摘要
以风光为代表的新能源发电功率准确预测是高比例新能源并网消纳的基础,为此提出一种变权重组合的风光混合预测模型,可实现风光发电功率的同时预测。首先考虑风光发电功率耦合相关性,分析风电场和光伏发电场站的关联特性,利用支持向量机、遗传算法优化的BP神经网络和径向基神经网络,得到风光发电功率的初步预测值,进一步采用方差-协方差权值动态分配法组合单一预测算法预测初值,构建基于变权重组合的风光发电功率混合预测模型,并以新疆某地区为案例进行分析。研究结果表明:变权重组合的混合预测模型优于单一预测算法和其它预测模型,组合模型的3个评价指标均优于单一预测算法,能够对风光发电功率做出有效的预测,验证了本文所提模型的有效性和优越性。
Accurate prediction of wind and solar power generation is crucial for the efficient integration and consumption of a high proportion of new energy in the grid.Therefore,a wind-photovoltaic hybrid prediction model based on variable weight combination is proposed.Firstly,considering the power coupling correlation between wind and solar power generation,the characteristics of wind farms and photovoltaic power stations are analyzed.Initial predicted values of wind and solar power generation are obtained using the support vector machine,BP neural network optimized by genetic algorithm and radial basis function neural network.Furthermore,the dynamic variance-covariance weight distribution method is applied to combine the initial predictions from individual algorithms,constructing a wind-photovoltaic hybrid prediction model based on variable weight combination.The model is applied to a case study in a region of Xinjiang.The results indicate that the hybrid prediction model with variable weight combination outperforms individual prediction algorithms and other prediction models.All three-evaluation metrics of the hybrid model are superior to those individual algorithms,effectively predicting wind and solar power generation and verifying the effectiveness and superiority of the proposed model.
作者
何玉灵
焦凌钰
孙凯
解奎
杜晓东
王海朋
张祥宇
HE Yuling;JIAO Lingyu;SUN Kai;XIE Kui;DU Xiaodong;WANG Haipeng;ZHANG Xiangyu(School of Mechanical Engineering,North China Electric Power University,Baoding 071003,China;Hebei Engineering Research Center for Advanced Manufacturing&Intelligent Operation and Maintenance of Electric Power Machinery,Baoding 071003,China;Electric Power Research Institute of State Grid Hebei Electric Power Co.,Ltd.,Shijiazhuang 050022,China;School of Electrical Engineering,North China Electric Power University,Baoding 071003,China)
出处
《华北电力大学学报(自然科学版)》
北大核心
2025年第6期49-59,共11页
Journal of North China Electric Power University(Natural Science Edition)
基金
国家自然科学基金面上项目(52177042)
河北省高等学校科学技术研究重点项目(ZD2022162)
河北省重点研发计划专项(21312102D)
中央高校基本科研业务费面上项目(2022MS095)。
关键词
风光混合预测
变权重组合预测模型
支持向量机
BP神经网络
径向基神经网络
短期预测
wind-photovoltaic hybrid prediction
variable weight combination prediction model
support vector machine
BP neural network
RBF neural network
short-term prediction
