Multi-View Spatiotemporal Synchronous Graph Learning for Adaptive Online Short-Term Voltage Stability Assessment
Chao Deng, Lipeng Zhu, Chang Liu, Hefeng Zhai, Baoye Tian, Ye Zhang, Baorong Zhou, Zexiang Zhu
Hunan University China Southern Power Grid (China)
阅读操作
确认中在文库中上传 PDF 后可生成中文音频讲解。
摘要与影响
How to efficiently achieve accurate data-driven short-term voltage stability (STVS) assessment largely remains a challenge in the field of power system stability online monitoring. However, due to the lack of synchronous processing of spatial and temporal STVS dynamics and the limited ability of single learner-based feature extraction, the majority of existing data-driven methods may fail to comprehensively capture STVS features. Consequently, this could compromise the reliability of STVS assessment under time-varying operating scenarios. To address these issues, this paper proposes a novel multi-view spatiotemporal synchronous graph learning approach to adaptive online STVS assessment. First, a spatiotemporal synchronous graph learning (STSGL) module is introduced to enable comprehensive spatiotemporal synchronous learning of transient responsive trajectories. Then, to help better capture STVS features, an adaptive temporal graph is elaborately designed to automatically weight different time steps in the dynamic trajectories. Further, by integrating temporal trajectories with spatial topology and temporal graph representations, a multi-view parallel learning mechanism is developed to adaptively select the most discriminative STVS features from multiple independent STSGL modules, thereby enabling more reliable STVS assessment in various scenarios. Numerical test results on a realistic regional power grid in South China demonstrate the effectiveness of the proposed approach.
逐年被引趋势
暂无年度引用数据
关键指标
同类平均 = 1
同领域 · 同年份 · 同类型
Google Scholar 与 OpenAlex 的被引统计范围不同,数值存在差异属正常。
AI 辅助阅读
依据:摘要
可就本文提问;依据不足时会说明。
学术脉络
学科主题
工程Power System Optimization and Stability
Power Systems Fault Detection · Microgrid Control and Optimization