全球能源互联网2025，Vol.8Issue(5)：594-601,8.DOI:10.19705/j.cnki.issn2096-5125.20240451
基于改进多层感知机的电力系统振荡稳定域构建
Constructing Oscillatory Stability Region Based on the Improved Multi-layer Perceptron
摘要
Abstract
Oscillations generated by dynamic interactions between power electronics and grid are becoming one of the critical issues in renewable power systems.It is important to evaluate oscillation risk in the practical power systems by the stability region of oscillations.Conventional methods to construct the oscillatory stability region include the pointwise method,predictor-corrector method and the random forest classifier,etc.However,the existing methods have their shortages in computational efficiency or accuracy.To fill this gap,this paper proposes a novel approach to construct the oscillatory stability region by multi-layer perceptron regression.First,the oscillation stability region is defined.Then,the Monte Carlo method is used to obtain operating points in the parameter space,and the stability dataset of the power system with different parameters is acquired through the impedance method.Finally,the oscillation stability region is calculated using an improved multi-layer perceptron model.The effectiveness of the proposed method is validated through its application in a real renewable energy grid-connected system in Ningxia power grid.关键词
振荡稳定域/多层感知机/新能源并网系统Key words
oscillatory stability region/multi-layer perceptron/renewable power system分类
信息技术与安全科学引用本文复制引用
任勇,李旭涛,汪林光,谢小荣..基于改进多层感知机的电力系统振荡稳定域构建[J].全球能源互联网,2025,8(5):594-601,8.基金项目
国网宁夏电力有限公司科技项目(5229DK230001). Science and Technology Project of State Grid Ningxia Electric Power Co.,Ltd.(5229DK230001). （5229DK230001）
