多维灰色模型阶数差异性优化与结构拓展研究  

Research on the orders differential optimization and structural expansion of multidimensional grey model

作　　者：尹凤凤 曾波[1] 余乐安[2] 毛翠微 白云[1] YIN Fengfeng;ZENG Bo;YU Le’an;MAO Cuiwei;BAI Yun(School of Management Science and Engineering,Chongqing Technology and Business University,Chongqing 400067,China;School of Economics and Management,Beijing University of Chemical Technology,Beijing 100029,China;Wealth Management School,Chongqing Finance and Economics College,Chongqing 401320,China)

机构地区：[1]重庆工商大学管理科学与工程学院,重庆400067 [2]北京化工大学经济管理学院,北京100029 [3]重庆财经学院财富管理学院,重庆401320

出　　处：《系统工程理论与实践》2023年第7期2166-2178,共13页Systems Engineering-Theory & Practice

基　　金：国家自然科学基金(72071023);重庆市教委科学技术研究重大项目(KJZD-M202300801);重庆市自然科学基金(CSTB2023NSCQ-MSX0365,CSTB2023NSCQ-MSX0380)。

摘　　要：当前多维灰色预测模型对因变量和自变量取相同阶数进行建模,忽略了不同变量数据特征及物理属性之间的差异,导致了模型性能稳定性较差.为此,文章对因变量与自变量阶数进行差异性定义与优化,并引入非线性修正项以拓展模型灰信息结构,在此基础上构建了一种新的多维灰色预测模型.利用该模型对油气管道腐蚀速率进行模拟与预测,结果显示其综合误差(0.321%)优于其它三种主流多维灰色预测模型(3.035%,2.212%和0.755%),实现了对油气管道腐蚀速率的有效建模.本研究改善了阶数对序列数据的预处理效果和对数据特征的挖掘能力,对提升多维灰色预测模型的建模能力具有积极意义.Currently,the multidimensional grey prediction model takes the same order of dependent and independent variables for modeling,ignoring the differences between data characteristics and physical properties of different variables,which leads to poor stability of model performance.To this end,the paper defines and optimizes the orders of dependent and independent variables differentially and introduces a nonlinear correction term to expand the grey information structure of the model,on which a new multidimensional grey prediction model is constructed.The new model is used to simulating and predicting the corrosion rate of oil and gas pipelines,and the results show that its comprehensive error(0.321%)is much smaller than the other three mainstream multidimensional grey prediction models(3.035%,2.212%and 0.755%respectively),achieving an effective modeling of the corrosion rate of pipelines.The research results improve the pre-processing effect of orders on sequence data and the mining ability of data features,which have positive significance for improving the modeling ability of multidimensional grey prediction models.

关 键 词：多维灰色预测模型 阶数差异性优化 模型结构拓展 管道腐蚀速率预测 

分 类 号：N941.5[自然科学总论—系统科学]