摘要
为满足风电运行、维护及调度管理需要,提高风电功率预测精度,提出了一种基于ARIMA与BP神经网络的组合风电功率预测方法。介绍了时间序列法与BP神经网络法的基本原理,采用了新的结合方式,综合考虑了风速、风向、以及风电场当地的物理限制,建立了预测模型。通过对某风电场的实测数据进行分析预测及对比,结果表明,该方法能有效提高风电功率预测精度,具有较好的实际应用价值。
In order to satisfy the wind power operation, maintenance and scheduling management, as well as improve the predictive accuracy, a hybrid wind power prediction method based on ARIMA （autoregressive integrated moving average） and BP （back propagation） neural network is proposed. First, the paper introduces the time-se- ries method and BP neural network. With a novel combination approach, the predictive model is established, which takes adeount of the wind speed, wind direction and the physical limitations of the wind plant. Compared with the real data of a wind plant, the result shows that the proposed method improves the accuracy of wind power prediction effectively, and it also has good practical value.
出处
《电力科学与工程》
2012年第12期50-55,共6页
Electric Power Science and Engineering
关键词
风功率
时间序列
神经网络
预测
wind power
time-series
neural network
prediction
