Special Session III: Data-AI Driven State Estimation and Stability Analysis 数据—人工智能驱动的状态估计与稳定性分析
 |
 |
| Session Chair: Asst. Researcher Peng Wang, Shandong University, China | Co-Chair: Assoc. Prof. Yiyan Li, Shanghai Jiao Tong University, China |
Summary:
With the large-scale integration of renewable energy resources, power electronic devices, and multi-source sensing infrastructures, modern power systems are becoming increasingly stochastic, highly coupled, and dynamically time-varying. These features impose new requirements on accurate state estimation, online stability assessment, and early-warning technologies. This special session focuses on data- and AI-driven methods for state awareness, dynamic modeling, stability analysis, and online risk prediction. Topics include multi-source heterogeneous measurement fusion, physics-informed and data-driven modeling, robust and distributed state estimation, low-frequency and wideband stability analysis, impedance identification, digital twins, and AI-assisted control. The session aims to promote the application of data-AI technologies in the secure and stable.
随着新能源、电力电子装备和多源感知设备的大规模接入，电力系统运行状态呈现强随机性、强耦合性和快速动态变化特征，对高精度状态估计、在线稳定分析和风险预警提出了更高要求。本特邀专场聚焦数据与人工智能驱动的状态感知、动态建模、稳定性评估和在线预警方法，重点讨论多源异构量测融合、物理机理与数据模型协同、鲁棒/分布式状态估计、低频与宽频稳定分析、阻抗辨识、数字孪生和AI辅助控制等关键技术，推动数据-AI方法在新型电力系统安全稳定运行中的应用。
Topics of interest include, but are not limited to:
1. Static and dynamic state estimation methods for power systems with high penetration of renewable energy resources （面向新能源高比例接入电力系统的静态/动态状态估计方法）
2. Multi-source information fusion and state awareness based on PMU, SCADA, WAMS, edge measurements, and equipment operation data （基于PMU、SCADA、WAMS、边缘量测和设备运行数据的多源信息融合与状态感知）
3. Robust state estimation methods considering measurement noise, bad data, communication delays, and missing data （考虑量测噪声、坏数据、通信延迟和缺失数据的鲁棒状态估计方法）
4. Power system state identification and dynamic prediction based on graph neural networks, deep learning, reinforcement learning, and foundation models （基于图神经网络、深度学习、强化学习和大模型的电力系统状态辨识与动态预测）
5. Physics-informed and data-driven methods for stability assessment, stability margin evaluation, and stability boundary identification （物理机理与数据驱动融合的稳定性评估、稳定裕度计算和稳定边界识别方法）
6. Low-frequency oscillation, wideband oscillation, and impedance-based stability analysis of renewable-energy-integrated power systems （新能源并网系统低频振荡、宽频振荡和阻抗稳定性分析）
7. Impedance identification, modal analysis, and oscillation source location methods for converter-dominated power systems （变流器主导电力系统的阻抗辨识、模态分析和振荡源定位方法）
8. Digital twins, model updating, and real-time simulation methods for online stability assessment and early warning （面向在线评估与预警的电力系统数字孪生、模型校正和实时仿真方法）
9. Data- and AI-assisted methods for stability control, oscillation damping, and operation optimization （数据—人工智能辅助的稳定控制、振荡抑制和运行优化方法）
10. Applications of trustworthy AI, explainable AI, and data security technologies in state estimation and stability analysis （可信人工智能、可解释人工智能和数据安全技术在状态估计与稳定分析中的应用）
Keywords:
Artificial Intelligence （人工智能）
Data-Driven Methods （人工智能）
Stability Analysis （稳定性分析）
Multi-Source Measurement Fusion （多源量测融合）
Wideband Oscillation （宽频振荡）
Impedance Identification （阻抗辨识）
Digital Twin （数字孪生）
Submission Deadline: September 20, 2026