缓冲算子修正的服装流行色灰色预测  

Grey prediction of garment color based on buffer operator correction

作　　者：周捷[1] 李健[1] ZHOU Jie;LI Jian(School of Apparel and Art Design,Xi′an Polytechnic University,Xi′an 710048,China)

机构地区：[1]西安工程大学服装与艺术设计学院,陕西西安710048

出　　处：《纺织高校基础科学学报》2019年第1期88-94,共7页Basic Sciences Journal of Textile Universities

基　　金：陕西省科技厅国际科技合作计划项目(2018KW-056);陕西省高等教育教学改革研究项目(省级重点17BZ037);中国纺织工业联合会教育教学改革项目(2017BKJGLX155)

摘　　要：针对服装流行色震荡数据的冲击扰动问题,提出缓冲算子修正的等维动态递补灰色模型,以还原数据本身特性。在采用弱化缓冲算子对2013—2019年服装流行色原始数据进行处理后,分别建立不同维数的等维动态递补GM(1,1)模型(BOGM),选择最优维数模型进行建模。为验证模型的优越性,以原始数据建立了经典GM(1,1)模型(EGM)、离散GM(1,1)模型(DGM)、等维动态递补GM(1,1)模型(NGM)以及等维动态递补离散GM(1,1)模型(NDGM)作为对比。结果表明,BOGM模型可有效解决震荡数据序列扰动问题,其平均预测误差及平均总误差分别仅为0.845,0.744,较EGM(5.694,4.137),DGM(5.259,3.952),NGM(6.007,4.289)和NDGM(5.743,3.800)模型均有了较大改善,能为服装流行色预测提供参考。In view of the impact disturbance problem of fashion color vibration data,an iso-dimensional dynamic complementary grey model modified by buffer operator is proposed to restore the characteristics of the data itself.After weakening buffer operator is used to process the original data of fashion color from 2013 to 2019,the iso-dimensional dynamic complementary GM(1,1)model(BOGM)with different dimensions is established respectively,and the optimal dimension model is selected as the test.At the same time,classical GM(1,1)model(EGM),discrete GM(1,1)model(DGM),equal-dimensional dynamic progressive complementary GM(1,1)model(NGM)and equal-dimensional dynamic progressive complementary discrete GM(1,1)model(NDGM)are established based on the original data.The results show that BOGM model can effectively solve the problem of vibration data series disturbance,and its average prediction error and total average error are only 0.845 and 0.744,respectively.Compared with EGM(5.694,4.137),DGM(5.259,3.952),NGM(6.007,4.289)and NDGM(5.743,3.800),the model of BOGM greatly improves,which can provide reference for fashion color prediction.

关 键 词：服装流行色 缓冲算子 等维动态递补 灰色模型 最佳维数 

分 类 号：TS941.7[轻工技术与工程—服装设计与工程]