Lifu Ding
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Lifu Ding has a total of 27 co-authors. Here are the top co-authors by number of shared publications:
Gangfeng Yan
College of Electrical Engineering, Zhejiang University, Hangzhou, China
4 shared publications- Ying Chen
Tsinghua University
4 shared publications - Tannan Xiao
State Key Laboratory of Power System Operation and Control, Department of Electrical Engineering, Tsinghua University, Beijing 100084, China
3 shared publications - Xiasheng Shi
Key Laboratory of Advanced Process Control for Light Industry, Jiangnan University, Wuxi, China
3 shared publications - Zhiyun Lin
Southern University of Science and Technology
2 shared publications Ao Guo
1 shared publication
Latest Publications
- 4 September 2026
- Published by Elsevier in Energy Conversion and Management
- Vol. 370
Breaking the dimensionality curse in floating offshore wind farms: A spatiotemporal decoupled reduced-order modeling framework
- 4 September 2026
- Published by Elsevier in Energy Conversion and Management
- Vol. 370
Breaking the dimensionality curse in floating offshore wind farms: A spatiotemporal decoupled reduced-order modeling framework
- Open Access
- 20 May 2026
- Published by Springer Nature in Nature Communications
- Vol. 17 (1)
Digital twin-driven fault diagnosis of power substations by multi-modal fusion learning
Substations are critical infrastructures for ensuring reliable power system operation. With increasing digitalization and system complexity, the rapid growth of multi-source data poses significant challenges for accurate and timely fault diagnosis. Existing approaches often struggle to effectively integrate heterogeneous data or adapt to varying operating conditions. To address these limitations, this study proposes a digital twin-driven fault diagnosis framework incorporating a multi-modal fusion model that integrates system topology, alarms, fault waveforms, and SCADA data through Graph Attention Networks and self-attention mechanisms. In this work, the method is validated on a 110 kV substation using 11,597 training and 3,890 testing scenarios generated in CloudPSS. Experimental results demonstrate over 95% accuracy in fault location, 97% in fault type identification, and 90% accuracy in protection failure detection under 30% data loss conditions. The deployed digital twin system further verifies the practical feasibility of the proposed approach, highlighting its robustness in complex operating environments.
- 20 May 2026
- Published by Springer Nature in Nature Communications
- Vol. 17 (1)
Open AccessDigital twin-driven fault diagnosis of power substations by multi-modal fusion learning
Substations are critical infrastructures for ensuring reliable power system operation. With increasing digitalization and system complexity, the rapid growth of multi-source data poses significant challenges for accurate and timely fault diagnosis. Existing approaches often struggle to effectively integrate heterogeneous data or adapt to varying operating conditions. To address these limitations, this study proposes a digital twin-driven fault diagnosis framework incorporating a multi-modal fusion model that integrates system topology, alarms, fault waveforms, and SCADA data through Graph Attention Networks and self-attention mechanisms. In this work, the method is validated on a 110 kV substation using 11,597 training and 3,890 testing scenarios generated in CloudPSS. Experimental results demonstrate over 95% accuracy in fault location, 97% in fault type identification, and 90% accuracy in protection failure detection under 30% data loss conditions. The deployed digital twin system further verifies the practical feasibility of the proposed approach, highlighting its robustness in complex operating environments.
- 24 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Power Systems
- Vol. 41 (3) , 2034-2048
Robust State Estimation for Distribution Systems via Graph Signal Processing
This paper addresses the challenges of accurate state estimation in distribution grids, particularly the issues of sparse measurements and outliers in modern systems. A robust state estimation method is proposed, utilizing graph signal processing (GSP). By incorporating constraint terms, first- and second-order Graph Low-Frequency Measurements are derived, which exhibit low-pass characteristics and enhance state estimation accuracy. Additionally, a sliding-window-based outlier measurement identification technique is introduced, effectively identifying and mitigating the impact of outlier data. Extensive testing demonstrates the effectiveness of the proposed method in improving both state estimation accuracy and outlier detection performance.
- 24 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Power Systems
- Vol. 41 (3) , 2034-2048
Robust State Estimation for Distribution Systems via Graph Signal Processing
This paper addresses the challenges of accurate state estimation in distribution grids, particularly the issues of sparse measurements and outliers in modern systems. A robust state estimation method is proposed, utilizing graph signal processing (GSP). By incorporating constraint terms, first- and second-order Graph Low-Frequency Measurements are derived, which exhibit low-pass characteristics and enhance state estimation accuracy. Additionally, a sliding-window-based outlier measurement identification technique is introduced, effectively identifying and mitigating the impact of outlier data. Extensive testing demonstrates the effectiveness of the proposed method in improving both state estimation accuracy and outlier detection performance.
- Open Access
- 1 January 2025
- Published by Institution of Engineering and Technology (IET) in Electronics Letters
- Vol. 61 (1)
Wake‐Aware Online Cluster Optimization for Flexible Wind Farm Yaw Control
The integration of wind energy into power systems is expanding rapidly, bringing with it the need for optimized wind farm operations to maximize power output and efficiency. This paper addresses the critical challenge of yaw control in wind farms, which involves adjusting turbine orientation to optimize alignment with wind direction while accounting for complex wake effects that can significantly impact downstream turbines. To tackle these issues, this paper proposed an online cluster optimization algorithm designed for real‐time adaptation to changing wind conditions. This algorithm features multi‐objective optimization to balance the maximization of power output against the operational costs associated with yaw adjustments, including potential mechanical wear. Through case studies, the proposed method demonstrates its ability to enhance the overall performance of wind farms, achieving a 15.24% increase in power generation compared to the baseline greedy strategy, by mitigating wake effects and optimizing yaw angles under dynamic operational constraints.
- 1 January 2025
- Published by Institution of Engineering and Technology (IET) in Electronics Letters
- Vol. 61 (1)
Open AccessWake‐Aware Online Cluster Optimization for Flexible Wind Farm Yaw Control
The integration of wind energy into power systems is expanding rapidly, bringing with it the need for optimized wind farm operations to maximize power output and efficiency. This paper addresses the critical challenge of yaw control in wind farms, which involves adjusting turbine orientation to optimize alignment with wind direction while accounting for complex wake effects that can significantly impact downstream turbines. To tackle these issues, this paper proposed an online cluster optimization algorithm designed for real‐time adaptation to changing wind conditions. This algorithm features multi‐objective optimization to balance the maximization of power output against the operational costs associated with yaw adjustments, including potential mechanical wear. Through case studies, the proposed method demonstrates its ability to enhance the overall performance of wind farms, achieving a 15.24% increase in power generation compared to the baseline greedy strategy, by mitigating wake effects and optimizing yaw angles under dynamic operational constraints.
- 6 November 2024
- Published by Elsevier in Applied Energy
- Vol. 378
Topology-aware fault diagnosis for microgrid clusters with diverse scenarios generated by digital twins
- 6 November 2024
- Published by Elsevier in Applied Energy
- Vol. 378
Topology-aware fault diagnosis for microgrid clusters with diverse scenarios generated by digital twins
- 21 October 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Industrial Informatics
- Vol. 21 (2) , 1150-1159
Observer-Based Multiagent Deep Reinforcement Learning: A Fully Distributed Training Scheme
With the extensive application of multiagent reinforcement learning (MARL), it encounters great obstacles in more complex engineering applications. Realistic environment in MARL is partially observable to the agents. Previous algorithms are not fully distributed to address the instability associated with partial observability, which presents difficulties in deploying. In this article, a theoretical analysis of partial observability of the policy-based method is given, and a fully distributed multiagent proximal policy optimization based on a distributed observer is proposed. In the algorithm, each agent has its own independent critic and actor, and performs policy iteration locally by its observers to achieve global optimization. With a fully distributed mapping method, actions are bound to adjacent spaces that satisfy constraints. After subjecting the algorithm to rigorous testing on a generalized constrained optimization model, the obtained results unequivocally demonstrate its high stability and superior decision-making prowess.
- 21 October 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Industrial Informatics
- Vol. 21 (2) , 1150-1159
Observer-Based Multiagent Deep Reinforcement Learning: A Fully Distributed Training Scheme
With the extensive application of multiagent reinforcement learning (MARL), it encounters great obstacles in more complex engineering applications. Realistic environment in MARL is partially observable to the agents. Previous algorithms are not fully distributed to address the instability associated with partial observability, which presents difficulties in deploying. In this article, a theoretical analysis of partial observability of the policy-based method is given, and a fully distributed multiagent proximal policy optimization based on a distributed observer is proposed. In the algorithm, each agent has its own independent critic and actor, and performs policy iteration locally by its observers to achieve global optimization. With a fully distributed mapping method, actions are bound to adjacent spaces that satisfy constraints. After subjecting the algorithm to rigorous testing on a generalized constrained optimization model, the obtained results unequivocally demonstrate its high stability and superior decision-making prowess.
- Open Access
- 26 February 2024
- Published by Elsevier in International Journal of Electrical Power & Energy Systems
- Vol. 157
Distributed energy management of multi-area integrated energy system based on multi-agent deep reinforcement learning
- 26 February 2024
- Published by Elsevier in International Journal of Electrical Power & Energy Systems
- Vol. 157
Open AccessDistributed energy management of multi-area integrated energy system based on multi-agent deep reinforcement learning
- 4 July 2022
- Published by Wiley in International Journal of Intelligent Systems
- Vol. 37 (11) , 8387-8411
Multiagent reinforcement learning for strictly constrained tasks based on Reward Recorder
- 4 July 2022
- Published by Wiley in International Journal of Intelligent Systems
- Vol. 37 (11) , 8387-8411
Multiagent reinforcement learning for strictly constrained tasks based on Reward Recorder
- 16 March 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Power Systems
- Vol. 38 (1) , 204-217
Target-Value-Competition-Based Multi-Agent Deep Reinforcement Learning Algorithm for Distributed Nonconvex Economic Dispatch
With the increasing expansion of the power grid, economic dispatch problems have received considerable attention. A multi-agent coordinated deep reinforcement learning algorithm is proposed to deal with distributed nonconvex economic dispatch problems. In the algorithm, agents run independent reinforcement learning algorithms and update their local Q-functions with a newly defined joint reward. The double network structure is adopted to approximate the Q-function so that the offline trained model can be used online to provide recommended power outputs for time-varying demands in real-time. By introducing the reward network, the competition mechanism between the reward network and the target network is established to determine a progressively stable target value, which achieves coordination among agents and pledges the losses of the Q-networks to converge well. Theoretical analysis is given and case studies are conducted to prove the advantages compared with existing approaches.
- 16 March 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Power Systems
- Vol. 38 (1) , 204-217
Target-Value-Competition-Based Multi-Agent Deep Reinforcement Learning Algorithm for Distributed Nonconvex Economic Dispatch
With the increasing expansion of the power grid, economic dispatch problems have received considerable attention. A multi-agent coordinated deep reinforcement learning algorithm is proposed to deal with distributed nonconvex economic dispatch problems. In the algorithm, agents run independent reinforcement learning algorithms and update their local Q-functions with a newly defined joint reward. The double network structure is adopted to approximate the Q-function so that the offline trained model can be used online to provide recommended power outputs for time-varying demands in real-time. By introducing the reward network, the competition mechanism between the reward network and the target network is established to determine a progressively stable target value, which achieves coordination among agents and pledges the losses of the Q-networks to converge well. Theoretical analysis is given and case studies are conducted to prove the advantages compared with existing approaches.