摘要
为改善系统电能质量和电压稳定性,文中提出了一种结合深度神经网络和粒子群优化算法的无功电压优化方法。该方法通过采用图卷积神经网络对电力系统运行状态进行特征提取,辨识出对无功电压优化最有效的控制节点。在压缩控制节点后采用粒子群优化法对无功电压优化模型进行求解,避免了传统算法对求解初值要求高、计算耗时长的问题。以某地电网数据为算例进行了实验验证,不同方法求解所得到的系统无功功率分布和节点电压分布情况表明,所提方法在优化系统无功功率、改善节点电压方面相比于其他方法均有一定提高,系统线路平均负载率仅为0.67%。
In order to improve the power quality and voltage stability of the system,a reactive power and voltage optimization method combining deep neural network and Particle Swarm Optimization algorithm is proposed.This method identifies the most effective control node for reactive power and voltage optimization by feature extraction of power system operation state using graph Convolutional neural network.After compressing the control node,the Particle swarm optimization method is used to solve the reactive power and voltage optimization model,which avoids the problem of traditional algorithms requiring high initial value and long calculation time.The experimental verification was conducted using a certain power grid data as an example.The distribution of reactive power and node voltage obtained by different methods showed that the proposed method has a certain improvement compared to other methods in optimizing system reactive power and improving node voltage.The average load rate of the system line is only 0.67%.
作者
牛浩明
鲁怡兰
张立清
王维洲
NIU Haoming;LU Yian;ZHANG Liqing;WANG Weizhou(Electric Power Research Institute of State Grid Gansu Electric Power Company,Lanzhou 730030,China;School of Mechanical Electronic and Information Engineering,China University of Mining and Technology-Beijing,Beijing 100083,China;College of Information and Electrical Engineering,China Agricultural University,Beijing 100083,China;State Grid Gansu Electric Power Company,Lanzhou 730030,China)
出处
《电子设计工程》
2025年第3期115-119,共5页
Electronic Design Engineering
基金
国家电网公司科技项目(5100-202333003A-1-1-ZN)。
关键词
粒子群算法
新能源
电压优化
电网规划
电能质量
Particle Swarm Optimization algorithm
new energy
voltage optimization
power grid planning
power quality
