基于K-Adaboost数据挖掘的配电网负荷预测Load Forecasting of Distribution Network Based on K-Adaboost Data Mining
刘伟,张锐锋,彭道刚
LIU Wei,ZHANG Ruifeng,PENG Daogang
摘要(Abstract)：
气象因素是造成配电网负荷波动的主要原因,在利用斯皮尔曼相关系数分析气象因素与配电网负荷相关性的基础上,着重把握部分气象因素与配电网负荷的联系,针对这类气象因素对配电网负荷的影响,提出一种基于数据挖掘聚类分析和Adaboost的负荷预测方法。首先对历史负荷数据进行预处理,然后应用K均值聚类算法对待测点气象因素进行分析,选择与待测点同类气象因素的历史负荷作为训练样本,最后采用Adaboost算法建立配电网负荷预测模型。通过实例证明K-Adaboost预测模型比BP神经网络预测模型更加稳定并且更贴近实际负荷。
Meteorological factors are the main cause of load fluctuation in distribution network. On the basis of analyzing the correlation between meteorological factors and distribution network load with Spearman correlation coefficient, the relationship between some meteorological factors and distribution network load is emphatically grasped, in view of the influence of such weather factors on meteorological sensitive load, and a load forecasting method based on data mining clustering analysis and Adaboost is proposed. First, the historical load data is preprocessed, then the K means clustering algorithm is used to analyze the meteorological factors of the test points, and the historical load of the similar meteorological factors is selected as the training sample. Finally, the Adaboost algorithm is used to establish the distribution network load forecasting model.The example shows that the K-Adaboost prediction model is more stable and closer to the actual load than the BP neural network prediction model.
关键词(KeyWords)：
配电网负荷;负荷预测;K均值算法;Adaboost
distribution network load;load forecasting;K-means;Adaboost
基金项目(Foundation): 上海市“科技创新行动计划”社会发展领域项目(16DZ1202500)
作者(Author):
刘伟,张锐锋,彭道刚
LIU Wei,ZHANG Ruifeng,PENG Daogang
DOI: 10.19585/j.zjdl.201901018
参考文献(References)：
- [1]高志远,姚建国,曹阳,等.智能电网发展机理研究初探[J].电力系统保护与控制,2014,42(5):116-121.
- [2]万志宏,陈亮,文福拴.计及温度影响的短期负荷预测时间序列模型[J].华北电力大学学报(自然科学版),2011,38(3):61-66.
- [3]万昆,柳瑞禹.区间时间序列向量自回归模型在短期电力负荷预测中的应用[J].电网技术,2012,36(11):77-81.
- [4]黄宇腾,韩翊,赖尚栋.深度神经网络在配电网公变短期负荷预测中的应用研究[J].浙江电力,2018,37(5):1-5.
- [5]曹松钱.提高配电网短期负荷预测准确率的研究[J].浙江电力,2014,33(11):9-14.
- [6]吴亚雄,谢敏.基于BP神经网络灰色回归组合模型的年最大负荷预测[J].南方能源建设,2017,4(2):46-50.
- [7]王小君,毕圣,徐云鹍,等.基于数据挖掘技术和支持向量机的短期负荷预测[J].电测与仪表,2016,53(10):62-67.
- [8]ZENG X Y,SHU L,HUANG G M,et al.Triangular fuzzy series forecasting based on grey model and neural network[J].Applied Mathematical Modelling,2016,40(3):1717-1727.
- [9]王奔,冷北雪,张喜海,等.支持向量机在短期负荷预测中的应用概况[J].电力系统及其自动化学报,2011,23(4):115-121.
- [10]李晓峰,黄国兴,高巍巍,等.数据挖掘和动态神经网络在负荷预测中的研究与应用[J].计算机科学,2014,41(4):219-222.
- [11]柳进,于继来,唐降龙.基于数据挖掘的电网高峰负荷预测系统[J].计算机工程,2005(1):9-11.
- [12]曹莹,苗启广,刘家辰,等.AdaBoost算法研究进展与展望[J].自动化学报,2013,39(6):745-758.
- [13]张良均,杨坦,肖刚,等.MATLAB数据分析与挖掘实战[M].北京:机械工业出版社,2015.
- [14]TAN P N,STEINBACH M,KUMAR V.数据挖掘导论[M].北京:人民邮电出版社,2005.
- [15]刘畅,刘天琪,陈振寰,等.基于KPCA和BP神经网络的短期负荷预测[J].电测与仪表,2016,53(10):57-61.
- [16]YU F,XU X.A short-term load forecasting model of natural gas based on optimized genetic algorithm and im proved BP neural network[J].Applied Energy,2014,134(134):102-113.
- [17]翟玮星.基于卡尔曼滤波的短期负荷多步预测修正模型研究[J].浙江电力,2014,33(7):20-23.