短期光伏功率高精度组合预测模型  

High-precision Combination Prediction Model for Short-term PV Power

作　　者：万谨瑜 宁勇 孙静[1] 赵振兴 龚军辉[1] 龚希文 WAN Jinyu;NING Yong;SUN Jing;ZHAO Zhenxing;GONG Junhui;GONG Xiwen(College of Electrical and Information,Engineering Hunan Institute of Engineering,Xiangtan 411104,China)

机构地区：[1]湖南工程学院电气与信息工程学院,湘潭411104

出　　处：《湖南工程学院学报(自然科学版)》2024年第4期1-8,共8页Journal of Hunan Institute of Engineering(Natural Science Edition)

基　　金：湖南省教育厅科研创新平台开放基金项目(20k036).

摘　　要：为了提高光伏功率预测的精确性,提出一种基于变分模态分解(variational mode decomposition,VMD)-蜣螂优化算法(dung beetle optimizer,DBO)-核极限学习机(kernel based extreme learning machine,KELM)的组合光伏功率预测模型.该模型对光伏发电影响因素进行分析筛选,选出与光伏输出功率高度相关的因素作为输入,并采用变分模态分解(VMD)将光伏原始功率数据分解为多个特征模态函数(intrinsic mode function,IMF).然后,将分解得到的IMF分量分别输入KELM光伏功率预测模型,同时通过DBO优化算法对KELM初始输入权重进行优化,从而提高核极限学习机的泛化能力.最后,将各IMF分量预测结果叠加求和得到最终预测结果.仿真结果表明,该组合光伏功率预测模型相较于KELM、VMD-KELM模型具有更好的预测精度.In order to improve the accuracy of photovoltaic power prediction,this paper proposes a combined photovoltaic power prediction model based on variational mode decomposition(VMD)-dung beetle optimizer(DBO)-kernel based extreme learning machine(KELM).This algorithm analyzes and screens the influencing factors of photovoltaic power generation,selects factors highly related to photovoltaic output power as inputs,and uses variational mode decomposition(VMD)to decompose the original photovoltaic power data into multiple intrinsic mode functions(IMF).Then,the decomposed IMF components are separately input into the KELM photovoltaic power prediction model,and the initial input weights of the KELM are optimized by using the DBO optimization algorithm to improve the generalization ability of the kernel limit learning machine.Finally,the predicted results of each IMF component are superimposed and summed to obtain the final prediction result.The simulation results show that the photovoltaic power prediction model proposed in this paper has better prediction accuracy compared to KELM and VMD-KELM models,proving the effectiveness of the proposed method.

关 键 词：光伏功率预测 蜣螂优化算法 变分模态分解 核极限学习机 

分 类 号：TM615[电气工程—电力系统及自动化]