Photovoltaic power prediction method and system based on feature migration

The invention provides a photovoltaic power prediction method and system based on feature migration, and the method comprises the steps: obtaining the historical power generation data of a photovoltaic power station, removing an abnormal value, and taking the data as a target value of model training; obtaining a training data set through the source domain training sample data and the target domain training sample data based on a migration component analysis method; and respectively training an LSTM model and a Lasso model to predict photovoltaic power, and carrying out photovoltaic power combined prediction on a prediction result by using a fixed weight coefficient to obtain a final prediction result. According to the method, the model prediction precision is improved, and the accuracy and stability of photovoltaic power generation short-term power prediction are improved. 本发明提出一种基于特征迁移的光伏功率预测方法及系统，获取光伏电站历史发电数据，去除异常值，作为模型训练的目标值；基于迁移成分分析法通过源域训练样本数据和目标域训练样本数据得到训练数据集；分别训练LSTM和Lasso模型进行光伏功率的预测，并将预测结果以固定权重系数进行光伏功率