基于径向基神经网络风电场风速预测.PDFVIP

2010 The Second China Energy Scientist Forum Wind Speed Prediction Based on RBF Neural Network Shoudao Huang, Lang Dai, Keyuan Huang, Sheng Ye College of electrical and information engineering, Hunan university, Changsha, China, 410082 Email :shoudaohuang@, langzi3025@163.com Abstract: Wind speed forecasting is very important to wind farms and power system operation. By use of radial basis function (RBF) neural network the short-term wind speed forecasting is researched. On the premise of without taking into account numerical weather prediction data, with wind speed sequences is used as the input variable, the model was constructed based on the nonlinear approach ability of radial basis function neural networks that was used for the short-term wind speed prediction. Using the model, we have achieved 1h ahead forecasting of the wind speed and its prediction error analysis. The results show that neural network structure and the selection of input sample have a certain impact on the prediction results. The forecasting accuracy of the method is better than BP network. Keywords: short-term wind speed forecasting; artificial neural networks; radial basis function; wind farm 基于径向基神经网络的风电场风速预测 黄守道，戴 浪，黄科元，叶 盛 湖南大学电气与信息工程学院，长沙,中国，410082 Email:shoudaohuang@, langzi3025@163.com 摘 要：风速预测对风电场和电力系统的运行都具有重要意义。利用径向基函数（RBF ）神经网络研 究短期风速预测。在不考虑数值天气预报数据的前提下，以风速序列作为模型输入变量，建立了基于 径向基函数神经网络的非线性逼近功能的风速短期预测模型。利用该模型实现了对风速的提前 1h 预 测，并对其预测误差进行分析。研究结果表明，神经网络的结构和输入样本选择对预测结果有一定的 影响，和BP 网络比较，该方法具有更好的预测精度。 关键词：短期风速预测；人工神经网络；径向基函数；风电场 数学模型不同可分为持续预测法、自回归滑动平均模 1 引言