基于模糊神经网络的风速预测研究  

Research on wind speed prediction based on fuzzy neural network

作　　者：郭华旺[1] 董海鹰[1,2] 

机构地区：[1]兰州交通大学自动化与电气工程学院,甘肃兰州730070 [2]兰州交通大学光电技术与智能控制教育部重点实验室,甘肃兰州730070

出　　处：《电气传动自动化》2012年第3期1-5,13,共6页Electric Drive Automation

基　　金：长江学者与创新团队发展计划资助(IRT0629);甘肃省科技重大专项资助(0801GKDA058)

摘　　要：针对风速的不确定性给保障风力可靠性发电带来很大困难的问题,采用模糊神经网络的方法对大型风电场的风速进行预测,利用T-S模糊系统和神经网络的知识构建模糊神经网络预测模型,将风电场风电机组附近的气温、气压、空气湿度和风向等环境参数与风速预测模型的输入,对提前4小时和提前一天的风速进行预测,仿真结果表明该方法具有很高的精度。Aiming at the uncertainty of wind speed makes troubles in ensuring the reliability of wind power generation, the fuzzy neural network methods are adopted to predict the wind speed on large wind farms. To use the knowledge of T-S fuzzy systems and neural network to construct the fuzzy neural network models. To take other parameters such as temperature, atmospheric pressure, air humidity and other environmental aspects as inputs to predict the level of wind in advanced in 4 hours and one day. The simulation results show that this method has a high precision.

关 键 词：风速 预测 模糊神经网络 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]