Weiming Shao
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Weiming Shao has a total of 56 co-authors. Here are the top co-authors by number of shared publications:
- Wenxue Han
China University of Petroleum (East China)
10 shared publications - Zhihuan Song
Zhejiang University
8 shared publications - Zhiqiang Ge
Southeast University
7 shared publications - Chihang Wei
Hangzhou Normal University
5 shared publications - Junghui Chen
Department of Chemical Engineering, Chung-Yuan Christian University, Taoyuan, Taiwan
5 shared publications Dongya Zhao
Department of Chemical Equipment and Control Engineering, College of New Energy China University of Petroleum (East China) Qingdao China
4 shared publications
Latest Publications
- 1 November 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Industrial Informatics
- Vol. 22 (11) , 9789-9800
A Nonlinear Robust Dynamical Generative Latent Variable Model for Industrial Soft Sensing
- 1 November 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Industrial Informatics
- Vol. 22 (11) , 9789-9800
A Nonlinear Robust Dynamical Generative Latent Variable Model for Industrial Soft Sensing
- Open Access
- 19 March 2026
- Published by MDPI AG in Sustainability
- Vol. 18 (6) , 3039
An Efficient Geomechanical Modeling and Intelligent Prediction Approach for Fault Slip in Underground Gas Storage During Long-Term Injection-Production Operation
The steady operation of underground gas storage (UGS) is significant for securing national energy. However, long-term cyclic injection-production operation causes the dynamic changes in formation stress, potentially leading to fault reactivation and slippage. This could affect the seal performance of the fault zone and cause disastrous consequences. In this paper, a mechanical analysis model for fault slip is constructed to study the dynamic seal performance in response to long-term injection-production cycles. An intelligent approach is proposed to predicate the fault slip value based on machine learning algorithms. It can realize long-term prediction of fault slip value under a new condition of injection-production operation. The study shows that (1) formation pressure tends to accumulate near the fault zone due to the low permeability, and the interface of the reservoir layer, cap layer, and fault zone is the seal weak position of UGS; (2) the response of fault slip is driven by the injection-production rate and the reservoir pressure. There is a significant coupling relationship between the fault slip value and the accumulated injection gas volume; (3) the intelligent prediction approach can capture the nonlinear dynamic characteristics of slip tendency accurately, and it exhibits good prediction performance and generalization ability under the new operating condition. This study effectively assesses the dynamic risk for fault slip of depleted hydrocarbon reservoir UGS during the long-term injection-production procedure. It provides an effective technical approach for fault slip tendency analysis and injection-production process optimization, which is important for the sustainable operation of UGS reducing the risk of seal failure and supporting gas storage security.
- 19 March 2026
- Published by MDPI AG in Sustainability
- Vol. 18 (6) , 3039
Open AccessAn Efficient Geomechanical Modeling and Intelligent Prediction Approach for Fault Slip in Underground Gas Storage During Long-Term Injection-Production Operation
The steady operation of underground gas storage (UGS) is significant for securing national energy. However, long-term cyclic injection-production operation causes the dynamic changes in formation stress, potentially leading to fault reactivation and slippage. This could affect the seal performance of the fault zone and cause disastrous consequences. In this paper, a mechanical analysis model for fault slip is constructed to study the dynamic seal performance in response to long-term injection-production cycles. An intelligent approach is proposed to predicate the fault slip value based on machine learning algorithms. It can realize long-term prediction of fault slip value under a new condition of injection-production operation. The study shows that (1) formation pressure tends to accumulate near the fault zone due to the low permeability, and the interface of the reservoir layer, cap layer, and fault zone is the seal weak position of UGS; (2) the response of fault slip is driven by the injection-production rate and the reservoir pressure. There is a significant coupling relationship between the fault slip value and the accumulated injection gas volume; (3) the intelligent prediction approach can capture the nonlinear dynamic characteristics of slip tendency accurately, and it exhibits good prediction performance and generalization ability under the new operating condition. This study effectively assesses the dynamic risk for fault slip of depleted hydrocarbon reservoir UGS during the long-term injection-production procedure. It provides an effective technical approach for fault slip tendency analysis and injection-production process optimization, which is important for the sustainable operation of UGS reducing the risk of seal failure and supporting gas storage security.
- 12 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Industrial Informatics
- Vol. 22 (6) , 5149-5160
A Conditional Gaussian Mixture Model-Guided Transformer for Soft Sensing in Multicondition Industrial Processes
Soft sensing technology plays a crucial role in the real-time monitoring and optimization of key industrial variables. Recently, Transformers have emerged as a promising tool for soft sensor development, owing to natural advantages in dealing with long-range temporal correlations and complex nonlinearities. However, the conventional Transformer-based soft sensors are developed in a global learning framework, suffering from performance degradation in processes with multiple working conditions. To address this limitation, a conditional Gaussian mixture model (CGMM)-guided Transformer is developed in this article. Specifically, a CGMM, differentiating the physical properties and distributional discrepancies of manipulated and process variables, is first designed for high-accuracy recognition of the working conditions. Then, a condition-adaptive Transformer is proposed to accommodate variations in working conditions by capturing localized spatial-temporal characteristics based on the CGMM. Experimental results on both a numerical example and an industrial case demonstrate the superiority of the CGMM-Transformer over baseline models.
- 12 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Industrial Informatics
- Vol. 22 (6) , 5149-5160
A Conditional Gaussian Mixture Model-Guided Transformer for Soft Sensing in Multicondition Industrial Processes
Soft sensing technology plays a crucial role in the real-time monitoring and optimization of key industrial variables. Recently, Transformers have emerged as a promising tool for soft sensor development, owing to natural advantages in dealing with long-range temporal correlations and complex nonlinearities. However, the conventional Transformer-based soft sensors are developed in a global learning framework, suffering from performance degradation in processes with multiple working conditions. To address this limitation, a conditional Gaussian mixture model (CGMM)-guided Transformer is developed in this article. Specifically, a CGMM, differentiating the physical properties and distributional discrepancies of manipulated and process variables, is first designed for high-accuracy recognition of the working conditions. Then, a condition-adaptive Transformer is proposed to accommodate variations in working conditions by capturing localized spatial-temporal characteristics based on the CGMM. Experimental results on both a numerical example and an industrial case demonstrate the superiority of the CGMM-Transformer over baseline models.
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Systems, Man, and Cybernetics: Systems
- Vol. 56 (3) , 2021-2035
A Causal Knowledge-Assisted Semi-Supervised Dynamical Generative Latent Variable Model for Industrial Quality Index Prediction
Generative latent variable models (GLVMs) are prevalent in developing soft sensors for predicting industrial process quality indices owing to their exceptional capability in extracting features, reducing data dimensionality, and explaining data generation mechanisms. Dynamical GLVMs (DGLVMs) are able to effectively deal with the dynamical characteristics of industrial data changes over time, exhibiting prominence in time-series prediction and dynamical process modeling. However, the existing DGLVM-based soft sensors share an intrinsic imperfection that they focus on capturing the temporal correlations between variables while disregarding the causal relationships between variables. Based on industrial data, it is difficult to recover the causalities between variables from the correlations, resulting in compromised generalization performance of the existing DGLVM-based soft sensors. In light of this limitation, a novel causal knowledge-assisted semi-supervised DGLVM (CK-SsDGLVM) is proposed, and an ad hoc parameter learning method based on the expectation–maximization (EM) algorithm is developed to train the CK-SsDGLVM. The performance of the CK-SsDGLVM is comprehensively evaluated using an artificial numerical case and an actual industrial process. The experimental results demonstrate that the CK-SsDGLVM could achieve superior generalization accuracy and interpretability compared to benchmark models.
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Systems, Man, and Cybernetics: Systems
- Vol. 56 (3) , 2021-2035
A Causal Knowledge-Assisted Semi-Supervised Dynamical Generative Latent Variable Model for Industrial Quality Index Prediction
Generative latent variable models (GLVMs) are prevalent in developing soft sensors for predicting industrial process quality indices owing to their exceptional capability in extracting features, reducing data dimensionality, and explaining data generation mechanisms. Dynamical GLVMs (DGLVMs) are able to effectively deal with the dynamical characteristics of industrial data changes over time, exhibiting prominence in time-series prediction and dynamical process modeling. However, the existing DGLVM-based soft sensors share an intrinsic imperfection that they focus on capturing the temporal correlations between variables while disregarding the causal relationships between variables. Based on industrial data, it is difficult to recover the causalities between variables from the correlations, resulting in compromised generalization performance of the existing DGLVM-based soft sensors. In light of this limitation, a novel causal knowledge-assisted semi-supervised DGLVM (CK-SsDGLVM) is proposed, and an ad hoc parameter learning method based on the expectation–maximization (EM) algorithm is developed to train the CK-SsDGLVM. The performance of the CK-SsDGLVM is comprehensively evaluated using an artificial numerical case and an actual industrial process. The experimental results demonstrate that the CK-SsDGLVM could achieve superior generalization accuracy and interpretability compared to benchmark models.
- 7 September 2025
- Published by Elsevier in Advanced Engineering Informatics
- Vol. 69
A physical causality-informed generative latent variable modeling paradigm for industrial virtual metrology
- 7 September 2025
- Published by Elsevier in Advanced Engineering Informatics
- Vol. 69
A physical causality-informed generative latent variable modeling paradigm for industrial virtual metrology
- 29 July 2025
- Published by Elsevier in Measurement
- Vol. 257
A neural networks-aided Bayesian semi-supervised nonlinear hidden Markov model for regressive soft sensing application
- 29 July 2025
- Published by Elsevier in Measurement
- Vol. 257
A neural networks-aided Bayesian semi-supervised nonlinear hidden Markov model for regressive soft sensing application
- 22 May 2025
- Published by Springer Nature
- p. 621-633
A Semi-supervised Robust Linear Dynamical System for Industrial Quality Variable Prediction
- 22 May 2025
- Published by Springer Nature
- p. 621-633
A Semi-supervised Robust Linear Dynamical System for Industrial Quality Variable Prediction
- 23 April 2025
- Published by Wiley in Asia-Pacific Journal of Chemical Engineering
Data‐Driven Fault Detection for Multimode Industrial Processes Based on Mixture of Autoencoders
Data‐driven fault detection technique has been widely applied to industrial processes to reduce production losses caused by faults. Due to complicated chemical reactions and phase changes, the relationships between process variables are nonlinear. In particular, many industrial processes often operate under multiple conditions, resulting in multimode characteristics in the collected data and further complicating the nonlinearities between process variables, rendering challenging difficulties for traditional fault detection methods. In view of this, this paper proposes a novel fault detection method for multimode industrial processes by developing a mixture of autoencoders (MixAE) model. The MixAE is first designed with Gaussian mixture model and multiple single‐hidden layer autoencoders; then, an efficient learning algorithm, based on expectation–maximization algorithm and gradient ascent algorithm, is developed to train the MixAE. By using the confidence index provided by each autoencoder, a comprehensive evaluation statistic is formulated for detecting anomalies. Finally, the proposed fault detection method is validated with a numerical example and an experimental three‐tank liquid level control system. The experimental results demonstrate that the proposed method achieves better fault detection performance for multimode industrial processes compared to traditional methods.
- 23 April 2025
- Published by Wiley in Asia-Pacific Journal of Chemical Engineering
Data‐Driven Fault Detection for Multimode Industrial Processes Based on Mixture of Autoencoders
Data‐driven fault detection technique has been widely applied to industrial processes to reduce production losses caused by faults. Due to complicated chemical reactions and phase changes, the relationships between process variables are nonlinear. In particular, many industrial processes often operate under multiple conditions, resulting in multimode characteristics in the collected data and further complicating the nonlinearities between process variables, rendering challenging difficulties for traditional fault detection methods. In view of this, this paper proposes a novel fault detection method for multimode industrial processes by developing a mixture of autoencoders (MixAE) model. The MixAE is first designed with Gaussian mixture model and multiple single‐hidden layer autoencoders; then, an efficient learning algorithm, based on expectation–maximization algorithm and gradient ascent algorithm, is developed to train the MixAE. By using the confidence index provided by each autoencoder, a comprehensive evaluation statistic is formulated for detecting anomalies. Finally, the proposed fault detection method is validated with a numerical example and an experimental three‐tank liquid level control system. The experimental results demonstrate that the proposed method achieves better fault detection performance for multimode industrial processes compared to traditional methods.
- 2 January 2025
- Published by IOP Publishing in Measurement Science and Technology
- Vol. 36 (2)
Development of Industrial Soft Sensor Based on Semi-Supervised Probabilistic Mixture of Echo State Networks
Echo state network (ESN) is considered an effective tool for soft sensor development owing to its advantages in handling process dynamics and nonlinearities. However, the representation capabilities of a single ESN model are limited for complex processes, leading the ESN-based soft sensors to compromised performance. Furthermore, the scarcity of labeled samples in industrial processes makes it difficult to fully utilize the advantages of the ESN. In view of this, a novel semi-supervised probabilistic mixture of ESN (SsPMESN) is proposed in this paper. In the SsPMESN, multiple localized ESNs are established, which covers the shortage of a single ESN model while avoiding employing the complex deep ESN. In addition, an efficient variational inference-based parameter learning algorithm is developed to train the SsPMESN. The performance of the SsPMESN is verified through both numerical and industrial examples. On the debutanizer column process, the SsPMESN-based soft sensor lifts the value of the generalization R2 index to as high as over 0.9, showing the effectiveness and applicability of the SsPMESN in soft sensor modeling for dynamic industrial processes.
- 2 January 2025
- Published by IOP Publishing in Measurement Science and Technology
- Vol. 36 (2)
Development of Industrial Soft Sensor Based on Semi-Supervised Probabilistic Mixture of Echo State Networks
Echo state network (ESN) is considered an effective tool for soft sensor development owing to its advantages in handling process dynamics and nonlinearities. However, the representation capabilities of a single ESN model are limited for complex processes, leading the ESN-based soft sensors to compromised performance. Furthermore, the scarcity of labeled samples in industrial processes makes it difficult to fully utilize the advantages of the ESN. In view of this, a novel semi-supervised probabilistic mixture of ESN (SsPMESN) is proposed in this paper. In the SsPMESN, multiple localized ESNs are established, which covers the shortage of a single ESN model while avoiding employing the complex deep ESN. In addition, an efficient variational inference-based parameter learning algorithm is developed to train the SsPMESN. The performance of the SsPMESN is verified through both numerical and industrial examples. On the debutanizer column process, the SsPMESN-based soft sensor lifts the value of the generalization R2 index to as high as over 0.9, showing the effectiveness and applicability of the SsPMESN in soft sensor modeling for dynamic industrial processes.