Multi-agent reinforcement learning for power-system control
Cooperative decision-making across heterogeneous devices and action spaces for volt–var and load-frequency control.
Multi-agent RL / Heterogeneous resources / Grid control
Ph.D. in Electrical Engineering
郑晓东
Intelligent decision-making and control for power systems
Xiaodong Zheng received his Ph.D. in Electrical Engineering from Xi’an Jiaotong University in 2026, supervised by Prof. Hui Cao. He previously earned an M.S. in Electrical Engineering from the University of Southern California and a B.S. in Electrical Engineering from Florida Institute of Technology. His research focuses on multi-agent reinforcement learning, power-system time-series forecasting, and large-language-model agents, with applications to active distribution networks, coordinated energy storage scheduling, and power-system simulation.

Cooperative decision-making across heterogeneous devices and action spaces for volt–var and load-frequency control.
Multi-agent RL / Heterogeneous resources / Grid control
Graph learning, reinforcement learning, and time-series foundation models for charging-station scheduling, urban load forecasting, and wind-power forecasting.
Graph learning / Time series / Storage scheduling
LLM-based task planning, simulation-tool orchestration, result validation, and reproducible multi-agent training environments.
Language models / Task planning / Simulation environments
Electric Power Systems Research, 265: 114258Published online (2027 issue)Journal article
A dual-stream large–lightweight framework for urban load forecasting through trend–fluctuation decomposition.
@article{urbanload2027,
title = {Making large time series models work for urban load forecasting: A dual-stream large-lightweight hybrid framework with trend–fluctuation decomposition},
author = {Xiaodong Zheng and Tiansen Niu and Junkai Tan and Qingshu Guan and Xinyao Dong and Shuangsi Xue and Hui Cao},
journal = {Electric Power Systems Research},
year = {2027},
volume = {265},
pages = {114258},
doi = {10.1016/j.epsr.2026.114258},
note = {Published online; assigned to the 2027 issue}
}Neurocomputing, 699: 134376PublishedJournal article
@article{graphrl2026,
title = {Graph-Oriented Deep Reinforcement Learning Approach for Vehicle Routing Problems},
author = {Xiaodong Zheng and Qingshu Guan and Shuangsi Xue and Hui Cao and Badong Chen},
journal = {Neurocomputing},
year = {2026},
volume = {699},
pages = {134376},
doi = {10.1016/j.neucom.2026.134376}
}Power System Technology, 50(5): 1973–1984Published; in ChineseJournal article
Multi-charging-station cooperative scheduling with virtual aggregated energy storage, coupling traffic prediction and multi-agent reinforcement learning.
@article{storage2026,
title = {耦合交通预测多智能体强化学习的虚拟聚合储能多充电站协同调度},
author = {郑晓东 and 石天卓 and 张盼盼 and 张晓彤 and 薛霜思 and 丁涛 and 曹晖},
journal = {Power System Technology},
year = {2026},
volume = {50},
number = {5},
pages = {1973--1984},
doi = {10.13335/j.1000-3673.pst.2025.0918}
}IEEE Transactions on Power Systems, 41(1): 413–424PublishedJournal article
@article{frequency2026,
title = {Order-Based Heterogeneous Agents Reinforcement Learning Method With the Coordination of Distribution Network and Transmission Network for Load Frequency Control},
author = {Shixuan Yu and Xiaodong Zheng and Tianzhuo Shi and Ruilin Chen and Shuangsi Xue and Hui Cao},
journal = {IEEE Transactions on Power Systems},
year = {2026},
volume = {41},
number = {1},
pages = {413--424},
doi = {10.1109/TPWRS.2025.3598951}
}IEEE Transactions on Industrial Informatics, Early AccessEarly AccessJournal article
@article{resilientcontrol2026,
title = {Fixed-Time Convergent Resilient Critic-Learning Control With Asymmetric Input-State Constraints Under Hybrid FDI–DoS Attacks},
author = {Junkai Tan and Shuangsi Xue and Qingshu Guan and Xiaodong Zheng and Hui Cao and Badong Chen},
journal = {IEEE Transactions on Industrial Informatics},
year = {2026},
doi = {10.1109/TII.2026.3714681},
note = {Early Access}
}Renewable Energy, 258: 124969PublishedJournal article
@article{windforecast2026,
title = {Learning dynamic inter-farm dependencies for wind power forecasting via adaptive sparse graph attention network},
author = {Kai Qu and Shuangsi Xue and Xiaodong Zheng and Dapeng Yan and Hui Cao},
journal = {Renewable Energy},
year = {2026},
volume = {258},
pages = {124969},
doi = {10.1016/j.renene.2025.124969}
}Neurocomputing, 674: 132945PublishedJournal article
@article{stimi2026,
title = {STIMI: A masked image modeling framework for spatiotemporal wind speed reconstruction},
author = {Kai Qu and Shuangsi Xue and Xiaodong Zheng and Hui Cao},
journal = {Neurocomputing},
year = {2026},
volume = {674},
pages = {132945},
doi = {10.1016/j.neucom.2026.132945},
url = {https://doi.org/10.1016/j.neucom.2026.132945}
}Automation of Electric Power Systems, 50(7): 206–217PublishedJournal article
@article{adpdann2026,
title = {基于ADP-DANN的小样本风电集群短期功率预测方法},
author = {曲凯 and 闫大鹏 and 郑晓东 and 薛霜思 and 曹晖},
journal = {Automation of Electric Power Systems},
year = {2026},
volume = {50},
number = {7},
pages = {206--217},
doi = {10.7500/AEPS20250810002},
url = {https://kns.cnki.net/KCMS/detail/detail.aspx?dbcode=CAPJ&dbname=CAPJLAST&filename=DLXT20260211001}
}Journal of Electronic Measurement and Instrumentation, 40(3): 46–57PublishedJournal article
@article{soundseparation2026,
title = {多尺度门控网络的变电站复杂声源分离研究},
author = {陈睿霖 and 曹晖 and 郑晓东 and 闫大鹏 and 薛霜思 and 曲凯 and 汲胜昌},
journal = {Journal of Electronic Measurement and Instrumentation},
year = {2026},
volume = {40},
number = {3},
pages = {46--57},
url = {https://kns.cnki.net/KCMS/detail/detail.aspx?dbcode=CAPJ&dbname=CAPJLAST&filename=DZIY2025122201X}
}Engineering Applications of Artificial Intelligence, 163: 113013PublishedJournal article
@article{quadcoptercontrol2026,
title = {Adaptive hierarchical control of quadcopters via safe reinforcement learning from human demonstration},
author = {Junkai Tan and Shuangsi Xue and Zihang Guo and Huan Li and Xiaodong Zheng and Hui Cao},
journal = {Engineering Applications of Artificial Intelligence},
year = {2026},
volume = {163},
pages = {113013},
doi = {10.1016/j.engappai.2025.113013},
url = {https://doi.org/10.1016/j.engappai.2025.113013}
}Electric Power Systems Research, 254: 112679PublishedJournal article
@article{tsdbc2026,
title = {TS-DBC: A teacher-student dual-branch classifier for few-shot acoustic fault detection in substations},
author = {Ruilin Chen and Xiaodong Zheng and Hui Cao and Tiansen Niu and Shixuan Yu and Fan Zhang and Shengchang Ji},
journal = {Electric Power Systems Research},
year = {2026},
volume = {254},
pages = {112679},
doi = {10.1016/j.epsr.2025.112679},
url = {https://doi.org/10.1016/j.epsr.2025.112679}
}International Conference on Learning Representations (ICLR) 2026, PosterPublished · PosterConference paper
@inproceedings{fzoo2026,
title = {FZOO: Fast Zeroth-Order Optimizer for Fine-Tuning Large Language Models towards Adam-Scale Speed},
author = {Sizhe Dang and Yangyang Guo and Yanjun Zhao and Xiaodong Zheng and Guang Dai and Ivor Tsang and Haishan Ye},
booktitle = {International Conference on Learning Representations (ICLR) 2026},
year = {2026},
url = {https://openreview.net/forum?id=NMlF3YjS8E}
}arXiv, arXiv:2602.08007PreprintPreprint
@misc{tsradam2026,
title = {From O(mn) to O(r²): Two-Sided Low-Rank Communication for Adam in Distributed Training with Memory Efficiency},
author = {Sizhe Dang and Jiaqi Shao and Xiaodong Zheng and Guang Dai and Yan Song and Haishan Ye},
howpublished = {arXiv},
year = {2026},
url = {https://arxiv.org/abs/2602.08007},
eprint = {2602.08007},
archivePrefix = {arXiv},
note = {Preprint}
}Xi’an Jiaotong University, Doctoral dissertationDoctoral dissertationThesis
@phdthesis{doctoralthesis2026,
title = {基于多智能体强化学习的主动配电网异质性协同分布式控制方法研究},
author = {郑晓东},
school = {Xi’an Jiaotong University},
year = {2026}
}IEEE Transactions on Smart Grid, 16(3): 2115–2126PublishedJournal article
Sensitivity-based ordering and heterogeneous multi-agent learning for distributed volt–var control in active distribution networks.
@article{shom2025,
title = {Sensitivity-Based Heterogeneous Ordered Multi-Agent Reinforcement Learning for Distributed Volt-Var Control in Active Distribution Network},
author = {Xiaodong Zheng and Shixuan Yu and Hui Cao and Tianzhuo Shi and Shuangsi Xue and Tao Ding},
journal = {IEEE Transactions on Smart Grid},
year = {2025},
volume = {16},
number = {3},
pages = {2115--2126},
doi = {10.1109/TSG.2025.3540416}
}2025 8th International Conference on Robotics, Control and Automation Engineering (RCAE), 576–584PublishedConference paper
@inproceedings{agenticplanning2025,
title = {Agentic Planning for Power System Simulation Task Orchestrations Based on Multi-Agent Systems},
author = {Xiaodong Zheng and Siheng Zhao and Tianzhuo Shi and Tao Wang and Ruilin Chen and Shixuan Yu and Shuangsi Xue and Hui Cao},
booktitle = {2025 8th International Conference on Robotics, Control and Automation Engineering (RCAE)},
year = {2025},
pages = {576--584},
doi = {10.1109/RCAE66389.2025.11355202}
}Intelligent Robotics and Applications, 74–89PublishedConference paper
@inproceedings{manipulatorcontrol2025,
title = {Fixed-Time Fault-Tolerant Control for Robotic Manipulators Based on Bias Neural Network with Input Saturation},
author = {Zihang Guo and Shuangsi Xue and Huan Li and Junkai Tan and Xiaodong Zheng and Liangliang Xia and Hui Cao},
booktitle = {Intelligent Robotics and Applications},
year = {2025},
pages = {74--89},
doi = {10.1007/978-981-96-0798-3_7},
url = {https://doi.org/10.1007/978-981-96-0798-3_7}
}2025 Joint International Conference on Automation-Intelligence-Safety (ICAIS) & International Symposium on Autonomous Systems (ISAS), 1–6PublishedConference paper
@inproceedings{eventconsensus2025,
title = {Consensus Control via Edge-Event-Triggered Mechanism for Heterogeneous Multi-Agent Systems Under Disturbances},
author = {Huan Li and Xiaodong Zheng and Shuangsi Xue and Hui Cao and Junkang Ni},
booktitle = {2025 Joint International Conference on Automation-Intelligence-Safety (ICAIS) & International Symposium on Autonomous Systems (ISAS)},
year = {2025},
pages = {1--6},
doi = {10.1109/ICAISISAS64483.2025.11052104},
url = {https://doi.org/10.1109/ICAISISAS64483.2025.11052104}
}The International Journal of Intelligent Control and Systems, 30(4)PublishedJournal article
@article{weightedsign2025,
title = {Weighted-Sign Based Sequential Synchronization Control via Residual Reinforcement Learning},
author = {Xiaodong Zheng and Shuangsi Xue and Hui Cao},
journal = {The International Journal of Intelligent Control and Systems},
year = {2025},
volume = {30},
number = {4},
doi = {10.62678/ijics202512.10255},
url = {https://doi.org/10.62678/ijics202512.10255}
}SSRNPreprintPreprint
@misc{windimputationpreprint2025,
title = {Spatiotemporal Context-Aware Imaging for Robust Imputation of Wind Speed Data in Regional Wind Farms},
author = {Kai Qu and Shuangsi Xue and Xiaodong Zheng and Dapeng Yan and Hui Cao},
howpublished = {SSRN},
year = {2025},
doi = {10.2139/ssrn.5314740},
url = {https://doi.org/10.2139/ssrn.5314740},
note = {Preprint}
}Intelligent Robotics and Applications, 344–359PublishedConference paper
@inproceedings{expertap2025,
title = {ExpertAP: Leveraging Multi-unit Operational Patterns for Advanced Turbine Anomaly Prediction},
author = {Yuanxian Liang and Xiaodong Zheng and Dongjun Pang and Wenhao Zhu and Maoyuan Liu and Shuangsi Xue and Hui Cao},
booktitle = {Intelligent Robotics and Applications},
year = {2025},
pages = {344--359},
doi = {10.1007/978-981-96-0783-9_24},
url = {https://doi.org/10.1007/978-981-96-0783-9_24}
}IEEE Transactions on Systems, Man, and Cybernetics: Systems, 54(2): 1214–1225PublishedJournal article
@article{leaderfollowing2024,
title = {Leader-Following Formation Tracking of Multiagent Systems Using Adaptive Scaling Mechanism Under Spatial Constraints},
author = {Shuangsi Xue and Jinyu Liu and Hui Cao and Xiaodong Zheng and Huan Li and Jie Zhang},
journal = {IEEE Transactions on Systems, Man, and Cybernetics: Systems},
year = {2024},
volume = {54},
number = {2},
pages = {1214--1225},
doi = {10.1109/TSMC.2023.3323995},
url = {https://doi.org/10.1109/TSMC.2023.3323995}
}Proceedings of 2021 5th Chinese Conference on Swarm Intelligence and Cooperative Control, Lecture Notes in Electrical Engineering 934: 1112–1122PublishedConference paper
@inproceedings{flocking2023,
title = {Two-Stage Flocking Formation Control Based on Constrained Boundary},
author = {Xiaodong Zheng and Tao Wang and Shuangsi Xue and Jinyu Liu and Hui Cao and Feihu Hu},
booktitle = {Proceedings of 2021 5th Chinese Conference on Swarm Intelligence and Cooperative Control},
year = {2023},
volume = {934},
pages = {1112--1122},
doi = {10.1007/978-981-19-3998-3_106},
url = {https://doi.org/10.1007/978-981-19-3998-3_106}
}Journal of China Three Gorges University (Natural Sciences), 44(2): 100–106PublishedJournal article
@article{firefightingkinematics2022,
title = {面向变电站智能消防机械臂的高精度运动学模型研究},
author = {汤锦慧 and 伍发元 and 辛建波 and 郑晓东 and 薛霜思},
journal = {Journal of China Three Gorges University (Natural Sciences)},
year = {2022},
volume = {44},
number = {2},
pages = {100--106},
doi = {10.13393/j.cnki.issn.1672-948x.2022.02.016},
url = {https://doi.org/10.13393/j.cnki.issn.1672-948x.2022.02.016}
}Smart Power, 49(8): 46–54PublishedJournal article
@article{edgeenergy2021,
title = {基于边缘计算的能量自治区域调度策略},
author = {袁性忠 and 张鹭 and 罗迪 and 赖轩达 and 李雪 and 郑晓东},
journal = {Smart Power},
year = {2021},
volume = {49},
number = {8},
pages = {46--54}
}2020 Chinese Automation Congress (CAC), 3374–3379PublishedConference paper
@inproceedings{bluetoothmesh2020,
title = {A Cost-efficient Smart IoT Device Controlling System Based on Bluetooth Mesh and Cloud Computing},
author = {Xiaodong Zheng and Shuangsi Xue and Hui Cao and Feng Wang and Mingke Zhang},
booktitle = {2020 Chinese Automation Congress (CAC)},
year = {2020},
pages = {3374--3379},
doi = {10.1109/CAC51589.2020.9326634},
url = {https://doi.org/10.1109/CAC51589.2020.9326634}
}Complexity, 2020: 8840784PublishedJournal article
@article{fuzzycontrol2020,
title = {Fuzzy Model-Based Asynchronous Control for Markov Switching Systems with Stochastic Fading Channels},
author = {Fayuan Wu and Jinhui Tang and Zhuang Liu and Qi Xiao and Xiaodong Zheng and Shuangsi Xue},
journal = {Complexity},
year = {2020},
volume = {2020},
pages = {8840784},
doi = {10.1155/2020/8840784},
url = {https://doi.org/10.1155/2020/8840784}
}No works match these filters.
A multi-agent reinforcement learning platform for power-system control research, with standardized training environments and simulation cases.
The showcase presents steady-state scenarios and the research platform.
Sensitivity-based heterogeneous ordered multi-agent reinforcement learning for distributed volt–var control, with code accompanying the IEEE Transactions on Smart Grid paper.
A literature workflow connecting paper discovery, Zotero, Notion, and LLM-assisted summaries.
Ph.D. in Electrical Engineering
Advisor: Prof. Hui Cao
M.S. in Electrical Engineering
B.S. in Electrical Engineering
Magna Cum Laude
For research collaboration, open-source projects, and academic exchange.