摘要
风电场模型对风电场和电力系统的运行都具有重要意义。为克服机理建模方法的不足,在论述神经网络建模原理基础上,采用误差反向传播(BP)网络拟合风电场的静态特性模型;在神经网络静态模型的基础上,进一步建立动态实时仿真模型。通过数据样本进行预处理、训练和测试,使得网络模型达到精度要求,确定每层的权值,从而能很好地拟合建模对象的性能。用采集到的另一些数据进行验证,且与实际结果进行比较,以验证智能建模方法的可行性和优越性。
Wind farm model has a great significance for wind farm and power system operation.In order to overcome the deficiencies of mechanism modeling method,static characteristics were approached by using error back propagation（BP） network model and the discussion results of the neural network modeling principle.And then based on the neural network static model,dynamic real-time simulation model was developed.The data samples were pretreated,trained and tested,which made the network model precise and the weight of each floor determined,and so the simulating properties of the object model could be improved.The feasibility and superiority of intelligent modeling were verified by the collecting data and their comparison with the actual results.
出处
《中国电力》
CSCD
北大核心
2010年第9期79-82,共4页
Electric Power
基金
国家自然科学基金资助项目(50877053)
天津市自然科学基金资助项目(09JCYBJC07100)
关键词
风电场
神经网络建模
误差方向传播网络
wind farm
neural network modeling
error back propagation（BP） network
