说明：  针对传统光伏功率预测特征提取不足导致预测精度不高的问题，提出一种双通道网络 模型进行光伏功率预测。首先将光伏功率历史数据进行归一化处理，再将数据送入两个并行的卷积神经网络( Convolutional Neural Network，CNN) 进行特征提取，经融合层融合送入长短期记忆网络( Long Short-Term Memory，LSTM) 进行光伏功率预测。采用地中海气候光伏发电数据集进行测试，结果表明所提出的方法与单通道网络相比平均绝对误差( Mean-Absolute Error，MAE) 减小了 12. 3%，均方根误差( Ｒoot-Mean-Square Error，ＲMSE) 减小了 3%，实现了更高的预测精度。
(Aiming at the problem that the traditional PV power prediction feature is insufficiently extracted and the prediction accuracy is not high, a dual-channel network model is proposed for PV power prediction. First, the photovoltaic power historical data is normalized, and then the data is sent to two parallel convolutional neural networks for feature extraction, and the fusion layer is fed to the long-term and short-term memory network for photovoltaic power prediction. Using the Mediterranean climate photovoltaic power generation data set for testing, the results show that the proposed method reduces the average absolute error by 12.3% and the root mean square error by 3% compared with a single channel network, achieving higher prediction accuracy .)