摘要
针对新型电力系统实时运行期间,各个新能源机组的一次调频能力难以定量评估问题,源于数据驱动理论,提出一种基于卷积神经网络(convolutional neural networks, CNN)和层次分析法的新能源机组调频能力综合评估方法。以新能源机组有功出力、调频持续时间以及机组容量为指标,建立评估指标体系;采用卷积神经网络技术合理预测不同指标的数值,并通过层次分析法确定各指标间的相对权重。在PSCAD/EMTDC仿真软件中搭建包含风电场和光伏电站的IEEE 3机9节点模型进行仿真验证,算例结果表明所提出指标可以在数值上定量反映出各个新能源机组的调频能力,也验证了文中所提评估方法的有效性。
Aiming at the problem that it was difficult to quantitatively evaluate the primary frequency regulation ability of each renewable energy unit during the real-time operation of the renewable power system, a comprehensive evaluation method for the frequency regulation ability of renewable energy units based on convolutional neural networks(CNN) and analytic hierarchy process was proposed from the data-driven theory. Taking the active power output, frequency regulation duration and the capacity of renewable energy units as indicators, the evaluation index system was established. CNN technology was used to reasonably predict the value of different indicators, and the relative weight of each indicator was also determined by analytic hierarchy process. In PSCAD/EMTDC simulation software, an IEEE 3-machine 9-node model including wind farm and photovoltaic power station was built. The example results showed that the proposed index could quantitatively reflect the frequency regulation ability of renewable energy units in numerical terms, which verified the effectiveness of the proposed evaluation method.
作者
王智伟
徐海超
郭相阳
马炯
褚云龙
陈前昌
卢治
WANG Zhiwei;XU Haichao;GUO Xiangyang;MA Jiong;CHU Yunlong;CHEN Qianchang;LU Zhi(Northwest Branch,State Grid Corporation of China,Xi'an 710048,Shaanxi,China;Nanjing Branch,China Electric Power Research Institute Co.,Ltd.,Nanjing 210003,Jiangsu,China;Key Laboratory of Power System Intelligent Dispatch and Control of Ministry of Education(Shandong University),Jinan 250061,Shandong,China)
出处
《山东大学学报（工学版）》
CAS
CSCD
北大核心
2022年第5期70-76,共7页
Journal of Shandong University（Engineering Science）
关键词
新能源电源
一次调频能力
数据驱动理论
评估指标体系
卷积神经网络
层次分析法
renewable energy
primary frequency regulation ability
data-driven theory
evaluation index system
convolutional neural networks
analytic hierarchy process
