摘要
为了精准预测光伏发电输出功率,文章提出了一种基于多气象要素降维、优化后的变分模态分解(OVMD)技术、自适应t分布的麻雀搜索算法(t SSA)和最小二乘法向量机(LSSVM)的光伏发电输出功率预测模型。利用OVMD技术对输入光伏时间序列数据进行分解处理,引入t SSA对利用各模态分量建立的LSSVM模型进行参数寻优,搭建了基于OVMD-t SSA-LSSVM算法的光伏功率预测模型,并使用了中国东南沿海某地区3 a的气象数据和实时的光伏输出功率数据进行模型性能验证,通过与SVM,LSSVM,VMD-LSSVM和VMDSSA-LSSVM 4种模型的预测性能对比,OVMD-t SSA-LSSVM模型的预测精度和拟合效果均最优。实验数据表明,该模型的平均绝对百分比误差(MAPE)和均方根误差(RMSE)分别小于3%和0.35,决定系数(R-Square)超过了97%。最后,通过光伏气象要素降维处理,进一步提升了OVMD-t SSA-LSSVM模型性能。
In order to accurately predict the output power of photovoltaic power generation,this paper proposes a photovoltaic power generation output power prediction model based on multi meteorological element dimensionality reduction,variational modal decomposition(OVMD),adaptive t-distribution sparrow search algorithm(t SSA)and least square normal vector machine(LSSVM).In this paper,OVMD is used to decompose the input photovoltaic time series data,tSSA algorithm is introduced to optimize the parameters of LSSVM model established by using each modal component,and a photovoltaic power prediction model based on OVMD-t SSA-LSSVM algorithm is built.The performance of the model is verified by using 3-year meteorological data and real-time photovoltaic output power data of a certain area in the southeast coast of China,by comparing the prediction performance with SVM,LSSVM,VMD-LSSVM and VMD-SSA-LSSVM,it can be concluded that the prediction accuracy and fitting effect of OVMD-t SSA-LSSVM model are better than other models.The experimental data show that the average determination percentage error(MAPE)and root mean square error(RMSE)of the model are less than 3%and0.35 respectively,and the determination coefficient(R-square)exceeds 97%.Finally,the performance of OVMD-t SSA-LSSVM model is further improved through the dimension reduction of photovoltaic meteorological elements.
作者
杨凌升
李伟
Yang Lingsheng;Li Wei(School of Electronics and Information Engineering,Nanjing University of Information Science&Technology,Nanjing 210044,China)
出处
《可再生能源》
CAS
CSCD
北大核心
2022年第9期1157-1165,共9页
Renewable Energy Resources
基金
国家自然科学基金项目(41401572)
江苏省高校优势学科项目(PAPD)。
