分布式能源2026，Vol.11Issue(3)：75-82,8.DOI:10.16513/j.2096-2185.DE.25100347
一种基于改进CNN的短期日前新能源出力预测方法
An Improved CNN Approach for Short-Term Day-Ahead New Energy Output Prediction
摘要
Abstract
To address the issues of data noise interference,feature scale discrepancy,and insufficient multiscale meteorological pattern modeling in photovoltaic power forecasting,a prediction method based on dynamic data preprocessing and gated dense multiscale convolutional neural network(GDMS-CNN)is proposed.Firstly,an anomaly detection mechanism based on dynamic sliding-window Z-score is established,and missing values are processed via covariance-weighted multivariate interpolation.Secondly,an adaptive piecewise normalization algorithm is adopted to eliminate feature dimensional differences,and cloud-cover correction factor and atmospheric attenuation factor are constructed to enhance physical feature representation.Finally,a GDMS-CNN is designed,wherein the feature extraction efficiency is optimized by depthwise separable convolution modules,densely connected dilated convolution blocks are constructed to capture multiscale spatiotemporal correlation features,and an asymmetric gated channel attention mechanism is embedded to dynamically recalibrate feature weights.Experimental results demonstrate that the proposed method reduces the root mean square error(RMSE)by 16.4%compared with the optimal baseline model genetic algorithm-variational mode decomposition-echo state network(GA-VMD-ESN),and by 43.4%compared with the traditional random forest.The proposed method provides a novel solution for photovoltaic output forecasting and effectively enhances the reliability of power grid dispatching.关键词
光伏发电预测/动态数据预处理/门控密集多尺度卷积神经网络(GDMS-CNN)/异常值检测/自适应归一化/多变量插值Key words
photovoltaic power forecasting/dynamic data preprocessing/gated dense multiscale convolutional neural network(GDMS-CNN)/anomaly detection/adaptive normalization/multivariate interpolation分类
能源科技引用本文复制引用
王宣元,季震,孙巍,裴宇婷,孔帅皓,王泽森..一种基于改进CNN的短期日前新能源出力预测方法[J].分布式能源,2026,11(3):75-82,8.基金项目
This work is supported by Science and Technology Project of State Grid Jibei Electric Power Co.,Ltd.(No.52018K24000A). 国网冀北电力有限公司科技项目(52018K24000A) （No.52018K24000A）
