentation and Measurement
Verified
Journal Article
Closed access
YLYixiang Lu
Yixiang Lu
ZWZhenya Wang
Zhenya Wang
DZDe Zhu
De Zhu
QGQingwei Gao
Qingwei Gao
DSDong Sun
Dong Sun
Bearing is one of the most important transmission components and breaks down more frequently than other parts in rotating equipment. The state detection of the bearing has a great significance for their working performance and safety in industrial production. In this article, we propose a very effective fault diagnosis algorithm based on clustering and sparse representation for rolling bearing. To obtain a robust cluster center for the acquired noisy samples, the samples are first clustered by using their frequency spectrums instead of directly in time-domain waveforms. Then, for each class of samples, an adaptive redundant dictionary is trained over residuals between the original frequency spectrums and their component projected on the calculated cluster center. The noise contained in a test sample is reduced by performing sparse coding and representation with the trained dictionary. Finally, the test sample belonging to a specific category is identified by selecting the maximum cosine similarity value between the reconstructed sample and cluster centers. Experimental results show that the proposed algorithm performs well on both simulated signals and real signals and exhibits advantages over other fault diagnosis methods in terms of diagnosis accuracy.

15 February 2024 in IEEE Transactions on Instrumentation and Measurement
Verified
Journal Article
Closed access
YLYixiang Lu
Yixiang Lu
DTDaiqi Tang
Daiqi Tang
DZDe Zhu
De Zhu
QGQingwei Gao
Qingwei Gao
DZDawei Zhao
Dawei Zhao
JLJunwen Lyu
Junwen Lyu
Remaining Useful Life (RUL) prediction is one of the difficulties in prognostics and system health management (PHM) of bearings. Moreover, due to the limitations of test cost and equipment, it isn’t easy to obtain abundant full life cycle experimental samples of bearings, which cannot provide enough training samples to further improve the accuracy of RUL prediction. To address these problem, this paper proposes a rolling bearing RUL prediction method based on coupled diffusion probabilistic model and time attention mechanism (CDTA). This method first augments the original sequence with the coupled diffusion process, which preserves its randomness and reduces the difficulty of training the subsequent inference network. Then, it introduces a time attention unit (TAU) to enhance the network’s ability to extract and fuse temporal features and dependencies within and between samples, which improves prediction accuracy. This paper conducts experimental verification on PHM2012 datasets and shows that the proposed method achieves significant improvement in prediction accuracy compared with existing RUL prediction methods.

29 September 2022 in IEEE Transactions on Instrumentation and Measurement
Verified
Journal Article
Closed access
YXYi Xia
Yi Xia
CZChengzhi Zhang
Chengzhi Zhang
QYqiang ye
qiang ye
YLYixiang Lu
Yixiang Lu
RYRunyu Yang
Runyu Yang
YWYuhui Wu
Yuhui Wu
In real applications, the data distribution may be inconsistent due to the different working conditions of the equipment, resulting in significant performance degradation of a well-trained faults diagnosis model. To this end, unsupervised domain adaptation (UDA) technique is proposed to bridge the distribution gap between the source and the target domain. Most studies on UDA-based faults diagnosis put much attention on extracting domain-invariant features without considering how to improve the discriminant capability of these features in the target domain. In this study, the collaborative and conditional deep adversarial network (CCDAN) is proposed for cross-domain bearing faults diagnosis. In CCDAN, characteristic fault features at different scales are learned and utilized for domain adaptation. The domain classifier at each scale is fed with discriminative information conveyed in the label predictions, resulting in better alignment of feature distributions per class between different domains. By collaborative and adversarial learning (CAL) across different scales, the learned features are expected to be domain-invariant and also discriminant for faults diagnosis in the target domain. In addition, the multiple kernel variant of maximum mean discrepancies (MK-MMDs) is introduced to further reduce the feature distribution discrepancy. Experiments performed on two different datasets proved the effectiveness and superiority of our model.

4 March 2025 in Expert Systems with Applications
Verified
Journal Article
Closed access
YLYixiang Lu
Yixiang Lu
CCChao Cheng
Chao Cheng
DZDawei Zhao
Dawei Zhao
DZDe Zhu
De Zhu
QGQingwei Gao
Qingwei Gao

20 September 2022 in Digital Signal Processing
Verified
Journal Article
Closed access
YLYixiang Lu
Yixiang Lu
YQYue Qiu
Yue Qiu
QGQingwei Gao
Qingwei Gao
DSDong Sun
Dong Sun