摘要
智慧电网中的需求响应管理是一项最具挑战性的任务,因为每户家庭的负荷状况是不断变化的。现有的管理方法未能观察到这些用户的负载配置文件中隐藏的模式。因此,文中提出UDR算法来实现需求响应动态管理。该算法通过收集智慧住宅的用电数据,利用卷积神经网络(CNN)学习数据中的隐藏模式并输出不同的负载曲线。然后结合向量回归(SVR)模型,预测所有介入智慧电网的智慧住宅的总负荷消耗,将预测结果与电网的发电量进行比较,从而管理所连接智慧住宅的需求响应(DR)。
Demand response management in smart grids is one of the most challenging tasks because the load status of each household is constantly changing.Existing management methods fail to observe patterns hidden in the load profiles of these users.Therefore,a UDR algorithm to achieve dynamic management of demand response is proposed.The power consumption data of the smart house is collected,the convolutional neural network(CNN)is used to learn the hidden mode in the data and outputs different load curves.Then combined with the vector regression(SVR)model,predict the total load consumption of all smart homes involved in the smart grid.The predicted results are then compared to the amount of electricity generated by the grid to manage the demand response(DR)of the connected smart home.
作者
于宏文
朱海明
汤卫东
刘涛
徐遐龄
于文娟
肖大军
YU Hong-wen;ZHU Hai-ming;TANG Wei-dong;LIU Tao;XU Xia-ling;YU Wen-juan;XIAO Da-jun(Central China Branch of State Grid Co.,Ltd.,Wuhan 430077,China;NARI Group Corporation,Nanjing 211106,China;NARI Technology Co.,Ltd.,Nanjing 211106,China)
出处
《信息技术》
2020年第3期68-73,共6页
Information Technology
关键词
智慧电网
卷积神经网络
需求响应
智慧住宅
smart grid
convolution neural network
demand response
smart home
