基于CNN-SVR城市日负荷预测机制  

City Daily Load Forecasting Mechanism Based on CNN-SVR

作　　者：张功勋 姚方 曹赟 Zhang Gongxun;Yao Fang;Cao Yun(Department of Electric Power Engineering,Shanxi University,Taiyuan Shanxi 030013,China)

机构地区：[1]山西大学电力工程系,山西太原030013

出　　处：《电气自动化》2022年第5期38-40,共3页Electrical Automation

摘　　要：为了解决发电公司在计划、运营和控制城市日负荷方面面临的经济和技术挑战,考虑温度、湿度和风速等因素,设计了一个基于卷积神经网络支持向量回归机的预测模型。最后使用东南亚某邦调度中心的历史数据对模型进行了程序模拟,并获得了不错的日负荷预测表现。结果表明,预测模型对于实践具有很强的借鉴意义。In order to solve the economic and technical challenges faced by power generation companies in planning,operating and controlling the city’s daily load,considering factors such as temperature,humidity and wind speed,a prediction model based on a convolutional neural network support vector regression machine was designed.The CNN-SVR model was simulated programmatically using historical data from a dispatch center in a state in Southeast Asia,and a good daily load forecasting performance was obtained.The final result shows that the prediction model has a strong reference value for practice.

关 键 词：卷积神经网络 短期负荷 支持向量回归机 负荷预测 调度 

分 类 号：TM714[电气工程—电力系统及自动化]