Jia Wu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Jia Wu has a total of 16 co-authors. Here are the top co-authors by number of shared publications:
- Wenyan Tang
School of Instrumentation Science and Engineering, Harbin Institute of Technology, Harbin 150001, China
4 shared publications Kun Li
School of Automation and Electronic Information, Xiangtan University, Hunan, Xiangtan, China
2 shared publicationsWenyan Tang
School of Automation and Electronic Information Xiangtan University Xiangtan China
2 shared publications- Yongfang Xie
Central South University
2 shared publications Haihong Mo
School of Automation and Electronic Information Xiangtan University Xiangtan China
1 shared publicationHongwei Zhang
1 shared publication
Latest Publications
- Open Access
- 15 June 2026
- Published by Elsevier in SSRN Electronic Journal
Lightweight Fault Diagnosis of Industrial Circulating Water Pumps Using Multi-Level Information Fusion
Deep learning has greatly advanced intelligent fault diagnosis (FD), yet deploying it in industrial settings remains difficult because complementary multi-sensor information must be effectively utilized under strict computational limits. This paper proposes a Lightweight Multi-Level Information Fusion Approach for Fault Diagnosis (LMIF-FD) of circulating water pumps in industrial circulating water systems. The method combines data-level fusion, MobileNetV3-small (M3S), and a multi-scale 1D-ResNet (ML1R) to capture rich temporal and frequency features. Tri-axial vibration signals are first normalized and fused in the time domain, then transformed into RGB images using continuous wavelet transform (CWT) for frequency enhancement. These signals are fed into ML1R and M3S, and their outputs are combined via feature-level fusion. LMIF-FD effectively extracts discriminative fault patterns while substantially reducing model complexity and computational load. Experimental results on real and public datasets validate its diagnostic accuracy, real-time capability, and robustness under challenging industrial conditions.
- 15 June 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessLightweight Fault Diagnosis of Industrial Circulating Water Pumps Using Multi-Level Information Fusion
Deep learning has greatly advanced intelligent fault diagnosis (FD), yet deploying it in industrial settings remains difficult because complementary multi-sensor information must be effectively utilized under strict computational limits. This paper proposes a Lightweight Multi-Level Information Fusion Approach for Fault Diagnosis (LMIF-FD) of circulating water pumps in industrial circulating water systems. The method combines data-level fusion, MobileNetV3-small (M3S), and a multi-scale 1D-ResNet (ML1R) to capture rich temporal and frequency features. Tri-axial vibration signals are first normalized and fused in the time domain, then transformed into RGB images using continuous wavelet transform (CWT) for frequency enhancement. These signals are fed into ML1R and M3S, and their outputs are combined via feature-level fusion. LMIF-FD effectively extracts discriminative fault patterns while substantially reducing model complexity and computational load. Experimental results on real and public datasets validate its diagnostic accuracy, real-time capability, and robustness under challenging industrial conditions.
- 19 April 2025
- Published by Elsevier in ISA Transactions
- Vol. 162, 1-10
Free-will arbitrary time optimization for multi-agent systems with time-varying function
- 19 April 2025
- Published by Elsevier in ISA Transactions
- Vol. 162, 1-10
Free-will arbitrary time optimization for multi-agent systems with time-varying function
- Open Access
- 24 November 2023
- Published by Association for Computing Machinery (ACM)
- p. 799-804
Flotation Industrial Process control based on Model-assisted deep reinforcement learning method
The flotation industrial process is the most widely used beatification method to realize the separation of mineral particles. In this article, we focus on researching the control of the flotation industrial process based on the model-assisted deep reinforcement learning method. Considering the flotation industrial process as a Markov decision process, our control method is designed based on twin delayed deep deterministic gradient algorithm, in which an auxiliary model is constructed for training to avoid the demands of a large number of training data. The experimental results show that the performance of the flotation process can be effectively guaranteed through the proposed method.
- 24 November 2023
- Published by Association for Computing Machinery (ACM)
- p. 799-804
Open AccessFlotation Industrial Process control based on Model-assisted deep reinforcement learning method
The flotation industrial process is the most widely used beatification method to realize the separation of mineral particles. In this article, we focus on researching the control of the flotation industrial process based on the model-assisted deep reinforcement learning method. Considering the flotation industrial process as a Markov decision process, our control method is designed based on twin delayed deep deterministic gradient algorithm, in which an auxiliary model is constructed for training to avoid the demands of a large number of training data. The experimental results show that the performance of the flotation process can be effectively guaranteed through the proposed method.
- 23 November 2022
- Published by Wiley in Asian Journal of Control
- Vol. 25 (4) , 2838-2852
Neural network‐based event‐triggered cluster quasi‐consensus for unknown multiagent systems with directed topology
This paper studies cluster quasi‐consensus problem for a class of unknown nonlinear multiagent systems (MASs) with directed communication topology. First, a distributed continuous neural network (NN)‐based adaptive protocol is presented for solving this problem by introducing reference model to each agent. Then, taking limited communication resource and energy consumption into account, a distributed event‐triggered cluster quasi‐consensus protocol is proposed. Different from the existing results, two event‐triggered mechanisms are constructed in the proposed event‐triggered protocol to reduce communication load and control update frequency as possible. The sufficient conditions that guarantee cluster quasi‐consensus under the both proposed protocols are obtained, respectively. Zeno behavior is proved to be excluded. Finally, simulation results verify the effectiveness of the proposed protocols.
- 23 November 2022
- Published by Wiley in Asian Journal of Control
- Vol. 25 (4) , 2838-2852
Neural network‐based event‐triggered cluster quasi‐consensus for unknown multiagent systems with directed topology
This paper studies cluster quasi‐consensus problem for a class of unknown nonlinear multiagent systems (MASs) with directed communication topology. First, a distributed continuous neural network (NN)‐based adaptive protocol is presented for solving this problem by introducing reference model to each agent. Then, taking limited communication resource and energy consumption into account, a distributed event‐triggered cluster quasi‐consensus protocol is proposed. Different from the existing results, two event‐triggered mechanisms are constructed in the proposed event‐triggered protocol to reduce communication load and control update frequency as possible. The sufficient conditions that guarantee cluster quasi‐consensus under the both proposed protocols are obtained, respectively. Zeno behavior is proved to be excluded. Finally, simulation results verify the effectiveness of the proposed protocols.
- 5 August 2021
- Published by SAGE Publications in Journal of Vibration and Control
- Vol. 28 (23-24) , 3559-3574
Data-driven tracking consensus for a class of unknown nonlinear multi-agent systems
This study investigates the tracking consensus problem for a class of unknown nonlinear multi-agent systems A novel data-driven protocol for this problem is proposed by using the model-free adaptive control method To obtain faster convergence speed, one-step-ahead desired signal is introduced to construct the novel protocol Here, switching communication topology is considered, which is not required to be strongly connected all the time Through rigorous analysis, sufficient conditions are given to guarantee that the tracking errors of all agents are convergent under the novel protocol Examples are given to validate the effectiveness of results derived in this article
- 5 August 2021
- Published by SAGE Publications in Journal of Vibration and Control
- Vol. 28 (23-24) , 3559-3574
Data-driven tracking consensus for a class of unknown nonlinear multi-agent systems
This study investigates the tracking consensus problem for a class of unknown nonlinear multi-agent systems A novel data-driven protocol for this problem is proposed by using the model-free adaptive control method To obtain faster convergence speed, one-step-ahead desired signal is introduced to construct the novel protocol Here, switching communication topology is considered, which is not required to be strongly connected all the time Through rigorous analysis, sufficient conditions are given to guarantee that the tracking errors of all agents are convergent under the novel protocol Examples are given to validate the effectiveness of results derived in this article
- 30 January 2021
- Published by SAGE Publications in Journal of Vibration and Control
- Vol. 28 (7-8) , 882-891
Consensus of nonlinear multi–agent systems with distributed event–triggered impulsive control
This study addresses the consensus problem in nonlinear multi-agent systems with the distributed impulsive control method. To decrease communication wastage, a distributed event-triggered impulsive protocol is developed to achieve consensus. In the proposed control method, each agent receives neighbor information and performs impulse control only when its own trigger function exceeds a given tolerance. Then, through the Lyapunov method, the conditions that allow the multi-agent systems reach consensus are analyzed and the Zeno behavior is excluded. Finally, some numerical simulations are performed to validate the effectiveness of the proposed method.
- 30 January 2021
- Published by SAGE Publications in Journal of Vibration and Control
- Vol. 28 (7-8) , 882-891
Consensus of nonlinear multi–agent systems with distributed event–triggered impulsive control
This study addresses the consensus problem in nonlinear multi-agent systems with the distributed impulsive control method. To decrease communication wastage, a distributed event-triggered impulsive protocol is developed to achieve consensus. In the proposed control method, each agent receives neighbor information and performs impulse control only when its own trigger function exceeds a given tolerance. Then, through the Lyapunov method, the conditions that allow the multi-agent systems reach consensus are analyzed and the Zeno behavior is excluded. Finally, some numerical simulations are performed to validate the effectiveness of the proposed method.
- 1 March 2020
- Published by Wiley in Asian Journal of Control
- Vol. 23 (3) , 1536-1544
Tracking Control for General Second‐order Multi‐agent Systems with Communication Delay and Variable Topology
- 1 March 2020
- Published by Wiley in Asian Journal of Control
- Vol. 23 (3) , 1536-1544
Tracking Control for General Second‐order Multi‐agent Systems with Communication Delay and Variable Topology