基于模糊聚类法的光伏配电网负荷过载预测  

Load overload prediction of photovoltaic distribution network based on fuzzy clustering method

作　　者：许长清 李科 孙义豪 皇甫霄文 马杰 XU Changqing;LI Ke;SUN Yihao;HUANGFU Xiaowen;MA Jie(State Grid Henan Electric Power Company Economic and Technological Research Institute,Zhengzhou 450052,China)

机构地区：[1]国网河南省电力公司经济技术研究院,河南郑州450052

出　　处：《电子设计工程》2024年第6期77-80,85,共5页Electronic Design Engineering

基　　金：国家电网有限公司总部管理科技项目资助(5100-202256019A-1-1-ZN)。

摘　　要：配电网负荷过载对运行安全性造成消极影响,且光伏配电网负荷过载预测时,受天气影响,导致预测误差较大。为此,基于模糊聚类法提出一种新的光伏配电网负荷过载预测方法。分析光伏配电网因素,通过模糊聚类实现信息处理,确定聚类中心,分析样本相似度,根据分析结果完成分类识别。采用路径分析方法分析影响因素指标权重,计算权重系数。利用替代法通过构造新特征加快拟合速度,通过BP神经网络优化参数,完成配电网负荷过载预测。实验结果表明,所提方法在恶劣环境下预测误差低于5%,具有较强的抗干扰能力。The overload of the distribution network will have a negative impact on the operation safety,and when the overload of the photovoltaic distribution network is predicted,it will be affected by the weather,resulting in a large prediction error.Therefore,a new load overload prediction method for photovoltaic distribution network is proposed based on fuzzy clustering method.Analyze the factors of photovoltaic distribution network,realize information processing through fuzzy clustering,determine the cluster center,analyze the similarity of samples,and complete the classification and identification according to the analysis results.The path analysis method is used to analyze the weights of the influencing factors and calculate the weight coefficients.The substitution method is used to speed up the fitting speed by constructing new features,and the parameters are optimized by the BP neural network to complete the load overload prediction of the distribution network.The experimental results show that the prediction error of the proposed method is less than 5%in harsh environments,and it has strong anti⁃interference ability.

关 键 词：模糊聚类法 光伏配电网 负荷过载 过载预测 

分 类 号：TM715[电气工程—电力系统及自动化] TN01[电子电信—物理电子学]