Xiaodong Zhai
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
Research Subjects
Xiaodong Zhai has a total of 30 co-authors. Here are the top co-authors by number of shared publications:
- Fei Qiao
Tongji University
3 shared publications - Dongyuan Wang
College of Electronics and Information Engineering, Tongji University, Shanghai, China
2 shared publications Yandong Hou
School of Artificial Intelligence, Henan University
2 shared publications- Yandong Hou
School of Artificial Intelligence, Henan University, Zhengzhou, China
2 shared publications - Zhengquan Chen
School of Electrical Engineering, North China University of Water Resources and Electric Power, Zhengzhou 450045, China
2 shared publications - Chen Ding
School of Intelligent Manufacturing and Future Technology, Fujian Fuyao University of Science and Technology, Fuzhou, China
1 shared publication
Latest Publications
- 22 July 2026
- Published by IOP Publishing in Measurement Science and Technology
- Vol. 37 (29) , 296106
A novel remaining useful life prediction framework for incomplete degradation data with dual-layer transfer alignment and dual uncertainty quantification
In industrial applications, bearing remaining useful life (RUL) prediction typically uses data observed only up to the current operating stage. As a result, the target domain often contains incomplete degradation sequences rather than full life-cycle trajectories. This setting creates two major difficulties for transfer-learning models in practice. First, source-domain instances with inconsistent degradation trends can introduce negative transfer. Second, incomplete target-domain sequences can cause degradation-stage mismatch, which biases cross-domain alignment. Moreover, because the latter half of the target-domain degradation process is unobservable, uncertainty quantification based only on model uncertainty cannot fully reflect the true predictive risk. To address these issues, this paper proposes a novel bearing RUL prediction framework for incomplete degradation data with dual-layer transfer alignment and dual uncertainty quantification. Specifically, a dual-layer alignment mechanism is developed at both the instance and stage levels. At the instance level, gradient consistency is used to measure the similarity of degradation dynamics between the source and target domains, and transfer weights are dynamically assigned to suppress negative transfer. At the stage level, a partial optimal transport- correlation alignment alignment strategy is introduced to establish cross-domain stage correspondence and align within-stage statistical distributions, thereby alleviating stage mismatch bias. In addition, a dual uncertainty quantification strategy is constructed by combining Monte Carlo dropout for model uncertainty estimation with instance-level weights to characterize trajectory uncertainty from multiple source degradation patterns. Experiments on the XJTU-SY and PHM 2012 datasets demonstrate that the proposed method achieves superior RUL prediction accuracy and provides more reliable prediction intervals than representative transfer learning and uncertainty quantification methods under incomplete degradation data.
- 22 July 2026
- Published by IOP Publishing in Measurement Science and Technology
- Vol. 37 (29) , 296106
A novel remaining useful life prediction framework for incomplete degradation data with dual-layer transfer alignment and dual uncertainty quantification
In industrial applications, bearing remaining useful life (RUL) prediction typically uses data observed only up to the current operating stage. As a result, the target domain often contains incomplete degradation sequences rather than full life-cycle trajectories. This setting creates two major difficulties for transfer-learning models in practice. First, source-domain instances with inconsistent degradation trends can introduce negative transfer. Second, incomplete target-domain sequences can cause degradation-stage mismatch, which biases cross-domain alignment. Moreover, because the latter half of the target-domain degradation process is unobservable, uncertainty quantification based only on model uncertainty cannot fully reflect the true predictive risk. To address these issues, this paper proposes a novel bearing RUL prediction framework for incomplete degradation data with dual-layer transfer alignment and dual uncertainty quantification. Specifically, a dual-layer alignment mechanism is developed at both the instance and stage levels. At the instance level, gradient consistency is used to measure the similarity of degradation dynamics between the source and target domains, and transfer weights are dynamically assigned to suppress negative transfer. At the stage level, a partial optimal transport- correlation alignment alignment strategy is introduced to establish cross-domain stage correspondence and align within-stage statistical distributions, thereby alleviating stage mismatch bias. In addition, a dual uncertainty quantification strategy is constructed by combining Monte Carlo dropout for model uncertainty estimation with instance-level weights to characterize trajectory uncertainty from multiple source degradation patterns. Experiments on the XJTU-SY and PHM 2012 datasets demonstrate that the proposed method achieves superior RUL prediction accuracy and provides more reliable prediction intervals than representative transfer learning and uncertainty quantification methods under incomplete degradation data.
- 3 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 74 (00189456) , 1-12
SRPKD-IGhostNet: A Highly Precise and Lightweight Fault Diagnosis Framework Based on Sample-Related Progressive Knowledge Distillation and Multiscale Fusion Convolution
Deep learning-based fault diagnosis methods have been successfully deployed in a range of engineering applications. However, with the increasing complexity of network structures, it is difficult to perform deep learning training on systems with limited computational resources. To address this challenge, this study proposes a highly precise and lightweight fault diagnosis framework based on sample-related progressive knowledge distillation and multi-scale fusion convolution, called SRPKD-IGhostNet. The SRPKD is proposed to address the limitation that the teacher model is unable to adequately transfer the knowledge in traditional knowledge distillation. The method enables the student model to acquire more accurate and useful knowledge by transferring the attention-based sample-related knowledge in the middle layer of the teacher model and gradually reducing the erroneous output of the distillation process. Additionally, IGhostNet is designed as the student model in the SRPKD, employing a multi-scale fusion convolution block to replace traditional convolution, thereby reducing computational cost while maintaining high diagnostic accuracy. Extensive experiments are conducted on mechanical fault simulation platforms, such as MCDS, with limited computational resources. The results demonstrate that SRPKD-IGhostNet outperforms existing state-of-the-art lightweight fault diagnosis frameworks in accuracy and computational complexity, highlighting its significant potential for industrial applications.
- 3 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 74 (00189456) , 1-12
SRPKD-IGhostNet: A Highly Precise and Lightweight Fault Diagnosis Framework Based on Sample-Related Progressive Knowledge Distillation and Multiscale Fusion Convolution
Deep learning-based fault diagnosis methods have been successfully deployed in a range of engineering applications. However, with the increasing complexity of network structures, it is difficult to perform deep learning training on systems with limited computational resources. To address this challenge, this study proposes a highly precise and lightweight fault diagnosis framework based on sample-related progressive knowledge distillation and multi-scale fusion convolution, called SRPKD-IGhostNet. The SRPKD is proposed to address the limitation that the teacher model is unable to adequately transfer the knowledge in traditional knowledge distillation. The method enables the student model to acquire more accurate and useful knowledge by transferring the attention-based sample-related knowledge in the middle layer of the teacher model and gradually reducing the erroneous output of the distillation process. Additionally, IGhostNet is designed as the student model in the SRPKD, employing a multi-scale fusion convolution block to replace traditional convolution, thereby reducing computational cost while maintaining high diagnostic accuracy. Extensive experiments are conducted on mechanical fault simulation platforms, such as MCDS, with limited computational resources. The results demonstrate that SRPKD-IGhostNet outperforms existing state-of-the-art lightweight fault diagnosis frameworks in accuracy and computational complexity, highlighting its significant potential for industrial applications.
- 25 October 2025
- Published by Springer Nature in Signal, Image and Video Processing
- Vol. 19 (15) , 1-12
A novel multi-scale adaptive feature fusion framework for accurate and efficient bearing fault diagnosis
- 25 October 2025
- Published by Springer Nature in Signal, Image and Video Processing
- Vol. 19 (15) , 1-12
A novel multi-scale adaptive feature fusion framework for accurate and efficient bearing fault diagnosis
- 19 June 2025
- Published by Elsevier in Measurement
- Vol. 256
A novel fault diagnosis framework based on adaptive VAEGAN and optimal data selection for imbalanced data
- 19 June 2025
- Published by Elsevier in Measurement
- Vol. 256
A novel fault diagnosis framework based on adaptive VAEGAN and optimal data selection for imbalanced data
- 10 September 2024
- Published by Springer Nature in Signal, Image and Video Processing
- Vol. 18 (12) , 9363-9374
Mbda-net: a building damage assessment model based on a multi-scale fusion network
- 10 September 2024
- Published by Springer Nature in Signal, Image and Video Processing
- Vol. 18 (12) , 9363-9374
Mbda-net: a building damage assessment model based on a multi-scale fusion network
- 9 July 2024
- Published by Elsevier in Applied Acoustics
- Vol. 224
A high-performance rolling bearing fault diagnosis method based on adaptive feature mode decomposition and Transformer
- 9 July 2024
- Published by Elsevier in Applied Acoustics
- Vol. 224
A high-performance rolling bearing fault diagnosis method based on adaptive feature mode decomposition and Transformer
- 17 November 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 2832-2836
A Knowledge Graph-Based Muti-Agent Learning Method for Dynamic Scheduling of Flexible Job Shop
Dynamic scheduling is a complicated problem with high research value in the production field, because it requires a fast response to the dynamic disturbance event. In this paper, knowledge graph is combined with reinforcement learning to solve this problem. In the proposed method, knowledge graph model is constructed by a series of steps, including ontology-based knowledge modeling, knowledge extraction, and knowledge fusion. In the knowledge graph model, the state information of the shop floor is extracted from the production data, including the information about machine, job, and process. On this basis, reinforcement learning model is designed, which is composed of machine assignment agent and job sequence agent. Thus, the real-time scheduling decisions can be made rapidly under the dynamic production environment. The effectiveness of the proposed method is verified on a case study of aviation component production workshop. Compared with other methods, the proposed method obtains the best results.
- 17 November 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 2832-2836
A Knowledge Graph-Based Muti-Agent Learning Method for Dynamic Scheduling of Flexible Job Shop
Dynamic scheduling is a complicated problem with high research value in the production field, because it requires a fast response to the dynamic disturbance event. In this paper, knowledge graph is combined with reinforcement learning to solve this problem. In the proposed method, knowledge graph model is constructed by a series of steps, including ontology-based knowledge modeling, knowledge extraction, and knowledge fusion. In the knowledge graph model, the state information of the shop floor is extracted from the production data, including the information about machine, job, and process. On this basis, reinforcement learning model is designed, which is composed of machine assignment agent and job sequence agent. Thus, the real-time scheduling decisions can be made rapidly under the dynamic production environment. The effectiveness of the proposed method is verified on a case study of aviation component production workshop. Compared with other methods, the proposed method obtains the best results.
- 22 August 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 71 (00189456) , 1-11
Model Evolution Mechanism for Incremental Fault Diagnosis
As the key issue of equipment operation and maintenance, fault diagnosis has attracted extensive attention, in which deep learning is often applied to the construction of fault diagnosis models. However, the changing operation state of industrial systems will lead to the decline of model performance. Therefore, it is necessary and urgent to explore the dynamic update of models. This work establishes an incremental fault diagnosis (IFD) framework and further proposes a model evolution mechanism based on adaptive knowledge distillation (KD) and representative exemplar selection (MEMAR). First, when new samples pour in, KD is used to mine the potential correlation between old and new samples, with adaptive constraints on parameter optimization for model updating. Second, genetic algorithm (GA) is applied to selectively retain representative exemplar subsets so as to enhance the model’s ability to deal with dynamic data. Moreover, in order to further verify the effectiveness of MEMAR, the discrimination performance of the fault diagnosis model after multistage incremental learning is tested. The experimental results show that the proposed MEMAR can reduce the storage space of old samples and training time. Besides, it enables the fault diagnosis model based on deep learning to have excellent diagnostic performance for new and old samples.
- 22 August 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 71 (00189456) , 1-11
Model Evolution Mechanism for Incremental Fault Diagnosis
As the key issue of equipment operation and maintenance, fault diagnosis has attracted extensive attention, in which deep learning is often applied to the construction of fault diagnosis models. However, the changing operation state of industrial systems will lead to the decline of model performance. Therefore, it is necessary and urgent to explore the dynamic update of models. This work establishes an incremental fault diagnosis (IFD) framework and further proposes a model evolution mechanism based on adaptive knowledge distillation (KD) and representative exemplar selection (MEMAR). First, when new samples pour in, KD is used to mine the potential correlation between old and new samples, with adaptive constraints on parameter optimization for model updating. Second, genetic algorithm (GA) is applied to selectively retain representative exemplar subsets so as to enhance the model’s ability to deal with dynamic data. Moreover, in order to further verify the effectiveness of MEMAR, the discrimination performance of the fault diagnosis model after multistage incremental learning is tested. The experimental results show that the proposed MEMAR can reduce the storage space of old samples and training time. Besides, it enables the fault diagnosis model based on deep learning to have excellent diagnostic performance for new and old samples.
- 23 May 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 71 (00189456) , 1-12
A Novel Fault Diagnosis Method Under Dynamic Working Conditions Based on a CNN With an Adaptive Learning Rate
The fault diagnosis problem of prognostics and health management (PHM) is becoming more complicated and challenging due to the increasing demand for timeliness and adaptability under dynamic working conditions. This article studies a new adaptive fault diagnosis (AdaFD) scheme and proposes a deep learning-based model and algorithm to enhance both the timeliness of fault diagnosis and its adaptability to dynamic conditions. The primary AdaFD model is based on a convolutional neural network (CNN). The two key modules of AdaFD are: 1) adaptive and independent learning rate design for model training and 2) adaptive structure optimization approach for model application. During the training phase, we design an adaptive learning rate strategy by distinguishing the weight and bias parameters in the CNN model to accelerate the training process. In the application phase, the adaptive batch normalization (AdaBN) algorithm is adopted to improve the adaptability of the primary CNN model to dynamic working conditions and data noise. Finally, a case study is conducted to validate the superiority of AdaFD based on a benchmark rolling bearing dataset. The results show that the proposed fault diagnosis method has higher accuracy and better adaptability than other common methods and possesses advantages in terms of training efficiency.
- 23 May 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 71 (00189456) , 1-12
A Novel Fault Diagnosis Method Under Dynamic Working Conditions Based on a CNN With an Adaptive Learning Rate
The fault diagnosis problem of prognostics and health management (PHM) is becoming more complicated and challenging due to the increasing demand for timeliness and adaptability under dynamic working conditions. This article studies a new adaptive fault diagnosis (AdaFD) scheme and proposes a deep learning-based model and algorithm to enhance both the timeliness of fault diagnosis and its adaptability to dynamic conditions. The primary AdaFD model is based on a convolutional neural network (CNN). The two key modules of AdaFD are: 1) adaptive and independent learning rate design for model training and 2) adaptive structure optimization approach for model application. During the training phase, we design an adaptive learning rate strategy by distinguishing the weight and bias parameters in the CNN model to accelerate the training process. In the application phase, the adaptive batch normalization (AdaBN) algorithm is adopted to improve the adaptability of the primary CNN model to dynamic working conditions and data noise. Finally, a case study is conducted to validate the superiority of AdaFD based on a benchmark rolling bearing dataset. The results show that the proposed fault diagnosis method has higher accuracy and better adaptability than other common methods and possesses advantages in terms of training efficiency.