Zhixiong Liu
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
Zhixiong Liu has a total of 44 co-authors. Here are the top co-authors by number of shared publications:
Bin Xiang
Electric Power Research Institute of State Grid Hubei Electric Power Company, Ltd., Wuhan, China
3 shared publicationsGuiming He
Wuhan University,Wuhan,China
3 shared publications- Han Fu
State Grid Wuhan Power Supply Company, Wuhan 430070, China
3 shared publications - Wenpei Li
State Grid Hubei Electric Power Research Institute, Wuhan 430062, China
3 shared publications Bin Song
2 shared publications- Bin Yang
State Grid Wuhan Power Supply Company, Wuhan 430070, China
2 shared publications
Latest Publications
- Open Access
- 28 April 2026
- Published by Elsevier in SSRN Electronic Journal
GMA-STLF: Short-Term Load Forecasting Based on Explicit Causal Decomposition and Gating-First Modulated Attention
Short-Term Load Forecasting (STLF) is essential for the secure and economic operation of power systems, yet it faces the dual challenges of coupled multiple seasonalities and the look-ahead bias inherent in conventional decomposition algorithms. To address this, we propose GMA-STLF, a causality-constrained short-term load forecasting framework. The method employs strictly causal MSTL to explicitly decompose load into trend, daily, and weekly components, and introduces a Gating-First Modulated Attention (GMA) mechanism that dynamically fuses physical components with heterogeneous multi-source features through a "Filter-then-Interact" strategy. Building on this design, a two-stage residual correction pipeline is introduced: the Stage 1 GMA point-forecast model provides the base forecast, while the Stage 2 MLP corrector leverages realized weather observations (Oracle weather) to correct residual patterns, with especially pronounced gains on extreme-weather samples. Validation on the NYISO dataset demonstrates that the strictly causal Stage 1 achieves 4.11% MAPE; after Oracle-weather-assisted residual correction, this is further reduced to 3.54%, numerically surpassing iTransformer (3.83%) under this upper-bound setting and confirming the effectiveness of the residual correction mechanism.
- 28 April 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessGMA-STLF: Short-Term Load Forecasting Based on Explicit Causal Decomposition and Gating-First Modulated Attention
Short-Term Load Forecasting (STLF) is essential for the secure and economic operation of power systems, yet it faces the dual challenges of coupled multiple seasonalities and the look-ahead bias inherent in conventional decomposition algorithms. To address this, we propose GMA-STLF, a causality-constrained short-term load forecasting framework. The method employs strictly causal MSTL to explicitly decompose load into trend, daily, and weekly components, and introduces a Gating-First Modulated Attention (GMA) mechanism that dynamically fuses physical components with heterogeneous multi-source features through a "Filter-then-Interact" strategy. Building on this design, a two-stage residual correction pipeline is introduced: the Stage 1 GMA point-forecast model provides the base forecast, while the Stage 2 MLP corrector leverages realized weather observations (Oracle weather) to correct residual patterns, with especially pronounced gains on extreme-weather samples. Validation on the NYISO dataset demonstrates that the strictly causal Stage 1 achieves 4.11% MAPE; after Oracle-weather-assisted residual correction, this is further reduced to 3.54%, numerically surpassing iTransformer (3.83%) under this upper-bound setting and confirming the effectiveness of the residual correction mechanism.
- Open Access
- 22 April 2026
- Published by Elsevier in SSRN Electronic Journal
GMA-STLF: Short-Term Load Forecasting Based on Explicit Causal Decomposition and Gating-First Modulated Attention
Short-Term Load Forecasting (STLF) is essential for the secure and economic operation of power systems, yet it faces the dual challenges of coupled multiple seasonalities and the look-ahead bias inherent in conventional decomposition algorithms. To address this, we propose GMA-STLF, a causality-constrained short-term load forecasting framework. The method employs strictly causal MSTL to explicitly decompose load into trend, daily, and weekly components, and introduces a Gating-First Modulated Attention (GMA) mechanism that dynamically fuses physical components with heterogeneous multi-source features through a "Filter-then-Interact" strategy. Building on this design, a two-stage residual correction pipeline is introduced: the Stage 1 GMA point-forecast model provides the base forecast, while the Stage 2 MLP corrector leverages realized weather observations (Oracle weather) to correct residual patterns, with especially pronounced gains on extreme-weather samples. Validation on the NYISO dataset demonstrates that the strictly causal Stage 1 achieves 4.11% MAPE; after Oracle-weather-assisted residual correction, this is further reduced to 3.54%, numerically surpassing iTransformer (3.83%) under this upper-bound setting and confirming the effectiveness of the residual correction mechanism.
- 22 April 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessGMA-STLF: Short-Term Load Forecasting Based on Explicit Causal Decomposition and Gating-First Modulated Attention
Short-Term Load Forecasting (STLF) is essential for the secure and economic operation of power systems, yet it faces the dual challenges of coupled multiple seasonalities and the look-ahead bias inherent in conventional decomposition algorithms. To address this, we propose GMA-STLF, a causality-constrained short-term load forecasting framework. The method employs strictly causal MSTL to explicitly decompose load into trend, daily, and weekly components, and introduces a Gating-First Modulated Attention (GMA) mechanism that dynamically fuses physical components with heterogeneous multi-source features through a "Filter-then-Interact" strategy. Building on this design, a two-stage residual correction pipeline is introduced: the Stage 1 GMA point-forecast model provides the base forecast, while the Stage 2 MLP corrector leverages realized weather observations (Oracle weather) to correct residual patterns, with especially pronounced gains on extreme-weather samples. Validation on the NYISO dataset demonstrates that the strictly causal Stage 1 achieves 4.11% MAPE; after Oracle-weather-assisted residual correction, this is further reduced to 3.54%, numerically surpassing iTransformer (3.83%) under this upper-bound setting and confirming the effectiveness of the residual correction mechanism.
- Open Access
RETRACTED: Li et al. A Kalman Filter-Based Distributed Cyber-Attack Mitigation Strategy for Distributed Generator Units in Meshed DC Microgrids. Energies 2023, 16, 7959
The journal retracts the article titled “A Kalman Filter-Based Distributed Cyber-Attack Mitigation Strategy for Distributed Generator Units in Meshed DC Microgrids”
Open AccessRETRACTED: Li et al. A Kalman Filter-Based Distributed Cyber-Attack Mitigation Strategy for Distributed Generator Units in Meshed DC Microgrids. Energies 2023, 16, 7959
The journal retracts the article titled “A Kalman Filter-Based Distributed Cyber-Attack Mitigation Strategy for Distributed Generator Units in Meshed DC Microgrids”
- 10 September 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Applied Superconductivity
- Vol. 34 (8) , 1-5
Identification Winding Material of Distribution Transformer Based on Multi-Information Fusion
In transformer production, some manufacturers use aluminum to impersonate copper to reduce the manufacturing cost, which is difficult to be detected and can cause significant losses to the power grid. To address the problem of non-destructive identification of winding material in distribution transformers, we propose a multi- information recognition method by fusing the harmonic resistance coefficient and the appearance parameters of the transformer, namely volume and height. Then the winding material is identified by a trained support vector machine model. The test results demonstrate that our proposed method achieves 90% recognition accuracy for copper transformers and 100% accuracy for aluminum transformers with the test samples. Additionally, the proposed non-destructive method is easier to implement than other methods in engineering.
- 10 September 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Applied Superconductivity
- Vol. 34 (8) , 1-5
Identification Winding Material of Distribution Transformer Based on Multi-Information Fusion
In transformer production, some manufacturers use aluminum to impersonate copper to reduce the manufacturing cost, which is difficult to be detected and can cause significant losses to the power grid. To address the problem of non-destructive identification of winding material in distribution transformers, we propose a multi- information recognition method by fusing the harmonic resistance coefficient and the appearance parameters of the transformer, namely volume and height. Then the winding material is identified by a trained support vector machine model. The test results demonstrate that our proposed method achieves 90% recognition accuracy for copper transformers and 100% accuracy for aluminum transformers with the test samples. Additionally, the proposed non-destructive method is easier to implement than other methods in engineering.
- Open Access
RETRACTED: A Kalman Filter-Based Distributed Cyber-Attack Mitigation Strategy for Distributed Generator Units in Meshed DC Microgrids
To ensure fast dynamics and the stability of multiple distributed generator units (DGUs) in DC microgrids, communication links among the controllers of DGUs are generally adopted. However, those communication channels are vulnerable to cyber-attacks. To alleviate this hassle, a Kalman Filter (KF)-based distributed cyber-attack mitigation strategy, which is highly involved in both primary and secondary control, is proposed in this paper. The KF, as a robust state estimator, is utilized to accurately estimate the authentic terminal voltages and currents of the DGUs. Based on the discrepancies between the estimated and measured parameters of the systems under cyber-attacks, the proposed control can adaptively compensate the attack signals via an adaptive proportional integral (API) controller and a fractional API (FAPI) controller in cyber-attack-mitigation layers. The main advantage of using the proposed control scheme compared to conventional schemes is the fast dynamic response. The simulation results verify this merit by comparing the adopted KF and comparing it with conventional artificial neural networks (ANN), while the experimental results validate that effectiveness of the proposed control and showcase the superiority of the FAPI control in terms of its perfect compensation for different types of cyber-attacks.
Open AccessRETRACTED: A Kalman Filter-Based Distributed Cyber-Attack Mitigation Strategy for Distributed Generator Units in Meshed DC Microgrids
To ensure fast dynamics and the stability of multiple distributed generator units (DGUs) in DC microgrids, communication links among the controllers of DGUs are generally adopted. However, those communication channels are vulnerable to cyber-attacks. To alleviate this hassle, a Kalman Filter (KF)-based distributed cyber-attack mitigation strategy, which is highly involved in both primary and secondary control, is proposed in this paper. The KF, as a robust state estimator, is utilized to accurately estimate the authentic terminal voltages and currents of the DGUs. Based on the discrepancies between the estimated and measured parameters of the systems under cyber-attacks, the proposed control can adaptively compensate the attack signals via an adaptive proportional integral (API) controller and a fractional API (FAPI) controller in cyber-attack-mitigation layers. The main advantage of using the proposed control scheme compared to conventional schemes is the fast dynamic response. The simulation results verify this merit by comparing the adopted KF and comparing it with conventional artificial neural networks (ANN), while the experimental results validate that effectiveness of the proposed control and showcase the superiority of the FAPI control in terms of its perfect compensation for different types of cyber-attacks.
- Open Access
A Communication Encryption-Based Distributed Cooperative Control for Distributed Generators in Microgrids under FDI Attacks
To alleviate the hassle of false data injection (FDI) attacks on distributed generators (DGs) in microgrids, a communication encryption-based distributed cooperative control is proposed in this paper. Compared to the conventional distributed control strategies, the proposed control scheme is simpler with much less complex evaluation mechanism by upgrading the secondary control to a second-order control based on the finite-time control theory while combining an encryption strategy. The proposed algorithm provides constant injections to eliminate the impact of FDI attacks based on a robust communication system. The effectiveness and high efficiency of the proposed control scheme is validated in an IEEE 34 Node Test Feeder system with six DGs as a microgrid cyber-physical system (CPS) under different FDI attacks.
Open AccessA Communication Encryption-Based Distributed Cooperative Control for Distributed Generators in Microgrids under FDI Attacks
To alleviate the hassle of false data injection (FDI) attacks on distributed generators (DGs) in microgrids, a communication encryption-based distributed cooperative control is proposed in this paper. Compared to the conventional distributed control strategies, the proposed control scheme is simpler with much less complex evaluation mechanism by upgrading the secondary control to a second-order control based on the finite-time control theory while combining an encryption strategy. The proposed algorithm provides constant injections to eliminate the impact of FDI attacks based on a robust communication system. The effectiveness and high efficiency of the proposed control scheme is validated in an IEEE 34 Node Test Feeder system with six DGs as a microgrid cyber-physical system (CPS) under different FDI attacks.
- Open Access
- 3 July 2023
- Published by Institution of Engineering and Technology (IET) in IET Renewable Power Generation
- Vol. 17 (11) , 2779-2790
Power utilization efficiency improvement of distributed energy units in DC microgrids using primal‐dual subgradient method
To improve the power utilization efficiency of distributed energy units (DEUs) in DC microgrids, a centralized current iteration regulation approach is presented in this paper. The power utilization efficiency is partially represented by the power losses of DEUs, which can be modelled as quadratic functions of the output currents of DEUs. The power efficiency is also dependent on the line losses on the resistances between the DEUs and DC buses, which can be formulated in the basis of the output currents as well. Thus, the total power losses of DEUs comprise line and converter losses, which are verified to be a constrained‐convex function of the current allocation coefficients. In this paper, a primal‐dual subgradient method (PDSM) using only explicit addition and subtraction iterations is adopted to obtain the optimal current distribution coefficients. On this basis, an adaptive droop regulation scheme is applied to further control the output voltage references of DEUs, and proportional‐integral (PI) controllers are applied to track the adaptive voltages based on the current sharing errors. Both simulation and empirical results demonstrate the effectiveness of the proposed control to improve the power utilization efficiency of DEUs regardless parameter and load variations in DC microgrids.
- 3 July 2023
- Published by Institution of Engineering and Technology (IET) in IET Renewable Power Generation
- Vol. 17 (11) , 2779-2790
Open AccessPower utilization efficiency improvement of distributed energy units in DC microgrids using primal‐dual subgradient method
To improve the power utilization efficiency of distributed energy units (DEUs) in DC microgrids, a centralized current iteration regulation approach is presented in this paper. The power utilization efficiency is partially represented by the power losses of DEUs, which can be modelled as quadratic functions of the output currents of DEUs. The power efficiency is also dependent on the line losses on the resistances between the DEUs and DC buses, which can be formulated in the basis of the output currents as well. Thus, the total power losses of DEUs comprise line and converter losses, which are verified to be a constrained‐convex function of the current allocation coefficients. In this paper, a primal‐dual subgradient method (PDSM) using only explicit addition and subtraction iterations is adopted to obtain the optimal current distribution coefficients. On this basis, an adaptive droop regulation scheme is applied to further control the output voltage references of DEUs, and proportional‐integral (PI) controllers are applied to track the adaptive voltages based on the current sharing errors. Both simulation and empirical results demonstrate the effectiveness of the proposed control to improve the power utilization efficiency of DEUs regardless parameter and load variations in DC microgrids.
- Open Access
- 15 February 2023
- Published by Frontiers Media SA in Frontiers in Energy Research
- Vol. 11
Overheating fault alarming for compact insulated busways in buildings by gas sensing
Addressing the overheating fault detection and alarming of insulated busways in buildings, a system based on sensing volatile gases generated by the thermal degradation of the busduct insulation layer was proposed. By monitoring the concentration of volatile gases in the environment, the overheating fault of the busducts can be found early. The thermal degradation process of the busway insulating layer is analyzed, and the pyrolysis characteristic gas at low temperature is studied. The experimental platform has been built, by which the relation between the concentration of volatile gases and temperature of the insulated layer has been studied. By testing, the concentration of volatile organic compounds (VOCs) is proposed as the basis for judging the overheating fault in the alarming system. With the collected samples for training and testing, the AdaBoost classifier is used to identify the overheating fault. Finally, the design of the overheating fault alarming system is given.
- 15 February 2023
- Published by Frontiers Media SA in Frontiers in Energy Research
- Vol. 11
Open AccessOverheating fault alarming for compact insulated busways in buildings by gas sensing
Addressing the overheating fault detection and alarming of insulated busways in buildings, a system based on sensing volatile gases generated by the thermal degradation of the busduct insulation layer was proposed. By monitoring the concentration of volatile gases in the environment, the overheating fault of the busducts can be found early. The thermal degradation process of the busway insulating layer is analyzed, and the pyrolysis characteristic gas at low temperature is studied. The experimental platform has been built, by which the relation between the concentration of volatile gases and temperature of the insulated layer has been studied. By testing, the concentration of volatile organic compounds (VOCs) is proposed as the basis for judging the overheating fault in the alarming system. With the collected samples for training and testing, the AdaBoost classifier is used to identify the overheating fault. Finally, the design of the overheating fault alarming system is given.
- Open Access
- 26 February 2022
- Published by Elsevier in Energy Reports
- Vol. 8, 243-250
Pointer meter recognition in UAV inspection of overhead transmission lines
- 26 February 2022
- Published by Elsevier in Energy Reports
- Vol. 8, 243-250
Open AccessPointer meter recognition in UAV inspection of overhead transmission lines
- Open Access
- 26 February 2022
- Published by Elsevier in Energy Reports
- Vol. 8, 259-268
Load characteristics forecasting of Hubei power grid up to year 2030 with the development of electric vehicles
- 26 February 2022
- Published by Elsevier in Energy Reports
- Vol. 8, 259-268
Open AccessLoad characteristics forecasting of Hubei power grid up to year 2030 with the development of electric vehicles