基于小波分解与极限学习机的光伏电站辐照度短期预测研究

Short-term Prediction of Photovoltaic Power Plants Radiation based on Decomposition and Extreme Learning Machine

文章就传统光伏发电站太阳能辐照强度预测精度不理想的问题,给出基于相对关联数据输入的WD-ELM太阳能辐照强度预测方法,提出对日相似度与前趋势相似度进行多种目标优化来选取相似日集的改进相似日算法,并组织建立了广义回归神经网络功率预测模型,提前一天预测分辨率为15 min的太阳能辐照强度。预测成果表明,此方法相对传统太阳能光伏发电站的辐照度预测更为精确。

For prediction accuracy of solar radiation intensity is not ideal question to traditional photovohaic power plants, the prediction method of WD-ELM solar radiation intensity based on the relative associated data input is proposed. An improved similar day algorithm was proposed, taking the multi- objective optimization on the day-similarity and the previous-trend-similarity to select the similar day set. And then, the GRNN power forecasting model was created, solar radiation intensity was forecasted with 15min＇ s resolution 1 day ahead finally. The prediction result shows that the GRNN model can obtain a higher forecasting accuracy than the traditional method.