基于神经网络的风功率预测问题研究  

Research on Wind Power Forecasting Problem Based on Neural Network

作　　者：张涛[1] 张新燕[1] 王维庆[1] 

机构地区：[1]新疆大学电气工程学院,新疆乌鲁木齐830047

出　　处：《四川电力技术》2013年第1期16-18,47,共4页Sichuan Electric Power Technology

基　　金：新疆大学博士基金项目(BS100122);国家自然基金项目(50867004);国家"十一五"863项目(2009AA05Z445)

摘　　要：为了更准确地预报风力发电机出力,给出了一种基于神经网络的风功率预测方法。介绍了风功率预测的现状和相关标准,基于一般预测方法准确率、合格率低等因素的考虑,尝试采用BP神经网络对其进行优化。计算数据和仿真结果表明优化后的预测数据准确率、合格率更高,对风电场出力预测有一定的帮助。In order to forecast the output power of wind turbines more accurately, a wind power forecasting method based on the neural network is presented. The present situation of wind power forecasting and the relevant standards are introduced. Based on the consideration for the low accuracy and qualification rate of the general forecasting methods, the BP neural network is a-dopted to optimize it. The calculation data and simulation results show that the prediction data after the optimization are more accurate and qualified, and it is helpful for the output prediction of wind power.

关 键 词：风力发电 功率预测 神经网络 优化 MATLAB 仿真 

分 类 号：TM714[电气工程—电力系统及自动化]