Xueyi Zhang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Xueyi Zhang has a total of 15 co-authors. Here are the top co-authors by number of shared publications:
- Kaixiang Peng
Key Laboratory of Knowledge Automation for Industrial Processes of Ministry of Education, School of Automation and Electrical Engineering University of Science and Technology Beijing Beijing China
8 shared publications - Chuanfang Zhang
University of Science and Technology Beijing
5 shared publications - Jie Dong
Key Laboratory of Knowledge Automation for Industrial Processes of Ministry of Education, School of Automation and Electrical Engineering University of Science and Technology Beijing Beijing China
4 shared publications - Liang Ma
Key Laboratory of Knowledge Automation for Industrial Processes of Ministry of Education, School of Automation and Electrical Engineering, University of Science and Technology Beijing, Beijing, China
3 shared publications - Xin Qin
Key Laboratory of Knowledge Automation for Industrial Processes of Ministry of Education, School of Automation and Electrical Engineering, University of Science and Technology Beijing, Beijing, China
3 shared publications - Chi Zhang
Key Laboratory of Knowledge Automation for Industrial Processes of Ministry of Education, School of Automation and Electrical Engineering, University of Science and Technology Beijing, Beijing, China
1 shared publication
Latest Publications
- 23 March 2026
- Published by Wiley in The Canadian Journal of Chemical Engineering
- Vol. 104 (9) , 4923-4945
Wavelet‐guided dual‐graph transformer for low‐frequency oscillation detection and root cause localization in industrial processes
Industrial process data often contain intermittent, low frequency oscillations that are weak, slowly drifting in frequency, and propagate across multiple coupled control loops, conditions under which fixed thresholds, pairwise test and uncalibrated scores perform poorly. To address these challenges, a wavelet‐guided dual‐graph spatiotemporal transformer framework is proposed for plant‐wide oscillation detection and root‐cause localization with calibrated confidence. First, a complex Morlet continuous wavelet transform is applied to identify oscillatory events and to define two analysis windows per event: an early window capturing onset dynamics and a peak window centred on maximal oscillatory energy. Second, for each window, Granger‐causality and band‐limited coherence graphs are constructed, yielding four directed graphs that encode causal and coherence structure over the specific oscillation interval. Third, a dual‐graph message‐passing model fuses these graphs at each time step, and a temporal transformer aggregates contextual information to produce per‐sensor logits. Finally, a sigmoid output head provides root‐cause probabilities for each channel, and post‐hoc calibration converts raw scores into reliable probabilities suitable for operational thresholding. On real hot strip mill data, the proposed framework outperforms strongest granger baselines for root cause Localisation, achieving top‐1 accuracy of 0.90 and a top‐3 accuracy of 0.95, while the continuous wavelet transform (CWT)‐ridge detector improves oscillations event detection. Reliability analysis and a row‐normalized calibration matrix yields ECE = 0.142 and Brier = 0.049, isotonic calibration slightly pulls mid‐range probabilities towards the centre and makes high confidence predictions a bit stronger without changing the order of predictions. Early/peak Granger‐causality and coherence heatmaps provide transparent, auditable evidence, yielding a practical end‐to‐end deployable solution for plant‐wide low‐frequency oscillation detection and root‐cause localization.
- 23 March 2026
- Published by Wiley in The Canadian Journal of Chemical Engineering
- Vol. 104 (9) , 4923-4945
Wavelet‐guided dual‐graph transformer for low‐frequency oscillation detection and root cause localization in industrial processes
Industrial process data often contain intermittent, low frequency oscillations that are weak, slowly drifting in frequency, and propagate across multiple coupled control loops, conditions under which fixed thresholds, pairwise test and uncalibrated scores perform poorly. To address these challenges, a wavelet‐guided dual‐graph spatiotemporal transformer framework is proposed for plant‐wide oscillation detection and root‐cause localization with calibrated confidence. First, a complex Morlet continuous wavelet transform is applied to identify oscillatory events and to define two analysis windows per event: an early window capturing onset dynamics and a peak window centred on maximal oscillatory energy. Second, for each window, Granger‐causality and band‐limited coherence graphs are constructed, yielding four directed graphs that encode causal and coherence structure over the specific oscillation interval. Third, a dual‐graph message‐passing model fuses these graphs at each time step, and a temporal transformer aggregates contextual information to produce per‐sensor logits. Finally, a sigmoid output head provides root‐cause probabilities for each channel, and post‐hoc calibration converts raw scores into reliable probabilities suitable for operational thresholding. On real hot strip mill data, the proposed framework outperforms strongest granger baselines for root cause Localisation, achieving top‐1 accuracy of 0.90 and a top‐3 accuracy of 0.95, while the continuous wavelet transform (CWT)‐ridge detector improves oscillations event detection. Reliability analysis and a row‐normalized calibration matrix yields ECE = 0.142 and Brier = 0.049, isotonic calibration slightly pulls mid‐range probabilities towards the centre and makes high confidence predictions a bit stronger without changing the order of predictions. Early/peak Granger‐causality and coherence heatmaps provide transparent, auditable evidence, yielding a practical end‐to‐end deployable solution for plant‐wide low‐frequency oscillation detection and root‐cause localization.
- 12 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 13 (7) , 12702-12715
Trinity of Safety–Quality–Efficiency: Cloud–Edge–Device Collaborative Monitoring for Manufacturing Systems With Industrial Validation
Against the backdrop of Industry 4.0, complex industrial processes face heightened risks of performance degradation and safety hazards due to increasing scale and integration. To address this, we propose a novel cloud–edge–device collaborative framework for hierarchical performance monitoring and fault diagnosis. Our approach explicitly targets multiperformance coupling and cross-domain coordination in complex systems. Aligned with the architecture of manufacturing systems, a cloud–edge–device collaborative framework is designed, deeply integrating models for safety control, quality monitoring, and energy consumption prediction. The architecture strategically distributes tasks: 1) at the device layer, real-time control-loop safety is considered and the safety-related data processing and alignment are realized; 2) the edge layer deploys a dual-attention minimal gated unit network coupled with a quality-driven autoencoder (AE), enabling temporal and nonlinear feature extraction for quality-centric fault diagnosis within subprocesses; and 3) the cloud layer integrates self-attention minimal gated units with a broad learning system for multiperspective energy efficiency prediction and holistic evaluation. Finally, a hot strip mill process (HSMP) prototype system is designed and developed to validate the effectiveness and engineering value of the proposed framework through practical case studies.
- 12 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 13 (7) , 12702-12715
Trinity of Safety–Quality–Efficiency: Cloud–Edge–Device Collaborative Monitoring for Manufacturing Systems With Industrial Validation
Against the backdrop of Industry 4.0, complex industrial processes face heightened risks of performance degradation and safety hazards due to increasing scale and integration. To address this, we propose a novel cloud–edge–device collaborative framework for hierarchical performance monitoring and fault diagnosis. Our approach explicitly targets multiperformance coupling and cross-domain coordination in complex systems. Aligned with the architecture of manufacturing systems, a cloud–edge–device collaborative framework is designed, deeply integrating models for safety control, quality monitoring, and energy consumption prediction. The architecture strategically distributes tasks: 1) at the device layer, real-time control-loop safety is considered and the safety-related data processing and alignment are realized; 2) the edge layer deploys a dual-attention minimal gated unit network coupled with a quality-driven autoencoder (AE), enabling temporal and nonlinear feature extraction for quality-centric fault diagnosis within subprocesses; and 3) the cloud layer integrates self-attention minimal gated units with a broad learning system for multiperspective energy efficiency prediction and holistic evaluation. Finally, a hot strip mill process (HSMP) prototype system is designed and developed to validate the effectiveness and engineering value of the proposed framework through practical case studies.
- 9 January 2026
- Published by Elsevier in Neurocomputing
- Vol. 671
An exergy-related fault diagnosis framework based on fault-agnostic meta-learning for industrial multimode batch processes
- 9 January 2026
- Published by Elsevier in Neurocomputing
- Vol. 671
An exergy-related fault diagnosis framework based on fault-agnostic meta-learning for industrial multimode batch processes
- 30 October 2025
- Published by IOP Publishing in Measurement Science and Technology
- Vol. 36 (10) , 106218
A novel quality-related generative data-driven fault diagnosis method for complex industrial processes with incomplete data
Spatiotemporal incomplete data pose significant challenges to the robust diagnosis of quality-related faults in complex industrial environments. Missing data, often attributed to sensor failures, communication issues, or storage limitations, can compromise the reliability of diagnostic systems and diminish the effectiveness of conventional fault detection methods. To address these challenges, this paper proposes a novel integrated framework combining a generative quality-aware (QA) generative adversarial imputation network (GAIN)-based imputation with a graph-based spatiotemporal transformer encoder, a broad learning system (BLS) classifier and an enhanced uncertainty estimation mechanism. Compared to the standard GAIN, our proposed QA-GAIN incorporates a specific quality indicator into adversarial training, reducing the root mean square error (RMSE) by 23%–53% across missing rates of 10%–80% compared to traditional imputation methods. For downstream monitoring, the spatiotemporal transformer with graph convolution and uncertainty weighting reduces Hotelling’s T 2 false alarm rate (FAR) by up to 27% and improves the fault detection rate (FDR) by up to 15% compared to transformer-based baseline models under severe missingness. The BLS-inspired classifier uses latent features efficiently to enhance decision boundaries, while uncertainty calibration shows up to 64% lower expected calibration error (ECE) than standard Monte Carlo (MC) dropout, providing more reliable probability estimates and higher confidence in correct predictions. Evaluations on the steel industry hot strip mill process (HSMP) and the chemical industry Tennessee Eastman process (TEP) datasets confirm the proposed framework’s ability to recover missing data, understand spatiotemporal patterns and deliver strong and trustworthy predictions for quality-related fault diagnosis.
- 30 October 2025
- Published by IOP Publishing in Measurement Science and Technology
- Vol. 36 (10) , 106218
A novel quality-related generative data-driven fault diagnosis method for complex industrial processes with incomplete data
Spatiotemporal incomplete data pose significant challenges to the robust diagnosis of quality-related faults in complex industrial environments. Missing data, often attributed to sensor failures, communication issues, or storage limitations, can compromise the reliability of diagnostic systems and diminish the effectiveness of conventional fault detection methods. To address these challenges, this paper proposes a novel integrated framework combining a generative quality-aware (QA) generative adversarial imputation network (GAIN)-based imputation with a graph-based spatiotemporal transformer encoder, a broad learning system (BLS) classifier and an enhanced uncertainty estimation mechanism. Compared to the standard GAIN, our proposed QA-GAIN incorporates a specific quality indicator into adversarial training, reducing the root mean square error (RMSE) by 23%–53% across missing rates of 10%–80% compared to traditional imputation methods. For downstream monitoring, the spatiotemporal transformer with graph convolution and uncertainty weighting reduces Hotelling’s T 2 false alarm rate (FAR) by up to 27% and improves the fault detection rate (FDR) by up to 15% compared to transformer-based baseline models under severe missingness. The BLS-inspired classifier uses latent features efficiently to enhance decision boundaries, while uncertainty calibration shows up to 64% lower expected calibration error (ECE) than standard Monte Carlo (MC) dropout, providing more reliable probability estimates and higher confidence in correct predictions. Evaluations on the steel industry hot strip mill process (HSMP) and the chemical industry Tennessee Eastman process (TEP) datasets confirm the proposed framework’s ability to recover missing data, understand spatiotemporal patterns and deliver strong and trustworthy predictions for quality-related fault diagnosis.
- 9 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. PP (00189456) , 1
Hierarchical Spatiotemporal Characteristics Analysis-Based Quality-Related Fault Detection and Trustworthy Diagnosis for Complex Industrial Processes
With the improvement of production efficiency and the demand for product diversification, industrial processes present the characteristics of large-scale, long processes, and multi-process coupling. The analysis of hierarchical spatiotemporal characteristics, the imbalanced sample condition, and the uncertainty caused by out-of-distribution (OOD) scenarios including the changes of operating modes or new fault conditions bring severe challenges to traditional fault detection and diagnosis methods. In this paper, an integrated quality-related fault detection and trustworthy diagnosis scheme based on hierarchical spatiotemporal characteristic analysis is proposed. First, spatiotemporal graphs are formulated with reference to the spatial topology of industrial processes and the temporal correlation of data. A cascade spatiotemporal graph neural network (CSTGN) is proposed to extract intra-subprocess and inter-subprocess hierarchical spatiotemporal characteristics, and a bottleneck layer is designed to extract quality-related latent variables from the hierarchical spatiotemporal characteristics to achieve fault detection. Second, considering the imbalanced sample condition in real industrial processes, the CSTGN is transferred to the few-shot fault diagnosis task through model-based transfer learning. Finally, OOD scenarios may emerge subsequent to the deployment of the trained mode. Therefore, Bayesian deep learning is introduced into the CSTGN to perceive the uncertainty and achieve trustworthy fault diagnosis. The effectiveness of the proposed method is illustrated with a real hot strip mill process.
- 9 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. PP (00189456) , 1
Hierarchical Spatiotemporal Characteristics Analysis-Based Quality-Related Fault Detection and Trustworthy Diagnosis for Complex Industrial Processes
With the improvement of production efficiency and the demand for product diversification, industrial processes present the characteristics of large-scale, long processes, and multi-process coupling. The analysis of hierarchical spatiotemporal characteristics, the imbalanced sample condition, and the uncertainty caused by out-of-distribution (OOD) scenarios including the changes of operating modes or new fault conditions bring severe challenges to traditional fault detection and diagnosis methods. In this paper, an integrated quality-related fault detection and trustworthy diagnosis scheme based on hierarchical spatiotemporal characteristic analysis is proposed. First, spatiotemporal graphs are formulated with reference to the spatial topology of industrial processes and the temporal correlation of data. A cascade spatiotemporal graph neural network (CSTGN) is proposed to extract intra-subprocess and inter-subprocess hierarchical spatiotemporal characteristics, and a bottleneck layer is designed to extract quality-related latent variables from the hierarchical spatiotemporal characteristics to achieve fault detection. Second, considering the imbalanced sample condition in real industrial processes, the CSTGN is transferred to the few-shot fault diagnosis task through model-based transfer learning. Finally, OOD scenarios may emerge subsequent to the deployment of the trained mode. Therefore, Bayesian deep learning is introduced into the CSTGN to perceive the uncertainty and achieve trustworthy fault diagnosis. The effectiveness of the proposed method is illustrated with a real hot strip mill process.
- 23 October 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Industrial Informatics
- Vol. 20 (3) , 4397-4407
A Novel Quality-Related Distributed Fault Diagnosis Framework for Large-Scale Sequential Manufacturing Processes
Large-scale manufacturing processes are usually made up of multiple interrelated and distributed continuously subprocesses, which are transmitted and connected by information and quality flow. The characteristics of long processes, quality heritability between subprocesses, and dynamic-coupled variables bring severe challenges to conventional quality-related fault diagnosis. Against this background, a novel distributed diagnosis framework for quality-related faults is proposed in this article. First, the sequential manufacturing process is decomposed into multiple subprocesses based on mechanism knowledge. Second, a novel dual-attention quality-driven autoencoder method is designed as the model for local fault diagnosis. Deep nonlinear features are extracted under quality supervision; meanwhile, the dynamic information and the different correlations among variables are also considered. Then, based on the tandem structure of the manufacturing process, multiple dual-attention quality-driven autoencoder models corresponding to each subprocess are constructed and stacked into a distributed model. Bayesian inference is used to build global monitoring statistics. Moreover, after faults occur, the intervariable attention weights are achieved to identify faulty variables. Finally, the effectiveness and advantages of the proposed framework are demonstrated via a practical large-scale sequential manufacturing process, the hot strip mill process.
- 23 October 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Industrial Informatics
- Vol. 20 (3) , 4397-4407
A Novel Quality-Related Distributed Fault Diagnosis Framework for Large-Scale Sequential Manufacturing Processes
Large-scale manufacturing processes are usually made up of multiple interrelated and distributed continuously subprocesses, which are transmitted and connected by information and quality flow. The characteristics of long processes, quality heritability between subprocesses, and dynamic-coupled variables bring severe challenges to conventional quality-related fault diagnosis. Against this background, a novel distributed diagnosis framework for quality-related faults is proposed in this article. First, the sequential manufacturing process is decomposed into multiple subprocesses based on mechanism knowledge. Second, a novel dual-attention quality-driven autoencoder method is designed as the model for local fault diagnosis. Deep nonlinear features are extracted under quality supervision; meanwhile, the dynamic information and the different correlations among variables are also considered. Then, based on the tandem structure of the manufacturing process, multiple dual-attention quality-driven autoencoder models corresponding to each subprocess are constructed and stacked into a distributed model. Bayesian inference is used to build global monitoring statistics. Moreover, after faults occur, the intervariable attention weights are achieved to identify faulty variables. Finally, the effectiveness and advantages of the proposed framework are demonstrated via a practical large-scale sequential manufacturing process, the hot strip mill process.
- 27 March 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 72 (00189456) , 1-9
A Robust Fault Classification Method for Streaming Industrial Data Based on Wasserstein Generative Adversarial Network and Semi-Supervised Ladder Network
With the development of modern information technology, the collection, storage and transmission of information in the process industry has been gaining popularity. However, the massive streaming industrial data obtained in real time has some non-ideal characteristics, such as lack of labels and missing values, which greatly increases the difficulty of process monitoring in process industry. Therefore, a robust semi-supervised fault classification method is proposed in this paper. First, Wasserstein generative adversarial network (WGAN) and enhanced minimal gated unit (EMGU) are integrated to complete the missing data imputation of the incomplete unlabeled streaming industrial data, and then semi-supervised ladder network (SLN) is trained with the imputed unlabeled data and complete labeled data for fault classification. A case study on the hot rolling process demonstrates that the proposed method shows outstanding modeling and classification performance in lack of labeled data and missing data, compared to the other state-of-art deep learning methods.
- 27 March 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 72 (00189456) , 1-9
A Robust Fault Classification Method for Streaming Industrial Data Based on Wasserstein Generative Adversarial Network and Semi-Supervised Ladder Network
With the development of modern information technology, the collection, storage and transmission of information in the process industry has been gaining popularity. However, the massive streaming industrial data obtained in real time has some non-ideal characteristics, such as lack of labels and missing values, which greatly increases the difficulty of process monitoring in process industry. Therefore, a robust semi-supervised fault classification method is proposed in this paper. First, Wasserstein generative adversarial network (WGAN) and enhanced minimal gated unit (EMGU) are integrated to complete the missing data imputation of the incomplete unlabeled streaming industrial data, and then semi-supervised ladder network (SLN) is trained with the imputed unlabeled data and complete labeled data for fault classification. A case study on the hot rolling process demonstrates that the proposed method shows outstanding modeling and classification performance in lack of labeled data and missing data, compared to the other state-of-art deep learning methods.
- 12 July 2022
- Published by Wiley in The Canadian Journal of Chemical Engineering
- Vol. 101 (3) , 1485-1498
A novel quality‐related process monitoring method for multi‐unit industrial processes under incomplete data conditions
Modern industrial processes usually consist of multiple local units. With the wide application of distributed control systems, a large amount of industrial data is collected. During data acquisition and transmission, outliers and missing data commonly exist in different local units. Under such incomplete data conditions, the detection performance of traditional quality‐related process monitoring methods may be seriously degraded. This study develops a novel quality‐related process monitoring method for multi‐unit industrial processes. Considering the existence of outliers and missing data, incomplete data preprocessing is first performed on raw process data and quality data. Then, process variables are divided into several blocks according to the process knowledge. Afterwards, the quality‐related features extracted by variational information bottleneck (VIB) are used to train deep support vector data description (DSVDD). Finally, a quality‐related process monitoring strategy is designed at both block and global levels. Experimental results on the revised Tennessee Eastman process demonstrate that the proposed method has better monitoring performance under incomplete data conditions than other state‐of‐the‐art methods.
- 12 July 2022
- Published by Wiley in The Canadian Journal of Chemical Engineering
- Vol. 101 (3) , 1485-1498
A novel quality‐related process monitoring method for multi‐unit industrial processes under incomplete data conditions
Modern industrial processes usually consist of multiple local units. With the wide application of distributed control systems, a large amount of industrial data is collected. During data acquisition and transmission, outliers and missing data commonly exist in different local units. Under such incomplete data conditions, the detection performance of traditional quality‐related process monitoring methods may be seriously degraded. This study develops a novel quality‐related process monitoring method for multi‐unit industrial processes. Considering the existence of outliers and missing data, incomplete data preprocessing is first performed on raw process data and quality data. Then, process variables are divided into several blocks according to the process knowledge. Afterwards, the quality‐related features extracted by variational information bottleneck (VIB) are used to train deep support vector data description (DSVDD). Finally, a quality‐related process monitoring strategy is designed at both block and global levels. Experimental results on the revised Tennessee Eastman process demonstrate that the proposed method has better monitoring performance under incomplete data conditions than other state‐of‐the‐art methods.