电力系统及其自动化学报2024，Vol.36Issue(5)：1-8,8.DOI:10.19635/j.cnki.csu-epsa.001322
基于SARIMAX-SVR的光伏发电功率预测
Forecasting of Photovoltaic Power Generation Based on SARIMAX-SVR
摘要
Abstract
To improve the prediction accuracy of photovoltaic(PV)power generation,a prediction method for PV pow-er generation based on seasonal autoregressive integrated moving average with exogenous factors(SARIMAX)com-bined with optimized support vector regression(SVR)is proposed in this paper.First,the correlation feature method is used to cluster the key meteorological factors in meteorological conditions,thereby eliminating the data redundancy and reducing the complexity in the autoregressive integrated moving average with exogenous factors(ARIMAX)model.Sec-ond,a seasonal factor is introduced into the ARIMAX model,and a SARIMAX model is constructed to capture season-al changes in the data.Finally,the fitting residual of the SARIMAX model is used as input to the SVR model to further fit the nonlinearity of the data.The analysis of simulation examples shows that the proposed method can significantly im-prove the prediction accuracy of PV power generation.关键词
光伏发电/功率预测/差分自回归移动平均/季节性因子/支持向量回归Key words
photovoltaic(PV)power generation/power forecasting/autoregressive integrated moving average(ARI-MA)/seasonal factor/support vector regression(SVR)分类
信息技术与安全科学引用本文复制引用
周鑫,李燕,曾永辉,石鹏程..基于SARIMAX-SVR的光伏发电功率预测[J].电力系统及其自动化学报,2024,36(5):1-8,8.基金项目
湖南省自然科学基金资助项目(2022JJ30266) （2022JJ30266）
