摘要
随着越来越多的新能源接入电网,电网的频率特性发生了显著变化,现有的低频减载整定策略难以适应现代高比例新能源电网安全稳定运行的要求。文章提出了一种考虑风火协同的电网低频减载在线整定策略。在考虑风电参与电网调频的条件下,分析了不同风火占比以及不同负荷水平下的系统频率动态特性,基于系统多运行方式下的离线低频减载优化计算,采用深度长短时记忆网络算法进行在线低频减载参数整定,并以区域电网为实际算例进行了计算分析。结果表明,在线电网低频减载参数整定策略对于多源联合调频的高比例新能源电网具有较好的适用性。
With the increasing penetration of new energy,the frequency characteristics of the power grid have changed greatly.The existing under-frequency load shedding strategy is difficult to meet the requirements of safe and stable operation of modern high-proportion new energy power grids.This paper proposes an online setting strategy of low-frequency load shedding considering wind-thermal coordination.The dynamic characteristics of system frequency under various windthermal ratios and various load levels are studied in light of the contribution of wind power to grid frequency regulation.Based on the optimization calculation of offline underfrequency load shedding under multiple operating modes of the system,the deep long short-term memory network algorithm is used to set the online underfrequency load shedding parameters,and the regional power grid is taken as an actual example for calculation and analysis.The results show that the online low frequency load shedding parameter setting strategy has good applicability to the high proportion of new energy grid with multi-source joint frequency regulation.
作者
胡姝博
马欣彤
付尧
冯悦新
王欢
Hu Shubo;Ma Xintong;Fu Yao;Feng Yuexin;Wang Huan(State Grid Liaoning Electric Power Co.,LTD.Electric Power Research Institute,Shenyang 110055,China;Institute of Electric Power,Shenyang Institute of Engineering,Shenyang 110036,China)
出处
《可再生能源》
CAS
CSCD
北大核心
2023年第11期1547-1553,共7页
Renewable Energy Resources
基金
国网辽宁省电力有限公司管理科技项目资助(2022YF-70)。
关键词
深度长短时神经网络
低频减载
优化整定
deep time-short neural network
low frequency load reduction
optimal setting