灰色预测模型的改进及其在电力行业中的应用  

The Improvement of Grey Prediction Model and Its Application in Power Sector

作　　者：王亚[1] WANG Ya(School of Information Engineering,Xuchang University, Xuchang 461000, China)

机构地区：[1]许昌学院信息工程学院,河南许昌461000

出　　处：《许昌学院学报》2018年第6期68-72,共5页Journal of Xuchang University

基　　金：河南省科技厅项目(182102210500);基于下一代互联网技术创新项目(NGII20150610);许昌市科技局项目(20160213157)

摘　　要：针对已有的灰色聚类分析方法存在的不足和问题,对灰色预测问题进行了研究和分析,提出了一种基于改进的灰色聚类分析的预测模型.首先,基于灰色聚类分析的相关理论,对预测对象和数据进行灰色经典域和灰色节域的构建;其次,结合被预测对象的特征数据进行样本类别的聚类划分,形成与经典域相对应的分析类别;然后,基于灰色系统理论建立被预测对象与不同分析类别之间的灰色关联系数计算模型和灰色关联度计算模型,在此基础上,构建加权的灰色关联度计算模型,由此获得被预测对象的归属类别以及相应的分析数据.最后,以电力行业的电力负荷预测为具体分析案例进行了验证分析,说明了模型的有效性和可靠性.Aiming at shortcomings and problems of current grey theories for cluster analysis,this paper proposed an improved prediction model based on grey clustering analysis method by researching and analyzing the problem of grey prediction. First of all,based on related theories of grey clustering analysis,this paper constructed grey classical domain and joint domain for prediction objects and datum. Secondly,cluster partition for sample classification combining the feature data of predicted objects had been done to generate analysis categories corresponding to classical domain. Then based on grey system theory,this paper built a calculation model of grey relational coefficient and grey relational for predicted objects and different analysis classes,on that basis,constructed a weighted calculation model for grey correlation and obtained categories for predicted objects and relevant analysis data. Finally,experiments on power load forecasting in power sector proved that the prediction model proposed in this paper is effective and reliable.

关 键 词：灰色聚类分析 灰色预测 经典域 电力负荷预测 

分 类 号：TP311.13[自动化与计算机技术—计算机软件与理论]