Lei Xi
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
Lei Xi has a total of 89 co-authors. Here are the top co-authors by number of shared publications:
- Yanchun Xu
College of Electrical Engineering & New Energy, China Three Gorges University, Yichang 443002, China
6 shared publications - Xi Chen
College of Electrical Engineering and New Energy, China Three Gorges University, Yichang 443002, China
5 shared publications - Kailun Wei
China Three Gorges University
4 shared publications - Can Wang
China Three Gorges University
3 shared publications - Chun Sing Lai
Tianjin University
3 shared publications - Loi Lei Lai
Department of Electrical Engineering, School of Automation, Guangdong University of Technology, Guangzhou, China
3 shared publications
Latest Publications
- 1 November 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Dependable and Secure Computing
- Vol. 23 (6) , 12812-12825
A Physics-Informed Enhanced Spatiotemporal Convolutional Network Approach for FDIA Localization
- 1 November 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Dependable and Secure Computing
- Vol. 23 (6) , 12812-12825
A Physics-Informed Enhanced Spatiotemporal Convolutional Network Approach for FDIA Localization
- 1 November 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Smart Grid
- Vol. 17 (6) , 4907-4922
Microgrid Under-Frequency Load Shedding Strategy Integrating LMC-Shapley Value and Knowledge Distillation
As the final safeguard for maintaining microgrid stability, under-frequency load shedding (UFLS) can effectively prevent rapid frequency drops caused by system faults. However, a key challenge remains in how to fairly and effectively allocate load shedding across nodes to enhance overall system resilience. To address this issue, this paper proposes a microgrid UFLS strategy that integrates Shapley value with load multiplicity characteristics (LMC-Shapley) and knowledge distillation. The strategy first constructs a node contribution factor based on the load importance, economic loss from load shedding and frequency regulation effect. Then, based on the LMC-Shapley value method, the marginal contributions and allocation coefficients of each indicator are calculated to determine the load shedding amount for each node. To mitigate the computational complexity introduced by the LMC-Shapley method, a model lightweighting method based on knowledge distillation is proposed, allowing transfer of complex knowledge from a teacher model to a student model. Finally, based on the improved IEEE-37 node system microgrid model, the performance of the proposed strategy is verified. Simulation results reveal that the proposed strategy exhibits good frequency recovery performance.
- 1 November 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Smart Grid
- Vol. 17 (6) , 4907-4922
Microgrid Under-Frequency Load Shedding Strategy Integrating LMC-Shapley Value and Knowledge Distillation
As the final safeguard for maintaining microgrid stability, under-frequency load shedding (UFLS) can effectively prevent rapid frequency drops caused by system faults. However, a key challenge remains in how to fairly and effectively allocate load shedding across nodes to enhance overall system resilience. To address this issue, this paper proposes a microgrid UFLS strategy that integrates Shapley value with load multiplicity characteristics (LMC-Shapley) and knowledge distillation. The strategy first constructs a node contribution factor based on the load importance, economic loss from load shedding and frequency regulation effect. Then, based on the LMC-Shapley value method, the marginal contributions and allocation coefficients of each indicator are calculated to determine the load shedding amount for each node. To mitigate the computational complexity introduced by the LMC-Shapley method, a model lightweighting method based on knowledge distillation is proposed, allowing transfer of complex knowledge from a teacher model to a student model. Finally, based on the improved IEEE-37 node system microgrid model, the performance of the proposed strategy is verified. Simulation results reveal that the proposed strategy exhibits good frequency recovery performance.
- 23 October 2026
- Published by Elsevier in Electric Power Systems Research
- Vol. 265
Adaptive weight and temperature-decoupled soft actor-critic algorithm for automatic generation control
- 23 October 2026
- Published by Elsevier in Electric Power Systems Research
- Vol. 265
Adaptive weight and temperature-decoupled soft actor-critic algorithm for automatic generation control
- Open Access
- 10 June 2026
- Published by Springer Nature in Nature Communications
- Vol. 17 (1)
Detection and countermeasure integrated framework against false data injection attacks in multi-area coupled power systems
False data injection attacks (FDIA) pose severe threat to the secure operation of load frequency control (LFC) systems. The accuracy of state estimation in power systems directly determines the detection precision and countermeasure effectiveness against FDIA. However, the existing state estimation algorithms are constrained by linear assumptions and the static noise models, making them difficult to adapt to the multi-area coupling characteristics of LFC systems. Moreover, the statistical overlap between the state variables and the noise cannot be effectively handled, further limiting estimation accuracy. Thus, an integrated framework for joint detection and countermeasure against FDIA in multi-area coupled power systems is proposed. In the framework, the sigma-point sampling strategy of the unscented kalman filter is employed to efficiently capture the dynamic characteristics of multi-area coupled LFC systems, thereby improving the adaptability of state estimation algorithm. In addition, an adaptive covariance update mechanism is utilized to track the statistical characteristics of noise and states in real time, reducing the impact of their overlap. Experimental results demonstrate that the proposed framework achieves more accurate state estimation compared with various state estimation algorithms. Moreover, under multiple attack scenarios, it provides superior detection accuracy and more effective countermeasures against FDIA.
- 10 June 2026
- Published by Springer Nature in Nature Communications
- Vol. 17 (1)
Open AccessDetection and countermeasure integrated framework against false data injection attacks in multi-area coupled power systems
False data injection attacks (FDIA) pose severe threat to the secure operation of load frequency control (LFC) systems. The accuracy of state estimation in power systems directly determines the detection precision and countermeasure effectiveness against FDIA. However, the existing state estimation algorithms are constrained by linear assumptions and the static noise models, making them difficult to adapt to the multi-area coupling characteristics of LFC systems. Moreover, the statistical overlap between the state variables and the noise cannot be effectively handled, further limiting estimation accuracy. Thus, an integrated framework for joint detection and countermeasure against FDIA in multi-area coupled power systems is proposed. In the framework, the sigma-point sampling strategy of the unscented kalman filter is employed to efficiently capture the dynamic characteristics of multi-area coupled LFC systems, thereby improving the adaptability of state estimation algorithm. In addition, an adaptive covariance update mechanism is utilized to track the statistical characteristics of noise and states in real time, reducing the impact of their overlap. Experimental results demonstrate that the proposed framework achieves more accurate state estimation compared with various state estimation algorithms. Moreover, under multiple attack scenarios, it provides superior detection accuracy and more effective countermeasures against FDIA.
- Open Access
- 28 April 2026
- Published by Elsevier in SSRN Electronic Journal
Automatic generation control based on Exploration-Enhanced Contrastive Learning Optimal Adversary-Aware TD3 Algorithm
The large-scale integration of renewable energy sources introduces significant stochastic disturbances, posing substantial challenges to Automatic Generation Control (AGC) in modern power systems. Reinforcement learning (RL) has shown strong potential in addressing such stochastic optimization problems, particularly in continuous control scenarios. Among existing approaches, the Twin Delayed Deep Deterministic Policy Gradient (TD3) algorithm has been widely adopted due to its effectiveness in continuous action spaces. However, TD3 often suffers from insufficient exploration and reduced robustness when operating in highly stochastic environments. To address these challenges, this paper proposes an exploration-enhanced contrastive learning and optimal adversary-aware TD3 algorithm (EECL-OATD3) for AGC that jointly improves exploration efficiency and control robustness within a single learning architecture. The proposed algorithm introduces a coupled learning mechanism, in which intrinsic exploration incentives are adaptively generated to guide the agent toward under-explored regions of the state space, while adversarial perturbation modeling is incorporated into the policy evaluation process to explicitly account for worst-case disturbances. The effectiveness of the proposed algorithm is validated on an improved IEEE standard two-area load frequency control (LFC) model and the southwest china three-area LFC model with a high proportion of renewable energy . Simulation results demonstrate that the proposed method achieves faster convergence, reduced frequency deviations, and improved AGC performance compared with existing RL-based approaches.
- 28 April 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessAutomatic generation control based on Exploration-Enhanced Contrastive Learning Optimal Adversary-Aware TD3 Algorithm
The large-scale integration of renewable energy sources introduces significant stochastic disturbances, posing substantial challenges to Automatic Generation Control (AGC) in modern power systems. Reinforcement learning (RL) has shown strong potential in addressing such stochastic optimization problems, particularly in continuous control scenarios. Among existing approaches, the Twin Delayed Deep Deterministic Policy Gradient (TD3) algorithm has been widely adopted due to its effectiveness in continuous action spaces. However, TD3 often suffers from insufficient exploration and reduced robustness when operating in highly stochastic environments. To address these challenges, this paper proposes an exploration-enhanced contrastive learning and optimal adversary-aware TD3 algorithm (EECL-OATD3) for AGC that jointly improves exploration efficiency and control robustness within a single learning architecture. The proposed algorithm introduces a coupled learning mechanism, in which intrinsic exploration incentives are adaptively generated to guide the agent toward under-explored regions of the state space, while adversarial perturbation modeling is incorporated into the policy evaluation process to explicitly account for worst-case disturbances. The effectiveness of the proposed algorithm is validated on an improved IEEE standard two-area load frequency control (LFC) model and the southwest china three-area LFC model with a high proportion of renewable energy . Simulation results demonstrate that the proposed method achieves faster convergence, reduced frequency deviations, and improved AGC performance compared with existing RL-based approaches.