电力系统自动化2025，Vol.49Issue(10)：145-153,9.DOI:10.7500/AEPS20231224002
复杂模型下电力系统暂态稳定性量化分析的算例筛选
Case Filtering in Quantitative Analysis of Power System Transient Stability Under Complex Model
摘要
Abstract
To ensure the robustness of quantitative analysis of transient stability in power systems under complex factors such as large-scale renewable energy integration and hybrid transmission,and to improve the acceleration ratio of case filtering algorithms,artificial intelligence technology is used to integrate the quasi-quantity classification method that meets engineering application requirements under simple models with accurate quantitative analysis methods under complex models.First,mechanism feature indicators are extracted from trajectories derived via multi-step Taylor expansion to characterize the sensitivity to complex factors and multi-swing stability confidence.These indicators enable pre-evaluation of the impact of complex models and adaptive adjustment of classification thresholds.Then,a two-layer classification filtering is implemented by adaptively switching between numerical integration and multi-step Taylor expansion,approximating the effects of complex models while introducing adaptive variable step sizes to reduce computational burden.The effectiveness and robustness of the algorithm are evaluated in nearly 20 000 cases across four actual power systems in China.关键词
复杂模型/暂态稳定性/人工智能/算例筛选/敏感性分析/多步泰勒展开/自适应变步长Key words
complex model/transient stability/artificial intelligence/case filtering/sensitivity analysis/multi-step Taylor expansion/adaptive variable step size引用本文复制引用
彭慧敏,薛禹胜,刘庆龙,黄天罡,薛峰..复杂模型下电力系统暂态稳定性量化分析的算例筛选[J].电力系统自动化,2025,49(10):145-153,9.基金项目
国家电网有限公司科技项目:"双碳变革中电网主动支撑研究的应用准备"(5108-202218280A-2-50-XG) （5108-202218280A-2-50-XG）
已申请国家发明专利(申请号:202310335793.1). This work is supported by State Grid Corporation of China(No.5108-202218280A-2-50-XG). （申请号:202310335793.1）
