含高渗透率分布式电源的节点负荷预测方法  

Node load forecasting method of high permeability distributed power supply

作　　者：林梅芬[1,2] 潘文霞 LIN Meifen;PAN Wenxia(Department of Electrical Engineering, Fujian Water Conservancy and Power Professional Technical College, Yong′an 366000, China;College of Energy and Electrical Engineering, Hohai University, Nanjing 211100, China)

机构地区：[1]福建水利电力职业技术学院电力系,福建永安366000 [2]河海大学能源与电气学院,江苏南京211100

出　　处：《河南工程学院学报（自然科学版）》2020年第3期39-44,共6页Journal of Henan University of Engineering：Natural Science Edition

基　　金：国家自然科学基金资助项目(51707056);2018年福建省中青年教师教育科研项目(职业院校专项)(JZ180466)。

摘　　要：随着配电网中分布式电源比例的提高,其对节点负荷预测的影响变大,传统的负荷预测方法已难以满足预测精度要求。因此,针对高渗透率分布式电源的负荷预测问题,提出了一种基于小波神经网络和Elman神经网络的负荷预测方法。该方法具有较强的学习能力,避免了结构设计上的盲目性,具有结构简单、收敛速度快、精度高等特点。建立考虑天气类型指数的短期负荷预测模型,并应用小波神经网络和Elman神经网络相结合的预测模型进行训练,将训练结果与实际发电负荷进行对比可知,该模型有较强的预测能力,可预测光伏实际出力情况。根据各节点的实际用电负荷及由接入的分布式光伏所提供的用电负荷,可得出具有高渗透率分布式电源的节点预测负荷。最后,以实际地区的光伏发电数据和负荷数据算例验证了该方法的有效性。With the increase of the distributed power ratio in the distribution network,the influence of the distributed power supply on the load forecasting of the nodes is increasing,while the traditional load forecasting method is difficult to meet the prediction accuracy requirements.Aiming at the load forecasting problem of high permeability distributed power supply,a load forecasting method based on wavelet neural network and Elman neural network is proposed.The method has strong learning ability and avoids the blindness of structural design,and the structure is simple with high convergence speed and high precision.By establishing the short-term load forecasting model considering weather type index,using the wavelet neural network and Elman neural network to train the prediction model,then comparing the training results with the actual power load,the model can obtain the accurate prediction ability and predict the actual output of the photo voltaic(PV).Based on the actual electricity load of each node and the electricity load provided by the distributed(PV),the node forecast load with high permeability distributed power supply is obtained.Finally,an example is given to show the effectiveness of the proposed method.

关 键 词：高渗透率 分布式光伏 负荷预测 神经网络 

分 类 号：TM761[电气工程—电力系统及自动化]