基于模糊粗糙集与改进聚类的神经网络风速预测

Neural Network Wind Speed Prediction Based on Fuzzy Rough Set and Improved Clustering

提高风电功率预测精度是保障风电场和电力系统安全稳定运行的有效手段。神经网络方法已在风电功率预测中得到了广泛应用,并取得了不错的效果,而网络的输入变量与训练样本对其预测性能有着重要影响。基于此,提出一种基于模糊粗糙集与改进聚类的神经网络风速预测方法。采用模糊粗糙集对影响风电场风速的多种因素进行了属性约简,得到优化了的模型输入及各属性对风速的重要性;采用基于属性重要性的加权欧氏距离对传统聚类进行改进,建立了各聚类预测模型,并提取相似性较高的数据作为训练样本训练各类预测模型,对训练样本实现了优选;根据当前属性值选择匹配的模型对风速进行预测。以华北地区某风电场实际数据为例进行了实验,结果表明该方法能在较少的模型输入下有效地提高预测精度。

Improving wind power prediction accuracy is an effective means to ensure the safe and stable operation of wind farm and power system. Neural network methods have been widely applied to wind power prediction with satisfactory results, but the training set and the input variables affect its forecasting performance greatly. Based on that, an integrated neural networks approach combining fuzzy rough set and improved clustering was proposed in this paper. Fuzzy rough set was applied to carry on the attribute reduction for a variety of factors affecting wind speed to optimize the model input, and the importance of each attribute for wind speed was obtained. The traditional clustering was improved through the weighted Euclidean distance based on attributes＇ importance, and similar data were extracted as the training set to optimize the training set. Matching model was selected to carry out the wind speed prediction. Taking a wind farm in north China as an example, the experimental results show that the method can effectively improve the forecasting accuracy with less model input.