微型电脑应用2025，Vol.41Issue(4)：25-27,32,4.
基于Elman神经网络的短期光伏功率预测分析
Short-term Photovoltaic Power Prediction Based on Elman Neural Network
摘要
Abstract
In order to improve the efficiency of short-term photovoltaic power prediction in the next 24 hours,a short-term pho-tovoltaic power prediction method based on Elman neural network is designed.After the network structure and parameters are determined,the accuracy of the training results and the rationality of the algorithm are determined to achieve accurate predic-tion.The research findings indicate that cluster data exhibit high similarity,while inter-cluster data demonstrate distinct differ-ence characteristics.The optimized method is used to predict the power of the generation,and the relationship between the pre-dicted value and the measured value is established.The predicted data are close to the measured value,and the optimized clus-tering algorithm is used to obtain more accurate prediction results.After the optimization of the algorithm,the mean value of the prediction error decreases significantly,and the mean value of IMAE decreases by about 80％,so as to obtain more effective clustering results.The optimized clustering processing can effectively improve the short-term prediction accuracy.This re-search can be extended to other similar fields and has good practical value.关键词
短期光伏功率/Elman神经网络/预测/关联度/相似度Key words
short-term photovoltaic power/Elman neural network/prediction/correlation/similarity分类
信息技术与安全科学引用本文复制引用
郭韶昕,陈晓东,孙耀平,李峰..基于Elman神经网络的短期光伏功率预测分析[J].微型电脑应用,2025,41(4):25-27,32,4.基金项目
呼和浩特科技创新研发项目(201141D562) （201141D562）
