煤气与热力2025，Vol.45Issue(1)：26-31,6.
太阳辐照度机器学习模型预测能力分析
Analysis of Predictive Ability of Machine Learning Models for Solar Irradiance
摘要
Abstract
Weifang was selected as the study ar-ea,and the solar irradiance prediction models were es-tablished by using support vector machine(SVR),El-man neural network,long short-term memory network(LSTM),bidirectional long short-term memory network(Bi-LSTM),random forest(RF)and BP neural net-work.The prediction effect of the six models was eval-uated,and the importance of input variables was ana-lyzed.Elman neural network,RF,and BP neural net-work models can be used for solar irradiance predic-tion.Among the input variables,time has the greatest impact on the prediction effect of the model,followed by the number of days,cloud cover,and weather condi-tions.If any input variable is missing,the prediction effect of the model will be worse.When the input vari-ables are complete,the prediction of the model is opti-mal.关键词
太阳辐照度/机器学习/预测模型Key words
solar irradiance/machine learn-ing/predictive models分类
能源科技引用本文复制引用
安文含,刘建华,刘吉营..太阳辐照度机器学习模型预测能力分析[J].煤气与热力,2025,45(1):26-31,6.基金项目
山东省高等学校青创人才引育计划创新团队项目(鲁教科函[2021]51号) （鲁教科函[2021]51号）
山东省自然科学基金(ZR2021ME199) （ZR2021ME199）
