摘要
风光发电的大量接入,将引起配电网规划与运行特征的根本性改变,因而研究配电网中风光发电的选址定容问题具有重要意义。先利用数据驱动构建基于核极限学习机的容量选择模型;再以总投资成本和网络损耗最小为目标函数,以电压稳定性为评价指标对容量配置结果进行评估;最后采用IEEE 33节点系统作为算例进行仿真,将结果分别与支持向量机、粒子群和遗传算法进行比较。结果表明该容量配置方案能够起到节约成本、降低网络损耗及提高网络电压水平的作用,并能够为新能源接入配电网的投资方案提供一定的参考。
The extensive integration of wind and solar power generation will fundamentally change the planning and operational characteristics of the distribution network,therefore,studying the site selection and capacity determination of wind and solar power generation in the distribution network is of great significance.Firstly,a capacity selection model based on kernel extreme learning machine is constructed using data-driven approach;Then,the capacity allocation results are evaluated with the objective function of minimizing total investment cost and network loss,and voltage stability as the evaluation index;Finally,the IEEE 33 node system was used as an example for simulation,and the results were compared with support vector machine,particle swarm optimization,and genetic algorithm.The results show that the capacity configuration scheme obtained through this method can achieve cost savings,reduce network losses,improve network voltage stability,and provide certain reference for investment scheme of new energy access to distribution networks.
作者
涂菁菁
赵鹏
邹伟东
TU Jingjing;ZHAO Peng;ZOU Weidong(China Energy Engineering Group Investment Co.,Ltd.,Beijing 100122,China;School of Automation,Beijing Institute of Technology,Beijing 100081,China)
出处
《电气应用》
2024年第10期32-38,共7页
Electrotechnical Application
基金
国家自然科学基金青年基金资助项目(61906015)。
关键词
配电网
数据驱动
核极限学习机
有功损耗
电压稳定
distribution network
data driven
kernel extreme learning machine
active power loss
voltage stability
