摘要
针对传统评估方法难以有效评估新型电力系统暂态电压稳定性的问题,提出了基于改进深度信念网络(deep belief network,DBN)的暂态电压稳定评估模型。对新能源的不确定性进行建模,并通过最小绝对收缩和选择算子(least absolute shrinkage and selection operator,LASSO)的回归实现特征初筛获得了有效样本;改进了DBN中的受限玻尔兹曼机(restricted boltzmann machine,RBM),通过保留电网结构连接关系的方法来提高DBN模型的分组能力;采用局部线性解释器(local linear interpreter,LLI)分析失稳样本,解释系统特征与评估结果的关系。分析结果表明,所提方案能够有效提升风电并网系统暂态电压稳定性判定的准确性,从而验证了所提方法的有效性。
Aiming at the problem that traditional evaluation methods are difficult to effectively assess transient voltage stability under the new power system,this paper proposes a transient voltage stability evaluation model based on an improved deep belief network(DBN).The uncertainty of new energy sources is modeled,and effective samples are obtained through feature preliminary screening using least absolute shrinkage and selection operator(LASSO)regression.The restricted Boltzmann machine(RBM)in DBN is improved to enhance the grouping capability of the DBN model by retaining the connection relationships of the power grid structure.The local linear interpreter(LLI)is used to analyze unstable samples and explain the relationship between system features and evaluation results.Analysis results show that the proposed scheme can effectively improve the accuracy of transient voltage stability judgment for wind power grid-connected systems,thus verifying the effectiveness of the proposed method.
作者
刘颂凯
曹俊
吴宇恒
苏攀
李彦彰
艾宇坤
黄平川
LIU Songkai;CAO Jun;WU Yuheng;SU Pan;LI Yanzhang;AI Yukun;HUANG Pingchuan(School of Electrical Engineering and New Energy,China Three Gorges University,Yichang 443002,Hubei,China;Hubei Collaborative Innovation Center for New Energy Microgrid,Yichang 443002,Hubei,China;Wuhan Power Supply Company of State Grid Hubei Electric Power Co.,Ltd.,Wuhan 430010,Hubei,China)
出处
《电网与清洁能源》
北大核心
2025年第8期55-65,共11页
Power System and Clean Energy
基金
国家自然科学基金项目(52407118)。
关键词
风电并网系统
样本生成
深度信念网络
线性解释器
暂态电压稳定
wind-integrated power system
sample generation
deep belief network
linear interpreter
transient voltage stability
作者简介
刘颂凯(1990一),男,博士,副教授,研究方向为电力系统运行优化与控制;曹俊(2002-),男,硕士研究生,研究方向为电力系统运行优化与控制;吴宇恒(2001一),男,硕士研究生,研究方向为电力系统运行优化与控制;苏攀(1985一),男,硕士,助教,研究方向为输电线路工程;李彦彰(1989-),男,硕士研究生,研究方向为电力设备在线监测及故障诊断;艾宇坤(2001一),男,硕士研究生,研究方向为电力系统运行优化与控制;黄平川(2000一),男,硕士研究生,研究方向为电力系统运行优化与控制。
