基于灰色关联分析的短期风速预测方法

针对建立短期风速预测统计模型时输入变量的类型难以确定的问题,提出了一种基于灰色关联分析的短期风速预测方法。该方法以风速为基准序列,对温度、压强、相对湿度、露点等气象因素进行灰色关联分析,按照关联系数大小对气象因素进行排序,并根据排序结果选择风速和关联系数较大的气象因素作为输入变量构建LSTM模型,最后通过模型计算出预测结果。基于实测数据对该方法的有效性进行验证,结果表明,所提出的方法具有较高的预测精度。

Short-term Wind Speed Prediction Method Based on Grey Relation Analysis

In view of the problem of the type of input variables are difficult to determine when establishing a short-term wind speed prediction statistical model,a wind speed prediction method is proposed for wind farms based on grey relation analysis.The wind speed is used as the reference sequence and performs grey correlation analysis on other meteorological factors such as temperature,pressure,relative humidity and dew point.Then the meteorological factors are sorted according to the correlation coefficient,wind speed and the meteorological factors with large correlation coefficients are selected as input variables according to the sorted results.The LSTM model is built based upon.Finally the prediction results are calculated by the model.The effectiveness of the proposed method is verified on the basis of the actual data of a wind farm.The results show that the prediction accuracy of the proposed method is higher than that of a traditional method.