Yuxuan Zhang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Yuxuan Zhang has a total of 40 co-authors. Here are the top co-authors by number of shared publications:
- Yuanxiang Li
School of Aeronautics and Astronautics, Shanghai Jiao Tong University, Shanghai, China
10 shared publications - Lei Jia
School of Aeronautics and Astronautics, Shanghai Jiao Tong University, Shanghai, China
3 shared publications Huanchun Peng
Academy for Engineering and Technology, Fudan University, 220 Handan Rd, Shanghai, 200433, China
2 shared publications- Huanchun Peng
COMAC Shanghai Aircraft Customer Service Co.,Ltd., Shanghai, China
2 shared publications - Mengni Wang
School of Aeronautics and Astronautics, Shanghai Jiao Tong University, Shanghai, China
2 shared publications Xian Wei
2 shared publications
Latest Publications
- 4 November 2025
- Published by Springer Nature in Aerospace Systems
- p. 1-13
Weakly supervised update of civil aircraft maintenance knowledge graphs through in-context learning
- 4 November 2025
- Published by Springer Nature in Aerospace Systems
- p. 1-13
Weakly supervised update of civil aircraft maintenance knowledge graphs through in-context learning
- 22 October 2025
- Published by Springer Nature in Aerospace Systems
- Vol. 9 (2) , 235-248
Knowledge graph reasoning: learning long-path fault prioritization from aircraft maintenance records
- 22 October 2025
- Published by Springer Nature in Aerospace Systems
- Vol. 9 (2) , 235-248
Knowledge graph reasoning: learning long-path fault prioritization from aircraft maintenance records
- 17 July 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 12 (20) , 41562-41573
RmGPT: A Foundation Model With Generative Pretrained Transformer for Fault Diagnosis and Prognosis in Rotating Machinery
In industry, the reliability of rotating machinery is critical for production efficiency and safety. Current methods of Prognostics and Health Management (PHM) often rely on task-specific models, which face significant challenges in handling diverse datasets with varying signal characteristics, fault modes and operating conditions. Inspired by advancements in generative pretrained models, we propose RmGPT, a unified model for diagnosis and prognosis tasks. RmGPT introduces a novel generative token-based framework, incorporating Signal Tokens, Prompt Tokens, Time-Frequency Task Tokens and Fault Tokens to handle heterogeneous data within a unified model architecture. We leverage self-supervised learning for robust feature extraction and introduce a next signal token prediction pretraining strategy, alongside efficient prompt learning for task-specific adaptation. Extensive experiments demonstrate that RmGPT significantly outperforms state-of-the-art algorithms, achieving near-perfect accuracy in diagnosis tasks and exceptionally low errors in prognosis tasks. Notably, RmGPT excels in few-shot learning scenarios, achieving 82% accuracy in 16-class one-shot experiments, highlighting its adaptability and robustness. This work establishes RmGPT as a powerful PHM foundation model for rotating machinery, advancing the scalability and generalizability of PHM solutions. Code is available at:https://github.com/Pandalin98/RmGPT.
- 17 July 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 12 (20) , 41562-41573
RmGPT: A Foundation Model With Generative Pretrained Transformer for Fault Diagnosis and Prognosis in Rotating Machinery
In industry, the reliability of rotating machinery is critical for production efficiency and safety. Current methods of Prognostics and Health Management (PHM) often rely on task-specific models, which face significant challenges in handling diverse datasets with varying signal characteristics, fault modes and operating conditions. Inspired by advancements in generative pretrained models, we propose RmGPT, a unified model for diagnosis and prognosis tasks. RmGPT introduces a novel generative token-based framework, incorporating Signal Tokens, Prompt Tokens, Time-Frequency Task Tokens and Fault Tokens to handle heterogeneous data within a unified model architecture. We leverage self-supervised learning for robust feature extraction and introduce a next signal token prediction pretraining strategy, alongside efficient prompt learning for task-specific adaptation. Extensive experiments demonstrate that RmGPT significantly outperforms state-of-the-art algorithms, achieving near-perfect accuracy in diagnosis tasks and exceptionally low errors in prognosis tasks. Notably, RmGPT excels in few-shot learning scenarios, achieving 82% accuracy in 16-class one-shot experiments, highlighting its adaptability and robustness. This work establishes RmGPT as a powerful PHM foundation model for rotating machinery, advancing the scalability and generalizability of PHM solutions. Code is available at:https://github.com/Pandalin98/RmGPT.
- 22 August 2024
- Published by Springer Nature in Aerospace Systems
- Vol. 8 (1) , 149-161
Data-driven dynamic health index construction for diagnosis and prognosis of Engine Bleed Air system
- 22 August 2024
- Published by Springer Nature in Aerospace Systems
- Vol. 8 (1) , 149-161
Data-driven dynamic health index construction for diagnosis and prognosis of Engine Bleed Air system
- 22 June 2024
- Published by Elsevier in Expert Systems with Applications
- Vol. 255
Retrieving memory as prompts for continual relation extraction
- 22 June 2024
- Published by Elsevier in Expert Systems with Applications
- Vol. 255
Retrieving memory as prompts for continual relation extraction
- 19 August 2023
- Published by Elsevier in Expert Systems with Applications
- Vol. 235
Multi-information interaction graph neural network for joint entity and relation extraction
- 19 August 2023
- Published by Elsevier in Expert Systems with Applications
- Vol. 235
Multi-information interaction graph neural network for joint entity and relation extraction
- 17 October 2022
- Published by Elsevier in Expert Systems with Applications
- Vol. 213
Missing-edge aware knowledge graph inductive inference through dual graph learning and traversing
- 17 October 2022
- Published by Elsevier in Expert Systems with Applications
- Vol. 213
Missing-edge aware knowledge graph inductive inference through dual graph learning and traversing
- 20 December 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 22 (4) , 3334-3347
Adaptive Spatio-Temporal Graph Information Fusion for Remaining Useful Life Prediction
Accurate remaining useful life (RUL) prediction is of great significance for maintaining the safety and reliability of many industrial systems. In recent years, deep learning based-methods predict the RUL by automatically learning and fusing degradation features from signals and have shown great potential in improving prediction accuracy. However, these methods only focus on capturing degradation information from the sensor signals in the time domain while ignoring the characteristics of different sensors in the spatial domain. Our key motivation is that spatial characteristic is also critical in RUL prediction and cast the RUL prediction problem into a deep spatio-temporal graph fusion problem, which consists of two parts, i.e., spatial structures learning and spatio-temporal information fusion. We propose a framework, namely, adaptive spatio-temporal graph neural network (ASTGNN) to solve the above problem. In the spatial structures learning part, we propose two spatial graph convolution layers, i.e., ASTGNN-M and ASTGNN-A to learn the spatial structures adaptively from the time-varying signals. In the spatio-temporal information fusion part, we use dilation convolution to alleviate the over-smoothing problem when encountering long sensor signals. Moreover, based on ASTGNN, a model named adaptive spatio-temporal hypergraph neural network (ASTHGNN) is proposed for high-order spatio-temporal feature learning. The performances of ASTGNN-M, ASTGNN-A and ASTHGNN are investigated on the C-MAPSS turbofan engine dataset. Experimental results show that ASTHGNN achieves superior performance compared with the state-of-the-art methods and the proposed methods can effectively learn the graph and hypergraph structures of sensor signals.
- 20 December 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 22 (4) , 3334-3347
Adaptive Spatio-Temporal Graph Information Fusion for Remaining Useful Life Prediction
Accurate remaining useful life (RUL) prediction is of great significance for maintaining the safety and reliability of many industrial systems. In recent years, deep learning based-methods predict the RUL by automatically learning and fusing degradation features from signals and have shown great potential in improving prediction accuracy. However, these methods only focus on capturing degradation information from the sensor signals in the time domain while ignoring the characteristics of different sensors in the spatial domain. Our key motivation is that spatial characteristic is also critical in RUL prediction and cast the RUL prediction problem into a deep spatio-temporal graph fusion problem, which consists of two parts, i.e., spatial structures learning and spatio-temporal information fusion. We propose a framework, namely, adaptive spatio-temporal graph neural network (ASTGNN) to solve the above problem. In the spatial structures learning part, we propose two spatial graph convolution layers, i.e., ASTGNN-M and ASTGNN-A to learn the spatial structures adaptively from the time-varying signals. In the spatio-temporal information fusion part, we use dilation convolution to alleviate the over-smoothing problem when encountering long sensor signals. Moreover, based on ASTGNN, a model named adaptive spatio-temporal hypergraph neural network (ASTHGNN) is proposed for high-order spatio-temporal feature learning. The performances of ASTGNN-M, ASTGNN-A and ASTHGNN are investigated on the C-MAPSS turbofan engine dataset. Experimental results show that ASTHGNN achieves superior performance compared with the state-of-the-art methods and the proposed methods can effectively learn the graph and hypergraph structures of sensor signals.
- Open Access
- 5 October 2021
- Published by Springer Nature in Aerospace Systems
- Vol. 4 (4) , 367
Correction to: Spatio-temporal graph convolutional neural network for remaining useful life estimation of aircraft engines
- 5 October 2021
- Published by Springer Nature in Aerospace Systems
- Vol. 4 (4) , 367
Open AccessCorrection to: Spatio-temporal graph convolutional neural network for remaining useful life estimation of aircraft engines
- 30 January 2021
- Published by American Chemical Society (ACS) in Organic Letters
- Vol. 23 (4) , 1434-1439
Nickel-Catalyzed Hydrosilylation of Terminal Alkenes with Primary Silanes via Electrophilic Silicon–Hydrogen Bond Activation
We report a simple and effective nickel-based catalytic system, NiCl2·6H2O/tBuOK, for the electrophilically activated hydrosilylation of terminal alkenes with primary silanes. This protocol provides excellent performance under mild reaction conditions: exclusive anti-Markovnikov selectivity, broad functional group tolerance (36 examples), and good scalability (TON = 5500). However, the secondary and tertiary silanes are not suitable. Mechanistic studies revealed that this homogeneous catalytic hydrosilylation includes an electrophilically activated Si–H bond process without the generation of nickel hydrides.
- 30 January 2021
- Published by American Chemical Society (ACS) in Organic Letters
- Vol. 23 (4) , 1434-1439
Nickel-Catalyzed Hydrosilylation of Terminal Alkenes with Primary Silanes via Electrophilic Silicon–Hydrogen Bond Activation
We report a simple and effective nickel-based catalytic system, NiCl2·6H2O/tBuOK, for the electrophilically activated hydrosilylation of terminal alkenes with primary silanes. This protocol provides excellent performance under mild reaction conditions: exclusive anti-Markovnikov selectivity, broad functional group tolerance (36 examples), and good scalability (TON = 5500). However, the secondary and tertiary silanes are not suitable. Mechanistic studies revealed that this homogeneous catalytic hydrosilylation includes an electrophilically activated Si–H bond process without the generation of nickel hydrides.